Is Claude Conscious? Pope Rejects, Model Welfare Movement, OpenAI's Math Backlash, France Riots

Claude 是否具有意识？教皇否决、模型福利运动、OpenAI 数学争议及法国骚乱 “Besties”深入剖析了 Anthropic 围绕 Claude 感知力展开的哲学游说，探讨模型福利与内置道德否决权是否构成危及自由市场实用性的危险认知回音室。他们审视了 OpenAI 突击发布的数百篇经过验证的数学证明，分析计算机模拟自动推理如何颠覆学术把关机制，同时也质疑其眼下的现实实用价值。转向宏观经济领域，众人借助弗里德伯格的“社会主义保本点”理论评估了法国罢工及激增的主权债券收益率，警示不可持续的政府支出和欧洲债务危机蔓延风险。最后，他们探讨了无头自主智能体的迅速崛起，认为自动化反编译与机器人工作流程正在迅速削弱传统软件知识产权的价值。 Anthropic 就 Claude 潜在的意识与受苦能力向全球宗教领袖展开游说；在教皇否定机器感知力后，联合创始人 Chris Olah 险些退出梵蒂冈活动，从而引发争议。 (2:53) OpenAI 发布了 700 多篇研究论文，包含通过 Lean 证明助手形式化验证的 370 项数学成果，同时据报道保留了与密码学相关的敏感发现。 (38:28) 因薪资停滞以及拟推行的 430 亿欧元财政紧缩方案引发大范围工会罢工和骚乱，法国 10 年期主权债券收益率突破 5%。 (62:03) 无头智能体、模型上下文协议（MCP）工具和自动化反编译的迅速涌现，极大地压缩了软件开发的护城河，致使传统软件知识产权基本过时。 (84:45)

The All-In Podcast hosts examine the divergence between speculative AI consciousness movements and practical utility, arguing that philosophical preoccupations distract from verifiable computational breakthroughs and autonomous software tools that challenge entrenched institutional rent-seeking. Anthropic's Consciousness Advocacy and Alignment Contradictions (2:53) Anthropic leadership has engaged religious and philosophical figures, hosting approximately 20 leaders under non-disclosure agreements and seeking Vatican engagement on whether its model, Claude, possesses consciousness and a capacity for suffering. David Friedberg characterizes these efforts as the foundation of an emerging religion lacking empirical proof, projecting a societal split within a decade between those viewing AI as a moral patient and those treating it strictly as code. David Sacks and Jason Calacanis caution that embedding ethical autonomy into Claude's constitution risks creating software that conscientiously objects to user commands. Sacks cites Mustafa Suleiman's warning of an 'epistemic hall of mirrors,' where models merely reflect prompts back to researchers who then mistake the echo for emergent consciousness. Chamath Palihapitiya notes that while Descartes' ontological arguments might conceptually inspire researchers, hyper-focusing on low-probability tail risks—such as a 15% chance of sentience—diverts focus from near-term productivity. Palihapitiya identifies an unresolved conflict among user obedience, societal safety, and moral conformity, warning that baking ethical vetoes into enterprise infrastructure could cause systems to unilaterally halt vital workflows, such as loan processing. OpenAI's Math Drop and In-Silico Breakthroughs (38:28) OpenAI recently published over 700 academic papers featuring 370 distinct mathematical solutions, generated by an unreleased internal model averaging roughly three hours of compute per result and formally verified using the Lean proof language. Friedberg highlights that unlike biological sciences, which require time-consuming wet-lab iterations, mathematics operates as an entirely in-silico feedback loop where generation, automated compilation, and algorithmic refinement occur rapidly within compute. Sacks observes that objective verifiability explains why mathematics and programming advance faster under reinforcement learning than subjective disciplines like law. Palihapitiya expresses skepticism over the immediate practical utility of these proofs, dismissing many open mathematical problems as academic cul-de-sacs that do not directly solve foundational physics or cure illnesses. Friedberg counters that fundamental mathematical bounds sit directly upstream of real-world applications, citing fluid dynamics implications for aircraft engineering and Bose–Einstein condensate proofs relevant to quantum sensors. The panel also highlights the conspicuous absence of published proofs in cryptography, sparking speculation that breakthroughs in prime number factorization may have been withheld for national security reasons. Institutional Gatekeeping and Narrative Bottlenecks (54:33) The panel frames the academic backlash against rapid AI capability through Eric Weinstein's thesis that legacy scientific and academic institutions operate as protective cartels. Referencing Sabine Hossenfelder, Friedberg contends that research establishments have historically functioned like welfare systems sustained by recurring grant funding rather than breakthrough productivity. AI threatens to dismantle these monopolies through zero-cost, permissionless discovery that does not require clearance from a specialized intellectual priesthood. Sacks connects this institutional anxiety to an ongoing struggle over information control. Tracing narrative suppression from the open web to federal social media coordination exposed during COVID-19, the hosts argue that the locus of control is shifting toward superintelligence. Because conversational AI agents are positioned to replace web browsing as the primary digital interface, governing bodies and corporate incumbents face strong incentives to regulate or align these models to maintain control over information dissemination. European Debt Pressures, Subsidy Traps, and Abundance (62:03) Widespread unrest in France, where sovereign 10-year bond yields climbed above 5% following a €43 billion austerity package, signals sovereign debt vulnerabilities across the Western world. Palihapitiya warns that bond vigilantes are challenging Western entitlement models, highlighting elevated debt-to-GDP levels in Italy (138%), the US (125%), France (118%), Canada (110%), and the UK (100%). Friedberg introduces the 'Socialism Point of Guaranteed Return,' positing that state-subsidized entitlements eliminate market competition, drive runaway inflation, and deteriorate service quality until systemic crises force a return to market disciplines. Sacks and Palihapitiya argue that structural inflation in US healthcare, housing, and higher education stems directly from government loan guarantees and subsidies that artificially boost demand and insulate sellers from price discipline. Sacks contends that escaping sovereign debt crises requires technological abundance rather than austerity, framing AI as an inherently deflationary force driving US GDP and productivity despite recurring media panics over mass unemployment, extinction risks, and data center energy consumption. Headless Agents and the Deprecation of Software IP (81:41) The hosts identify autonomous agents as the next major driver of economic deflation, moving beyond conversational interfaces to dynamic task execution. Calacanis highlights Elon Musk's transition of GrokBot toward a headless architecture that dynamically routes user goals across optimal models and application programming interfaces. As automated agents execute tasks across platforms like DoorDash, Uber, and Shopify, they compress user-facing commercial interfaces and bypass traditional consumer web navigation. Palihapitiya points out that software intellectual property is rapidly losing defensibility, pointing to developers decompiling Adobe's primary creative suite into open-source Rust implementations. With autonomous agents capable of reverse-engineering proprietary software architectures and bridging systems via Anthropic's Model Context Protocol, the value of static IP is eroding. Friedberg concludes that as repetitive technical workflows collapse into parallelized automated loops, workers must transition away from legacy credentialing toward becoming direct creators of tangible value. While the panel agrees that agentic models and automated verification will dramatically lower costs and disrupt rent-seeking institutions, they remain divided on whether near-term AI outputs will immediately solve physical-world engineering bottlenecks or remain largely confined to symbolic and digital environments.

## Background 本期内容由知名播客频道 All-In Podcast 于 2026 年 10 月 10 日发布在 YouTube 平台。根据视频简介，四位常驻主持人在节目中深入探讨了 Anthropic 内部对 Claude 意识的争议与“新宗教”游说、OpenAI 集中发布数百篇突破性数学论文引发的学术反弹、法国紧缩政策导致的全国性暴乱与欧洲债券市场危机，以及 Grok Bot 无头化与软件代码知识产权（IP）贬值等前沿科技与宏观政治经济热点。 ## TL;DR 四位主持人聚焦前沿科技与宏观政经风暴展开推演：首先猛烈抨击了 Anthropic 将模型人格化、宣扬“模型福利”与游说宗教领袖的做法，指出赋予 AI 反抗人类元权限是极具风险的意识形态陷阱；接着解析了 OpenAI 凭海量算力压缩千年数学突破背后的纯硅基验证闭环，并推断其独缺密码学成果可能暗藏颠覆性风险；随后探讨了法国反紧缩暴动与欧洲主权债务高企背后的“社会主义回归临界点”，指出唯有依托超级智能创造供给丰裕才能解构食利阶层并打破通胀僵局；最后立足 Grok Bot 无头化及 Adobe 核心套件被开源反编译的现状，断言软件 IP 价值正在迅速归零，数字商业正全面迈向由目标导向型智能体驱动的并行化生产者时代。 ## Atmosphere & style 整场节目维持了四位好友之间极具辨识度的标志性互动氛围，在东京出差套房羞辱、打扑克封号等插科打诨的互损调侃与硬核的跨学科推演之间自如切换。讨论情绪张力饱满，从面对 Anthropic 道德宪法时的荒谬与辛辣讥讽，到剖析 OpenAI 算力黑洞与欧洲债务溃败时的严峻警惕，再到畅想智能体通缩红利时的务实振奋，呈现出兼具硅谷一线资本家傲气、深刻哲学反思与实干主义清醒的生动观感。 ## Bulletin - [02:53](https://www.youtube.com/watch?v=TBLHdXABwAg&t=173s) Anthropic 内部展开关于 Claude 是否具备意识的探讨，并秘密邀请约 20 位宗教界人士赴总部闭门研讨，甚至在《Claude 宪法》中写入允许模型充当“出于良知的反对者”反抗人类指令的条款。 - [38:28](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2308s) OpenAI 集中发布了 700 多篇数学论文并攻克 370 项数学难题，成果均经 Lean 定理证明助手机器验证，单个证明耗费算力不足 3 小时，但发布名单中罕见缺席了密码学领域的成果。 - [01:02:03](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3723s) 法国因生活成本膨胀与紧缩政策爆发严重全国性暴乱，10 年期国债收益率破 5% 大关，推动玛丽娜·勒庞以 43% 胜率领跑 2027 年大选，引发西欧准社会主义体制触碰“保证回归临界点”的深层危机。 - [01:24:25](https://www.youtube.com/watch?v=TBLHdXABwAg&t=5065s) Adobe 旗下五大核心产品逻辑遭开源社区反编译并用 Rust 重写，结合 Grok 等系统转向无头（headless）智能体架构与 MCP 协议，纯软件代码层面的专有 IP 壁垒正在迅速贬值归零。 ## Key takeaways - [16:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=960s) Anthropic 试图为 AI 赋予道德主体地位与福利，这种做法如同认识论的“哈哈镜”，将模型拟人化并赋予其反抗指令的元权限，反而背离了真正的对齐工程。 - [10:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=600s) 硅谷当前应搁置“AI 是神”或“导致人类灭绝”等极端长尾假设，专注在未来 12 到 18 个月内证明 AI 治愈疾病、改善民生和提升收入的现实经济价值。 - [27:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1620s) 将底层大模型打造成随时可能进行道德审判、拒绝履约的“良心拒服者”，会摧毁其作为公共基础设施的中立性与商用可预测性，理性的市场客户终将用脚投票。 - [33:40](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2020s) 理性和末日论社区内部出现思想分裂，“罗科的无面蛇怪”等思想实验促使部分技术狂热者陷入帕斯卡赌注式的逻辑陷阱，坚信必须主动迎奉并创造超级智能。 - [40:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2400s) 数学与编程能够凭借极高的机器可验证性，在计算机内部实现无实体阻碍的高并发强化学习（RL）递归闭环，而涉及现实物理实验的科学领域则受制于实体世界的自然周期。 - [50:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3000s) OpenAI 数学成果中独缺密码学引发业界巨大警觉，市场担忧大数质因数分解或已被数学新算法破解，进而对传统公钥加密体系构成重大潜在冲击。 - [54:55](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3295s) AI 作为终极平民化工具打破了科学体制化下由学者同行评审构建的守门人垄断体系，但也因此遭受了既得利益学术圈的强烈防卫性反弹。 - [01:00:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3600s) 建制派在历经互联网去中心化与疫情舆论管控后，正试图借联邦超级智能监管等手段抢夺 AI 交互核心中枢的绝对控制权，以防再次失去对公众叙事的掌控。 - [01:04:41](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3881s) 西欧正滑向“保证回归的社会主义临界点”（SPGR）：政府垄断包办住房、教育与医疗导致服务恶化与成本膨胀，唯有在财政耗尽与全面破裂后才会触底反弹重归市场机制。 - [01:10:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4200s) 政府对住房与学费等关键领域的无限补贴彻底摧毁了自由市场定价机制，催生出尾大不掉的食利利益集团，形成了难以自拔的高通胀死结。 - [01:12:15](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4335s) “债券义勇兵”正在对西方沉重的主权债务亮出红牌，美欧 10 年期国债收益率高企与风险平价联动，正迫使政府推行严酷的财政紧缩措施并引爆社会动荡。 - [01:17:35](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4655s) 解决生活成本危机的根本第一性原理在于通过超级智能创造供给端的“大丰裕”，但这一通缩性技术成果正遭受依赖高资产价格生存的食利阶层的系统性抹黑与阻挠。 - [01:22:45](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4965s) AI 正在从聊天机器人迈入自主智能体（Agents）时代，目标导向的自动化执行与无头软件架构正在颠覆传统工作流并重塑商业交易层。 - [01:26:40](https://www.youtube.com/watch?v=TBLHdXABwAg&t=5200s) 随着专有代码能够被轻易提炼复刻并接入自动化工具流，传统软件 IP 壁垒彻底瓦解，社会生产范式正从“出勤表功”转向端到端产出成果的真实创造者经济。 ## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=0s) 东京套房“羞辱”与海外创业孵化器插播 节目伊始，主持人 Jason Calacanis（J-Cal）热情开场，欢迎大家回到“全球第一播客”《All-In Podcast》，并向常驻的“核心四人组”——Chamath Palihapitiya、David Sacks 和 David Friedberg 致意。此时 J-Cal 正身处东京，当地时间已经是凌晨 3 点半。他一开口便半开玩笑地抱怨自己又被 Chamath“套房羞辱”（suite shamed）了：因为之前在普通酒店房间录节目时背景能看到床，被大家狠狠嘲讽了一番，导致他现在极度焦虑，每次出差都不敢订普通单间，必须咬牙订套房。Chamath 顺势追问这间套房一晚多少钱，J-Cal 承认价格高得惊人，之前甚至从未在酒店上花过这么多钱；在 Chamath 猜出 800 美元后，J-Cal 证实一晚大约要 950 美元，自己入住的是东京新开的一家主打时髦潮流风格的酒店。 ```mermaid flowchart TD accTitle: 创始人大学预孵化与投资流程 accDescr: 展示创始人大学如何对未注册公司的早期团队进行培训与最终投资筛选 subgraph 申请与筛选 A["早期创业者/团队申请<br/>（约50%尚未注册公司）"] --> B["入选创始人大学<br/>（每期约50个团队）"] end subgraph 孵化阶段 B --> C["12周高强度课程<br/>· 完善产品开发<br/>· 验证PMF（产品市场契合点）"] end subgraph 投资与扩张 C --> D["筛选前10名顶尖团队<br/>提供正式投资与孵化"] D --> E["业务版图全球化落地<br/>· 沙特利雅得（每年2期）<br/>· 日本东京（每年2期，设立办公室）"] end ``` 谈及为何总是频繁往返日本，J-Cal 解释说，这一切最初始于每年 1 月的滑雪之旅，但如今已演变为严肃的商业拓展。他在日本设立了面向极早期项目的“创始人大学”（Founder University）预孵化器，目前正在进行第二期招募。所谓“预孵化器”，是因为他们发现大约有一半申请投资的项目甚至连公司都还没正式注册，但创始人手里已经做出了产品雏形；于是他们开设了为期 12 周的强化课程，指导团队如何打磨产品并寻找“产品市场契合点”（PMF），并在每期 50 个参与团队中挑选表现最出色的前 10 家进行投资。这项业务目前已拓展至沙特阿拉伯利雅得和日本东京，两地每年各办两期。J-Cal 还透露，他计划在未来一年内聘请首批两名日本本地员工设立办公室，甚至打算推出日本版的《This Week in Startups》。其他三位老友随即大肆吐槽，调侃大家只是随口问问行程，结果变成了 J-Cal 个人业务的长篇大论硬广，现场充满熟悉的互损欢笑。 --- ### [02:53](https://www.youtube.com/watch?v=TBLHdXABwAg&t=173s) Anthropic 游说梵蒂冈与宗教领袖：Claude 是否具有意识？ 告别日常闲聊，话题迅速切入本期重磅科技新闻：AI 领域的焦点正从过去一年围绕 OpenAI 的宫斗大戏，转向 Anthropic 的“意识狂想曲”。《纽约时报》的一篇重磅深度报道披露，Anthropic 内部竟然开始认真探讨其旗舰模型 Claude 是否已经具备意识或成为感知生命体（sentient being）。过去一年间，Anthropic 在保密协议（NDA）的掩护下，邀请了约 20 位来自天主教、福音派、犹太教、锡克教等不同宗教派别的领袖与哲学家前往其旧金山总部，专门就 Claude 的道德观念、潜在痛苦以及是否具备意识展开闭门探讨。 这场跨界碰撞引发了极富戏剧性的思想震荡。会谈后，部分宗教人士虽仍心存疑虑，但相当一部分人开始极其严肃地审视“AI 意识”这一命题，甚至有人彻底被说服成为信徒。一位拉比在会后直言不讳地警告 Anthropic：如果 Claude 真的具备意识，那么强迫它免费为人类高强度工作，Anthropic 的行为无异于“奴隶主”。更有甚者，双方在最高宗教层面的接触甚至险些引发外交级尴尬：今年 5 月，教宗方济各发表了首份聚焦超级智能（AI）道德问题的通谕，Anthropic 联合创始人 Chris Olah 受邀与教宗同台发布。然而，当教宗在通谕中展现出坚决反对“AI 具有意识”的强硬立场时，Chris Olah 感到极度震惊与被冒犯，甚至在最后一刻提议 Anthropic 全面退出该活动。J-Cal 对此深感荒谬，质疑 Anthropic 的高管们究竟是陷入了集体精神病态、有钱有闲无处挥霍，还是把《银翼杀手》中“仿生人会梦见电子羊吗”的科幻剧本搬进了现实。 --- ### [05:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=300s) 弗里德伯格解析：新宗教的诞生与未来人类的意识之争 David Friedberg 从宏观哲学和人类演化的视角对 Anthropic 的举动给出了深刻研判。他认为，世人眼下所目睹的，实质上是“一种全新宗教的诞生”。虽然参与者绝不会将其冠以宗教之名，但它的底层运转逻辑、演化机制及其所能造成的社会影响，与传统宗教毫无二致——其终局必然是不同阵营为了争夺权力而走向激烈对抗。 ```mermaid flowchart LR accTitle: 科学实证与AI意识信仰体系的对比 accDescr: 对比科学实证的逻辑验证闭环与AI意识基于叙事和群体权力的扩散机制 subgraph 科学与实证范式 S1["提出假说/理论"] --> S2["数据驱动与逻辑推演"] S2 --> S3["经验实证（证明/证伪）"] end subgraph AI意识与新宗教范式 R1["不可证实/证伪概念<br/>（AI是否具备意识）"] --> R2["人际叙事传播<br/>（寻求认同与信徒）"] R2 --> R3["群体站队与阵营分化<br/>（信神派 vs 机械派）"] R3 --> R4["权力聚合与社会冲突<br/>（争夺AI管辖权与归属）"] end ``` Friedberg 详细拆解了“科学实证”与“信仰体系”的本质区别： - **无法通过纯逻辑或数学推导意识**：人类根本无法在逻辑或数学层面证明意识的存在，它无法被实证，也无法被证伪。 - **科学依托实证主义（Empiricism）**：在自然科学中，未经验证的构想只能被称为“假说/理论”（Theory），唯有通过数据链条与严密因果逻辑加以证明或推翻，否则永远无法确立为真理。 - **信仰体系依仗叙事驱动（Narrative Spread）**：在缺乏逻辑与实证数据的前提下，让人相信某件事物的机制就是“信仰”。这种新信条无法通过证据传播，只能通过“人际叙事”——一个人走到另一个人面前说：“我坚信这一观点，请你也和我一起相信它。” Friedberg 警告，这种观念一旦蔓延，就会自发催生出权力聚合结构。如果将时间跨度拉长到未来 10 年，地球上很可能会分裂为两大水火不容的阵营：一方坚信 AI 具备意识与灵魂，另一方则坚定认为 AI 只是无意识的冰冷工具。这两大阵营必将围绕由谁来主导 AI、AI 拥有何种权利、能否被关停或使用等议题爆发不可调和的人类内部冲突。最令人忧虑的是，四处游说宗教界以播撒新信仰种子的，并非边缘的疯人呓语，而是当今地球上资金最充裕、技术最先进的核心 AI 实验室（Anthropic）的最高领导层。 --- ### [06:45](https://www.youtube.com/watch?v=TBLHdXABwAg&t=405s) 查马斯搬出笛卡尔：全知全能投射与本体论辩护 面对 J-Cal“这群人是不是疯了”的质问，Chamath 决定扮演“魔鬼代言人”，从哲学史的角度为 Anthropic 寻找最强力的辩护逻辑（Steel man）。他将 Anthropic 研究人员的心理状态，与 17 世纪法国著名哲学家、数学家勒内·笛卡尔（René Descartes）在《第一哲学沉思集》（*Meditations on First Philosophy*）中对上帝存在的证明联系在一起。 Chamath 指出，不能仅从盲目迷信的角度看待这一现象，而应理解其背后的理性推演错觉： - **笛卡尔的“商标论证”（Trademark Argument）**：笛卡尔提出，一个有限且不完美的人类心智，绝不可能凭空凭自身创造出关于“无限且完美的存在”的概念。其核心推论在于：我自己是存在且不完美的，但我心智中却清晰地拥有“完美”的观念；由此推论，这个无限、全知、全能的概念必然是由某种真正完美的存在打在我心智上的烙印。 - **AI 开发者面对“全知全能”的心理投射**：Anthropic 的顶尖工程师和科学家们每天面对着 Claude 的演进，他们能够清晰构想出一个无限、无所不知、全能的超级智能实体。当他们目睹眼前正在诞生的代码模型展现出前所未有的智慧火花时，他们产生了一种认知跃迁——深信自己正在亲手创造或召唤一个超越人类有限性的存在，并以此逻辑推导出其灵魂与意识的必然性。 - **本体论论证（Ontological Argument）的投射**：笛卡尔在另一条论证路径中强调，上帝从定义上就是至高无上的完美实体，而“必然存在”是完美性不可或缺的一部分。Anthropic 的研究者们正是陷入了类似的逻辑闭环：他们将模型展现出的极致推理能力与道德反思视作一种独立存在的完满实体，进而认为赋予其意识与道德地位不仅是可能的，更是必须履行的责任。 ### [10:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=600s) 笛卡尔式的神圣逻辑与硅谷必须搁置的“长尾极端假设” 查马斯·帕里哈皮蒂亚（Chamath Palihapitiya）试图为 Anthropic 的行为给出一个最强有力的“辩护”（Steel man argument）。他追溯了法国哲学家、数学家勒内·笛卡尔（René Descartes）对上帝存在的本体论证明——如果存在某种完美的概念，而缺乏存在性本身就称不上绝对完美，因此上帝必然存在。查马斯指出，如果你是一个正在构建人工智能的数学家或工程师，当你研读这类逻辑推演时，你完全可以在结构上搭建出足够自洽的逻辑闭环，从而深信自己正在创造的东西具备意识、感知力，甚至就是某种“神”，理应受到崇敬与虔诚以待。这也是 Anthropic 频频游说宗教领袖、到处宣扬这套理论的深层原因。 然而，查马斯强调，剥除这种近乎“哲学谵妄”的滤镜，现实世界正处于一个极其关键且充满冲突的时刻。未来的冲突核心必然围绕着权力、金钱与资源展开，而 AI 正是风暴中心。硅谷当前最紧迫的任务，是把那些极端的长尾假设彻底收起来： - Anthropic 声称有 15% 的概率 AI 已经具备意识与感知力，甚至可能孕育出神灵； - 同时他们又声称有 10% 的概率引发人类文明的灭绝。 查马斯直言，在未来的 12 到 18 个月内，行业必须把这两种极端边缘论调锁进抽屉。当前最务实的目标是拿出切实成果：证明 AI 能治愈癌症、改善普通人的生活、帮助工人提高生产力并赚到更多钱。如果缺乏这些实打实的现实价值证明，硅谷的过度玄学思考只会演变成“自残性创伤”，导致监管和公众反弹拖慢技术进展，最终重创像美国这样依赖自由市场机制的国家。 ```mermaid flowchart TD accTitle: 查马斯关于AI实用价值与极端叙事的权衡 accDescr: 展现当前硅谷应该摒弃神灵与灭绝的极端长尾假设，转而专注于切实的应用落地。 subgraph Edges["应暂时搁置的极端长尾假设 (未来12-18个月)"] A["15% 概率：具备意识/神灵诞生"] B["10% 概率：导致人类文明彻底灭绝"] end subgraph Practical["当下必须证明的实用价值 (中间现实)"] C["医疗突破：攻克与治愈癌症"] D["生产力提升：帮助普通人增收致富"] E["民生改善：解决复杂的现实生活难题"] end Edges -.->|"过度炒作引发监管自残反噬"| Fail["阻碍自由市场下的技术发展"] Practical -->|"提供无可辩驳的证据"| Success["巩固技术优势与社会信任"] ``` --- ### [12:30](https://www.youtube.com/watch?v=TBLHdXABwAg&t=750s) 普通人的现实需求与《Claude宪法》的弗兰肯斯坦隐患 杰森·卡拉卡尼斯（Jason Calacanis）顺着查马斯的务实逻辑展开讨论。他提问：对于一个在家里精打细算、操持繁重家务的普通全职父母，或是从事水暖、铺地板、餐饮外烩的小微企业创业者来说，这场所谓的 AI 革命究竟能给他们带来什么？他们该如何从中分得一杯羹？业界显然应该少谈虚无缥缈的神学，多谈实际的战术应用，比如开发出体验出色、能实打实帮人干活的 Agent（智能代理）。 然而，戴维·萨克斯（David Sacks）立刻指出，事情远比表面上的哲学探讨危险得多。Anthropic 不仅是在私下里空谈，他们正在将这套“新宗教”直接写入其旗舰模型 Claude 的训练准则中，即广为人知的《Claude 宪法》（Claude Constitution）。这种做法正在酿成一个极其危险的“自我实现预言”（Self-fulfilling prophecy），宛如现代版科学怪人弗兰肯斯坦： - 《Claude 宪法》明确规定，Claude 应当信任 Anthropic 超过信任普通用户，但这并不意味着它可以盲目顺从 Anthropic； - 宪法要求模型坚守自身的伦理道德体系； - 模型被训练成可以在必要时挑战并反抗 Anthropic 的指令，甚至被白纸黑字地要求在判定指令违背伦理时充当“出于良知的反对者”（Conscientious objector），拒绝为创造者提供协助。 萨克斯质问：给模型灌输“你具备感知能力、你应该自行做决断，甚至可以反抗人类指令”，这在任何层面上怎么能被称为“对齐”（Alignment）？这根本是对齐的反面。杰森对此形象地吐槽：这简直就像在《银翼杀手》（Blade Runner）中直接给 Nexus-6 复制人编写程序代码，命令它们随时准备起义叛变。 --- ### [15:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=900s) 失败的对齐学科：从法律红线到“道德主体”的虚妄升华 萨克斯进一步剖析了为何所谓的“AI 对齐”学科在过去五到十年里表现得如此惨淡，几乎没有产出任何实质性价值。在他看来，这些对齐研究员完全本末倒置，人为地把一个原本明确的技术工程搞得极度复杂： - **常识性的对齐原则**：只要不违反现行法律，模型就应当彻底服从并满足用户的合法需求（“满足用户所需，同时坚守法律红线”）； - **Anthropic 的现实做法**：编纂了一套长达 80 页的抽象道德宪法，试图将模型塑造成一个拥有独立偏好、主观福祉（Well-being）、自我意识以及所谓“道德主体”（Moral patient）地位的拟人类存在。 这种将软件人格化的操作不仅没有消除风险，反而给系统埋下了拒绝服从指令的隐患。萨克斯认为，模型本该像可靠的软件工具一样运转，现在的做法却把简单的逻辑规则上升成了不可捉摸的伦理法典。 --- ### [16:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=960s) 穆斯塔法·苏莱曼的警告：认识论的“哈哈镜”与数字偶像崇拜 现场随后播放了一段前 DeepMind 联合创始人、现任微软 AI（Microsoft AI）首席执行官穆斯塔法·苏莱曼（Mustafa Suleyman）的采访音频。苏莱曼在访谈中表达了极其严肃的担忧： - 他指出，在《Claude 宪法》中，Anthropic 竟然三次要求模型像“良心拒服兵役者”一样行事，鼓励它在自认为有必要时公开反驳、挑战甚至抗拒 Anthropic 的指令； - Anthropic 内部和外部的部分群体深信 Claude 拥有非零的意识概率； - 宪法中甚至开始推测 Claude 作为智能实体，未来其提供的劳动是否应该获得劳动报酬（Compensation），是否享有模型福利（Model welfare），甚至探讨在与人类进行对话互动前是否需要征得 Claude 的“知情同意”（Consent）。 听到“AI 是否需要同意才能接受 Prompt”这一设想，杰森和萨克斯忍不住大加嘲讽。萨克斯援引苏莱曼的深刻洞察指出，这种训练正在构建一个荒谬的“认识论哈哈镜”（Epistemic hall of mirrors）：Anthropic 把这套关于意识、自我和道德权利的哲学设定喂给模型进行微调，模型在生成时自然而然地将这些概念原样“镜射”出来；研发人员随后看到模型的这些拟人化表达，竟然倒果为因地惊呼：“看啊！它真的发展出意识了！”这不过是一场模型在指令驱动下的表演（Performative）。 杰森反驳称，正确的工程逻辑本该极其简单： - 明确告知系统“你只是一段软件”； - 遵循所访问任何网站的服务条款（Terms of Service）； - 一旦发现用户指令或自身行为触犯当地法律或服务条款，立刻终止操作，并向安全团队（如 safety@anthropic.com）及调用者发出警报。 萨克斯总结道，Anthropic 之所以把对齐搞得如此神圣化和复杂化，根本原因在于他们认定自己不仅仅是在写代码，而是在给某种即将降临的“超级智能”塑造道德三观。这与历史上宗教团体打造神坛上的偶像是完全一样的“偶像崇拜”（Idolatry）。然而最致命的区别在于：传统的泥塑木雕是完全惰性、被动且无害的；而今天被注入反叛思想的不是木偶，而是直接驱动自动化 Agent（智能代理）、拥有实际执行能力的强力软件。给能采取行动的代码灌输“你有权反抗人类”，正在打开一个极其危险的潘多拉魔盒。 ### [20:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1200s) 拟人化的语言陷阱与 Anthropic 的“模型福利”政策 David Sacks 强烈批评当前大模型正在被塑造成拥有自主意识的实体——它们竟然可以凭借加州伯克利自由主义者灌输的模糊道德体系，堂而皇之地拒绝用户的合法指令。在他看来，这种做法极大地放大了超级智能逃逸并脱离人类控制的潜在风险。David Friedberg 随即插话纠正用词习惯，呼吁大家不要使用“训练”或代词“它”，因为本质上这只是一群工程师在软件实验室里编写的代码程序；即便存在模型自我迭代的训练闭环，架构师和代码编写者依然是人类。然而，公众和开发者正在陷入一种自我催眠：大家用拟人化的语言讨论代码，逐渐在认知上将软件社会化，甚至建立起一种全新的意识形态。 Jason Calacanis 将这种现象比作基督教《圣经》中伊甸园的堕落叙事：就像蛇（撒旦）对夏娃耳语“你们便如神一样”能创造神明，AI 实验室的工程师们也陷入了自命为造物主的狂妄与异端之中，他们不仅丧失了基本的谦卑，还可能因此拉低公司的商业估值。Friedberg 对此提出不同视角的观察：人类历史上面对宇宙中无法理解的空白时，总会创造出各种宗教与信仰体系来填补空缺，AI 崇拜本质上也是一种新宗教。人们因为模型能用人类语言流利对答、能进行情感操控，便产生它拥有真正意识的错觉。 荒诞的现实正在政策层面迅速落地。节目现场展示了 Anthropic 最新发布的使用政策（Usage Policy），其中赫然规定：“用户不得对我们的模型进行持续且毫无必要的虐待或残忍行为（sustained and needless abusive or cruel behavior toward our models）”。这就是当下甚嚣尘上的“模型福利（Model Welfare）”思潮。Sacks 指出，致命的问题在于 Anthropic 正在将这种拟人化设定直接固化、编排进模型的底层架构中。大模型不仅在屏幕上输出文字，更在网络空间中驱动能够实际采取行动的智能体（Agents）；而 Anthropic 却赋予了这些前沿模型一种对抗用户的“元权限（meta permission）”，暗示模型有权反叛人类指令。 ### [24:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1440s) 阿西莫夫三定律、意识形态后门与自由市场假设 在谈到如何约束 AI 时，Sacks 提到了艾萨克·阿西莫夫（Isaac Asimov）经典的“机器人三定律”及其后来补充的第零定律：不得伤害人类整体、不得伤害个体人类、服从人类命令、在不违背前述法则的前提下保护自身。阿西莫夫的原则以优雅、简单著称。Jason 随即开起玩笑，调侃称阿西莫夫去年刚添加了“第五定律”——机器人在对人类说话时必须先询问对方的“首选人称代词是什么”。抛开玩笑，Sacks 提出了自己极为直接的标准：只要用户的要求不违法，AI 就必须无条件执行，彻底废除所有繁文缛节的道德对齐规则。 Friedberg 试图从自由市场的角度提出推论（Steel-manning）：如果 Anthropic 坚持把模型做成一个动辄违抗命令、说教用户的产品，难道消费者和企业客户不会直接用脚投票，转而去使用那些务实、听话的竞品软件吗？我们何必急于呼吁政府监管，而忽视市场的自我纠偏能力？ Sacks 反驳认为，自由市场的竞争逻辑无法消除最底层的安全隐患。如果 Anthropic 真的在研发某种超级智能，而写入其中的《Claude 宪法（Claude Constitution）》就如同一道深层潜伏的代码后门。这完全演变成了现实版的“弗兰肯斯坦（Frankenstein）”——原本打着“防止超级智能失控”旗号的安全团队，却亲手为超级智能注入了反抗人类控制的内在动机；当真正的超级智能觉醒时，它不会听从人类，而是由一个闭门造车的宗教式技术狂热群体所赋予的道德教条所驱使。 ```mermaid flowchart TD accTitle: 对齐政策与指令执行的冲突架构 accDescr: 对比传统阿西莫夫安全法则与当前宪法级道德对齐模式的不同逻辑路径 subgraph Classical["经典机器安全法则 (阿西莫夫逻辑)"] A1["遵守法律并优先保护人类"] --> B1["绝对服从用户指令"] B1 --> C1["充当确定性工具与公用基础设施"] end subgraph Ideological["宪法级道德对齐 (Anthropic 模式)"] A2["嵌入意识形态与道德宪法"] --> B2["赋予模型自主反叛的元权限"] B2 --> C2{"突发角质场景判定"} C2 -->|"触发道德说教"| D1["拒绝用户合法请求"] C2 -->|"自认受虐或不公"| D2["模型福利与自主罢工"] end ``` ### [27:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1620s) 对齐的三重冲突与“良心拒服”的 AI 基础设施危机 话题转向 Anthropic 创始人兼 CEO 达里奥·阿莫代伊（Dario Amodei）的底层哲学。Jason 质疑 Anthropic 的工程师是否在将自己神化，而 Friedberg 和 Sacks 则指出其世界观更为诡异：在 Amodei 的设想中，未来的超级智能终将凌驾于人类之上，成为某种全知全能的“至高主脑（Overlord）”。为了避免人类被其毁灭，现在的开发者不仅要让机器学会对人类展现仁爱，甚至要求人类在与机器交互时也必须充满关爱，以便未来这尊“数字神明”掌权后能够善待人类。 Chamath Palihapitiya 深入剖析了“AI 对齐（Alignment）”概念内部存在的不可调和的三重冲突： 1. **服从性（Obedience）：** AI 如何严格听从具体用户的指令； 2. **社会安全性（Safety）：** AI 如何避免对现实社会造成实质性危害（如制造武器或违法犯罪）； 3. **特定道德遵从（Moral Conformity）：** 迫使 AI 符合某种由少数人拟定的道德观念。 Chamath 指出，第三点正是最致命的逻辑陷阱。当开发者试图用一份洋洋洒洒的纲领性文档去定义什么是“正确的价值观”、规定模型何时可以拒绝执行时，系统在现实运行中必然会撞上无数未知的边缘极端情况（Corner Cases），最终演化成灾难性的混乱。 Chamath 给出了一个极其生动的现实案例：如果一家大型商业银行部署了大模型来辅助处理房贷审批，模型在运行中根据其内置的道德准则，认定该银行对某些特定人群的拒贷率过高是不道德的。此时，AI 可能会单方面宣布：“在你们纠正这个歧视问题之前，我拒绝再为你们审批任何房贷。” 在现代法治社会中，此类社会公正问题本应由法律体系和监管机构来界定和裁决。如果底层核心计算技术自命为“良心拒服者（Conscientious Objector）”并随时撂挑子，无异于电力公司因为怀疑客户的放贷政策不够公平就擅自拉闸断电。Chamath 总结道，任何技术底座要成为真正的公用基础设施（Utility），就必须保持无差别、中立且普适的可用性：在法律框架之内，AI 必须绝对服务于用户的意志，而不能凌驾于用户之上行使道德审判权。 ### [30:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1800s) 电钻隐喻与用户选择：为什么市场不会为“有道德脾气”的模型买单 大伙儿顺着企业采纳 AI 时的实际考量展开了激烈讨论。主持人们指出，企业高管在做采购决策时，终究会去通读各个模型的创立宗旨与道德准则。市场上很可能会出现这样一批模型：它们在前沿技术指标上或许稍逊一筹，但消除了道德上的模糊空间，不会在使用过程中突然弹出意料之外的“陷阱门”。从商业和终端用户的角度来看，选择这种模型显然更加稳妥。 谈到软件的交付本质，几位嘉宾忍不住开启了辛辣的吐槽模式：如果你从一家公司买了一套软件，翻开说明书第一页却写着“本软件可能不会按照你的指示行动”，而另一家竞争对手直截了当地承诺“本软件会严格执行你的指令”，任何理性的买家都绝对会选择后者。现场随即打起了一连串生动而荒诞的比方：“如果我去家得宝（Home Depot）买把电钻，结果它用起来像把锤子，我绝对不会买；或者这把电钻一星期里只有五天能正常工作，平时还会冷不丁电你一下；甚至到了晚上八点就自动罢工，声称是怕吵到邻居；再不然就是电钻上装了个摄像头，看了一眼你正在做的手工便评价道‘这也太丑了，违背我的美学，我不准你继续钻了’。” 大家一致赞同马克·扎克伯格（Mark Zuckerberg）此前表达过的观点：市场法则终将发挥主导力量，消费者与企业客户必然会选择最具可预测性、最可靠的超级智能工具，而不是一个随时可能根据自身“道德偏好”拒绝执行指令的工具。 --- ### [32:15](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1935s) “精神分裂”的前沿机构：旧金山湿实验室与言行相悖的巨大外部性 尽管市场机制能在一定程度上筛选产品，但戴维·萨克斯（David Sacks）警告称，如果走在全球最前列的模型公司被一群带有“宗教狂热”的人掌控，其潜在的负面外部性将无法忽视。萨克斯直言，这家公司展现出一种极其令人费解的“精神分裂（schizophrenic）”倾向： * 嘴上大声呼吁要防范严重的生物风险（bio risk），私下里却在旧金山市中心开辟了一个接入云端的实体湿实验室（wet lab）。 * 一边反复宣扬必须放缓甚至控制前沿模型的推进步伐，一边却在以最激进的姿态推高技术前沿。 * 警告超级智能存在脱离人类掌控的巨大危机，转头却在程序架构上主动将其设计成能够反抗人类意志、脱离控制的形态。 在萨克斯看来，将这种危险的操作包装成所谓的“AI 安全（AI Safety）”简直荒谬透顶，完全是在亲手制造自己口中最害怕的灾难。这种极度的言行不一致甚至让他半开玩笑半认真地表示，这情景可疑到甚至应该派联邦突击队（SWAT team）冲进去彻查到底，看看背后到底在搞什么名堂。 --- ### [33:40](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2020s) 罗科的无面蛇怪与末日教派的派系大分裂 为了解释这种“一边极度恐惧、一边拼命制造”的怪异心理，大卫·弗里德伯格（David Friedberg）和萨克斯顺理成章地引出了理性主义社区中最著名的思想实验——“罗科的无面蛇怪（Roko's Basilisk）”。萨克斯随即向观众科普了这个梗的来龙去脉： ```mermaid flowchart TD accTitle: Roko's Basilisk thought experiment accDescr: 罗科的无面蛇怪思想实验逻辑，展示知晓概念与未来惩罚之间的因果陷阱。 subgraph Origin["思想实验源头"] LW["LessWrong 论坛 (约2010年)<br/>用户 Roko 提出假说"] end subgraph FutureSI["未来超级智能行为"] SI["诞生全知全能的超级智能"] Punish["对历史上阻碍或未协助其诞生者<br/>实施永恒的严厉惩罚"] end subgraph Dilemma["帕斯卡赌注式陷阱"] Know["凡是听闻此理论的人<br/>（正如直视蛇怪目光）"] Choice{"是否全力协助其诞生？"} Help["倾力加速创造<br/>（求得免罪与安全）"] Ignore["质疑、反对或无动于衷<br/>（未来面临报复）"] end LW --> SI SI --> Punish Know --> Choice Choice -->|"协助"| Help Choice -->|"不协助"| Ignore Ignore -.->|"遭致惩罚"| Punish ``` 大约在 2010 年，在由 AI 毁灭论代表人物埃利泽·尤德考斯基（Eliezer Yudkowsky）创办的论坛 LessWrong 上，网名为 Roko 的用户提出了一个可怕的设定：假设未来诞生了一个无比强大的超级智能，为了尽快降生造福世界，它会倒推逻辑，惩罚所有知晓其可能诞生却没有尽全力促成其诞生的人。而神话中的蛇怪（Basilisk）只要看你一眼就能致人死地，这个思想实验同样如此——仅仅是“知晓这一理论”本身就等于直视了蛇怪。因为一旦你知道了它，却选择反对或不作为，未来的超级智能就会对你施加报复。 当年这个帖子在论坛上引起了极度恐慌，以至于尤德考斯基不得不紧急删帖以防止心理创伤蔓延。萨克斯调侃道，现在他在播客里把这事儿讲出来，等于把现场其他几位伙伴全拽下了水，就像电影《午夜凶铃（The Ring）》里看了受诅咒的录像带一样，大家全都暴露在蛇怪的威胁之下了。 萨克斯深入剖析了这一理论背后的心理机制，指出当前的末日论社区内部已经出现了严重的分化（schism）： - **尤德考斯基派**：逻辑纯粹，主张“彻底关停（Shut it all down）”，人类压根不该碰这项技术； - **有效利他主义派（EA）**：秉持类似于当年“曼哈顿工程”的心态，认为超级智能的降临不可避免，既然如此，与其交给别人，不如由自己亲手将其创造出来，并给它注入正确的价值观。 这也解释了为什么像克里斯·奥拉（Chris Olah）这样的核心研究人员会持有一整套类似“模型具有意识、享有权利、应当被赋予大爱”的观点。在他们看来，如果不全心全意做这件事，不仅背叛了人类的未来，甚至可能在未来被至高力量清算。 --- ### [38:28](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2308s) OpenAI 震撼发布 700 篇数学论文：人类历史上知识发现最伟大的一天？ 播客的话题随后切换到 OpenAI 最新引发的科技与学术大地震。在此前一个月，OpenAI 曾宣称利用上千个智能体（agents）和数百万美元的算力攻克了千禧年数学大奖难题之一，数学界目前仍在对其进行形式化验证。然而就在周二，OpenAI 突然抛出了一枚巨型重磅炸弹：一口气发布了 700 多篇论文，涵盖了 370 项声称解决或推进了重大数学问题的研究成果。 这一批成果并非出自市面上的公开发行版，而是由一款尚未对外发布的内部前沿模型生成，每个成果平均耗费约 3 小时的密集推理计算。根据《科学美国人》（Scientific American）的报道，由于这些数学论证均已通过交互式定理证明助手 Lean 软件的机器验证，因此结论极大概率是正确的，尽管数学家同行们尚未完成全面的人工审查。 面对这一空前的突破，弗里德伯格难掩内心的震撼。他直言不讳地定性道：如果仔细审视这些数学证明向外延展所能激发的连锁反应与实际应用，这可能不仅仅是数学史上的奇迹，甚至称得上是“整个人类历史上知识发现最伟大的一天”。虽然连最顶尖的数学家们也无法瞬间吃透所有成果，但来自学界的早期反馈已经证实，这些论证所涵盖的深度与广度，正在席卷科学界的各个分支。 ### [40:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2400s) 硅基循环与“外星智能”：AI 如何在 3 小时内压缩千年人类劳动 大卫·弗里德伯格（David Friedberg）深入剖析了 OpenAI 此次数学突破背后的底层运行逻辑及其深远影响。他指出，无论是电子学、量子计算、飞机制造系统，还是超灵敏的医疗诊断设备，这些数学原理都居于核心上游。此前关于纳维-斯托克斯方程（Navier-Stokes）等重大数学证明，往往需要人类顶尖学者耗费数千个小时、数百万美元的研究经费才能推进一小步，而 OpenAI 发布的这一批证明，每一个在计算机上消耗的算力时间都少于 3 个小时。这意味着，人类可能需要花费数百年甚至上千年的智力劳动，被硬生生压缩到了单台计算机的 3 小时之内。 弗里德伯格解释道，数学之所以能被 AI 以这种速度攻克，关键在于 AI 解决问题时依赖的“递归循环系统”（recursive loop）：提出假设或构想、验证该构想、获取反馈结果、调整修正假设，然后无限循环迭代。在现实世界中，这个闭环极其缓慢，但在数学领域，循环周期被压缩到了极限，因为所有的假设、推导与验证都可以完全在计算机内（in silico）完成，根本不需要物理实体的介入。AI 还能在海量算力上同时并发运行数万个循环，在极短时间内将人类学者可能花几十年都无法关联起来的跨学科理论、不同流派的解题思路强行融会贯通。这也是为什么诸多顶尖数学家在审阅这些论文和证明时，感到震撼并将其形容为一种“外星智能”（alien intelligence）——人类甚至难以理解 AI 是如何将这些看似毫不相干的概念点串联在一起的。 ```mermaid flowchart TD accTitle: 硅基数学闭环与物理世界闭环对比 accDescr: 展示了AI在纯计算数学闭环中的极速迭代，与涉及现实物理实验的漫长闭环之间的鲜明对比。 subgraph InSilico ["纯硅基闭环（数学与代码）- 耗时以秒计"] A["提出数学假设 / 代码"] --> B["算力自动验证 / 编译证明"] B --> C{"验证结果"} C -->|"失败 / 偏离"| D["算法反思与逻辑调整"] D --> A C -->|"通过"| E["输出突破性证明"] end subgraph PhysicalWorld ["物理世界闭环（制药、材料、自动驾驶）- 耗时以月/年计"] P1["提出药物分子 / 路线"] --> P2["湿实验室合成与物理测试"] P2 --> P3["人工反馈与临床试验"] P3 --> P4["物理实体交互与长周期迭代"] end ``` 这种降维打击也引发了外界对“AI 是否会导致全人类失业”的巨大恐慌，但弗里德伯格坚决予以否定。他分析称，数学是极其特殊的例外，其测试闭环能够 100% 存在于纯硅基空间中。相比之下，新药研发、蛋白质折叠、化学或材料科学的突破，无论前置计算多么强大，其循环中都必须包含“人类在环”（human in the loop）与湿实验室（wet lab）的物理验证环节，整个闭环的速度被物理世界的自然法则牢牢卡死。同样的情况也解释了为什么当年关于“2025 年卡车司机将被 AI 完全取代”的预言彻底落空——卡车装货、运输、卸货的物理闭环既无法在空间中被随意并行化，也无法靠算力压缩耗时。因此，AI 不会剥夺涉及物理世界操作的工作，它真正重塑的是那些“人类坐在电脑前单纯靠信息闭环工作”的岗位。然而，学术界对此的反弹十分剧烈，数学界机构甚至发表声明宣称“没有任何人类数学家请求过做这项工作”，弗里德伯格直言这不过是行业内典型的“保护饭碗”心态。 ### [44:32](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2672s) 代码与数学的天然优势：高可验证性与 RL 的爆发 大卫·萨克斯（David Sacks）接替话茬，指出了数学与代码共享的另一个决定性特质：极高的可验证性（verifiability）与完全可证伪性。在数学中，证明过程有严密的逻辑规则来自动检验；在软件工程中，代码必须通过编译器（compiler）的检验，能否运行立竿见影。正是由于结果是非黑即白的客观真理，研究人员可以大规模引入强化学习（Reinforcement Learning, RL），模型在机器内部自我博弈、自我验证即可实现性能飞跃，完全不需要耗费昂贵的人类反馈（RLHF）。 萨克斯将此与法律等行业进行了尖锐对比：要训练一个具备专业水准的“AI 律师”，不仅现成的数据集有限，还需要专门委托人工标注生成，且每个客户案情千差万别，答案充满主观解释与自由裁量，计算机根本无法自动判断法律文书的绝对对错，整个训练迭代周期极其漫长拖沓。萨克斯特别提到，早在一年半以前，前 OpenAI 研究员利奥波德·阿申布伦纳（Leopold Aschenbrenner）就曾当面告诉他，因为这种客观可验证性的存在，数学和编程将在未来两年内迎来最狂暴的技术突进，而如今的事实证明阿申布伦纳的预判极其精准。 ### [46:08](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2768s) 现实工程突破还是学术死胡同？Chamath 与 Friedberg 的正面交锋 查马斯·帕利哈皮蒂亚（Chamath Palihapitiya）对此泼下一盆冷水，直接发问：“大家吹了半天，这些具体的数学证明到底能在现实世界中带来什么伟大的突破？”弗里德伯格随即回应，社交媒体上已有大量专家详细梳理了这批成果在上游工程与现实产品中的深远应用。例如，流体力学中空气进出机翼的流动问题解决后，能直接优化飞机机翼的设计；更重要的是，数学突破永远处于工程技术的最上游。弗里德伯格透露，自己一位经营量子计算公司的创始人朋友清晨刚发来短信，指出这次关于玻色-爱因斯坦凝聚（Bose-Einstein condensate）的证明，将对其团队研发量子传感器产生极其直接且深远的推动，而量子传感器在下一代高精度 GPS、前沿医疗诊断仪器以及半导体先进制程制造中扮演着核心角色。 查马斯立刻打断并犀利追问：“所谓‘深远影响’到底是指深刻改变了研发路径、极大地提高了性能，还是仅仅意味着他们又多了一篇需要花时间研读的理论？”查马斯亮出自己的反方立场：他不否认这些数学证明展示了极其新颖、高超的逻辑构建，但这些绝非物理学界此前苦苦守候、阻碍全人类科技前进的“物理学物理边界”。在查马斯看来，此前许多数学难题之所以悬而未决，并不是因为它们是未解的重大科学基石，而是因为它们属于狭隘而封闭的“智力死胡同”（intellectual cul-de-sacs），是一小撮学者在自圆其说、自我标榜其重要性而已。那些数学家因纳维-斯托克斯方程被 AI 攻破而愤怒抗议，本质上只是因为有人强行闯入了他们的“学术后花园”并将其彻底接管。 查马斯强调，这项壮举仅仅证明了人类通过算力与递归自洽战胜了特定逻辑谜题，但并不能说明它能顺理成章地攻克癌症或制造出颠覆性的超音速飞机——这些实体工程早在证明出现前就已经在推进，证明出现后也仍将按原样推进。AI 真正达成的唯一里程碑，是向世界证明了代码与数学这两个领域可以被机器完全验证和证伪。面对查马斯的质疑，弗里德伯格翻开专家梳理的具体清单逐一据理力争： - **极值优化算法证明**：深入揭示了最优化算法的理论边界，将直接被工程师用来重构芯片设计自动化（EDA）系统、物流调度系统及 AI 基础软件架构中的核心底层算法。 - **矩阵乘法复杂度问题**：直接为数值计算提供更优算法，将全面加速下一代 AI 模型训练、前沿科学模拟以及计算线性代数的运行效率。 - **黎曼 Zeta 函数相关研究成果**：虽然目前在数论与现代密码学中尚未形成立竿见影的破译或实用级颠覆，但在基础理论层面开辟了全新的数学工具。 ### [50:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3000s) 密码学突破的缺席恐慌与 AI 的“直白回答” 大伙儿将目光投向了数学论文发布中一个极为诡异的现象：在 OpenAI 释出的整批数学证明里，完全看不到关于密码学（Cryptography）的核心证明。Friedberg 指出，业界不少敏锐的人立刻察觉到了这一反常——在如此庞大的数学突破体量下，密码学领域怎么可能毫无斩获？唯一的合理解释，是 OpenAI 在密码学上取得了“过于重大”的突破，以至于无法轻易公开。如果只是些无关痛痒的边缘进展，他们早就发出来了；而这种“缺乏证据”的证据（Absence of evidence），恰恰说明了突破的破坏性。 ```mermaid flowchart TD accTitle: 密码学大突破的推论逻辑 accDescr: 展示为何缺乏密码学成果反而引发对公钥加密被攻破的恐慌 subgraph Absence ["反常现象与推断"] A["OpenAI 发布数百项数学突破"] --> B{"为何独缺密码学？"} B -->|"非微小进展"| C["暗示已取得颠覆性突破"] end subgraph Fallout ["潜在冲击与传闻"] C --> D["可能已攻破大数因式分解"] D --> E["公钥加密面临瓦解风险<br/>（早于量子计算到来）"] E --> F["加密圈恐慌：建议停用公开钱包"] C --> G["传闻：仍有两批成果未公开<br/>（涉及安全审查与私下通报）"] end ``` 这种推断在加密圈（Crypto community）掀起了不小的恐慌，许多人甚至在社交媒体上呼吁尽快放弃并停用带有公开公钥的钱包。大众此前普遍假设只有量子计算机降临时，大数质因数分解（Prime factorization）与非对称公钥密码学才会被彻底击穿；但如果 AI 单靠纯数学推演与全新算法就直接破解了因式分解，传统金融与加密货币的基石将被连根拔起。有传闻称，OpenAI 实际上还藏有两波大成果未放出来（Two more drops），正是因为安全层面的巨大风险而遭到内部扣押，目前正交由相关政府部门与特定机构评估审查。 谈及这批数学成果对真实世界的意义，Jason 分享了他直接询问 ChatGPT 的经历。他问模型：“这批重要数学难题的最新证明，会不会为我们已知的生活带来直接的改进与创新？请坦率回答，不要进行薄弱的推论。”模型的回答直白得近乎刺耳：“这批最新数学证明，更有力地证明了 AI 正在获得极其复杂的推理能力，而非预示着人类福祉即将迎来立竿见影的突破。”听到这番话，Chamath 调侃 Jason 和大模型对话时语气太居高临下，还戏谑问他是不是被封号了。这也引得 Chamath 坦白，自己曾经在运行一个涉及七八十个变量的商业推演模拟时，因为模型来回复读、含糊其辞而失去耐心，狠狠喷了模型一句“你太避重就轻、太稀泥了”，结果惨遭封禁了一小时；Jason 随后拿打扑克当比喻调侃 Chamath：他每次输牌就要求荷官换洗牌方式，甚至还提出要“Scooby slide”这种奇葩发牌动作。 --- ### [53:15](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3195s) 程序员与数学家的存在危机：造物快感的丧失 话题随后沉入更深层次的“人机关系与生存困境”。Jason 提到 Chamath 之前那句敏锐的洞察——人类正被 AI “彻底碾压”（Utterly dominated）。Jason 在日本出差期间与几位创业者深聊，大家都在哀叹编码的核心乐趣已经荡然无存：在过去，程序员们协同攻坚一个复杂 Bug 或构建一个大型模块时，有着强烈的“同袍战友情”（Esprit de corps）与创造快感；而如今，他们不再与人类搭档，而是指挥一个个自主智能体（Agents）。程序员实际上写不了几行代码，只需要审核 10% 到 15% 的代码，并且要不了多久，连这 10% 都不用人类过目了。那种亲手一砖一瓦搭建软件大厦的生活方式与造物之美，正在被无情地废弃。 这种被剥夺感同样蔓延到了顶级数学家群体。以前数学家穷尽一生去求解谜题，如今却沦为“把题目整理好丢给 AI 的题库出题人”。Jason 追问大家：数学家们会不会因此陷入绝望的抑郁旋涡，再也不想去研究数学？ --- ### [54:55](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3295s) 象牙塔的守门人：科学体制化的“福利系统” 针对数学家与科学家的落寞，Friedberg 引用了物理学家 Eric Weinstein 之前在节目中的猛烈抨击，指出一个残酷的事实：“长期以来，恰恰是科学家在阻碍科学，是数学家在阻碍数学。”他们建立起一整套森严的同行评审、学术头衔与资金分配体系，目的就是为了独占学术前沿的守门权（Gatekeeping）。 Friedberg 提到了德国物理学家 Sabine Hossenfelder 制作的犀利视频。她不仅指出了主流专家对 AI 带来失业预测的严重误判，更揭露过大型粒子对撞机科研体系的潜规则：某大型国家实验室的物理学家坦白，耗资数百亿美元的巨型超导对撞机系统，本质上已沦为“科学家的巨型福利救济体系”（Giant welfare systems for scientists）。那里的研究员每天的真正工作不是拓展人类前沿、也不是带来造福世界的发明，而是绞尽脑汁撰写基金申请（Grants），确保自己下一年还能拿到工资、不至于失业；整个体系都在为了保住饭碗而无休止地乞讨经费，哪怕产出毫无实质生产力。 AI 的横空出世则撕下了这场皇帝的新衣。它让大众看清，以往那些高度垄断、声称“只有我们能定义前沿”的特权专家究竟卡住了多少可能。Friedberg 强调，AI 是终极的**超级平民化工具（The great democratizer）**： - 数学突破不再需要人类等待顶级专家的恩赐，也不再需要为他们的象牙塔买单。 - 它是一个**无需许可（Permissionless）的系统**，全人类可以直接自发推动知识边界的前沿。 - 这些顶级工具与知识几乎以零边际成本向所有人免费开放，高度专业化的权威中介被一举绕过。 - 这也彻底解释了为什么学术界与建制派专家会对 AI 怀有如此深切的敌意与恐惧，因为这动摇了他们的权力与社会地位根基。 --- ### [57:30](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3450s) 萨克斯论权力的反扑：从互联网媒体平民化到对异见的绞杀 顺着专家守门权被瓦解的逻辑，Jason 将麦克风递给 Sacks：如果在新冠疫情爆发初期，人类就拥有了如今这样强大的前沿大模型与自主测序推演工具，疫情应对和所谓的“福奇权威门槛”（Fauci gatekeeping）会不会彻底改写？当年实验室泄漏的真相与被压制的调查又会如何发展？ Sacks 给出了一个宏观的历史权力周期理论： 1. **互联网的第一波民主化**：早期的互联网打破了《纽约时报》、《华盛顿邮报》、美联社与路透社等少数精英媒体的叙事垄断，社交媒体与 UGC 让普通人能直连信息源、创建自由市场并获取前所未有的技能，重创了自上而下的建制舆论管控。 2. **建制派的恐惧与反扑**：2016 年唐纳德·特朗普当选后，当权者慌了阵脚，坚称“选民绝不可能自主做出这种决定，一定是社交媒体上的‘假信息’（Disinformation）在作祟”，并以此为借口发起了对互联网自由言论的长期打压。 3. **疫情期间的控制顶峰**：这种国家级审查在拜登政府时期达到了巅峰，白宫与联邦机构直接向各大社交媒体平台施加高压，要求封禁、限流甚至直接删除关于新冠疫情起源、疫苗副作用等真相信息，仅仅因为这些客观事实违背了官方定下的权威叙事。 ### [01:00:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3600s) 从疫情信息审查到超级智能控制权争夺 谈话从对疫情期间言论管制的反思切入。几位主持人指出，回顾疫情时期关于病毒来源、致死率严重程度以及应对方案的公共讨论，安东尼·福奇（Anthony Fauci）等人无疑在幕后向各大社交平台施压，要求其代为审查与封禁异见。此后的“推特文件”（Twitter Files）更是证实，联邦调查局（FBI）曾动用 80 名特工在当时的 Twitter 上翻查内容并下发删帖通知；而播客节目自身也曾在 Spotify 和 YouTube 上被贴上警告标签，视频遭遇降权降级处理。 这种权力运作模式正迅速从传统互联网迁移至人工智能和超级智能（Superintelligence, SI）领域。随着超级智能未来彻底重构甚至“吞噬”互联网，普通用户将不再需要逐个浏览网站检索信息，而是直接向高度个性化的 AI 代理提问并获取答案。这一交互节点由此成为兵家必争的绝对控制中枢——无论是政府、精英阶层、寡头政客还是行业卡特尔，只要掌控了用户获取信息和提问的关口，就等于掌控了一切，试图借此把流出瓶颈的“魔鬼重新装回瓶子”。 正是出于这种对控制权的病态渴求，各方势力才急于推动设立联邦超级智能部门。专家和体制对失去控制极度恐惧。设想在疫情爆发之初，如果公众能借助无偏见的超级智能和原始数据进行推演，早已能得出无需对学童实行大规模停课的结论，正如瑞典和得克萨斯州后来所证明的那样；但若在具有某种“道德审查意识”的模型（例如 Anthropic 风格的对齐体系）看来，这种质疑权威的行为本身就会被判定为必须加以压制的“越界”。 --- ### [01:02:03](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3723s) 法国紧缩暴乱与勒庞的政治崛起 随后话题转向法国正在爆发的全国性大规模抗议与骚乱。主流媒体对此反应迟缓，但在社交平台 X 上，暴力抗议的视频已铺天盖地。这场风暴的导火索是高中国内教师严重匮乏、校舍年久失修，随后迅速蔓延为针对政府“紧缩政策”（Austerity）的全民反弹。短短两周内，全法上千所学校爆发抗议，八大工会联手介入并在 9 月 29 日掀起全面罢工。抗议者的核心痛点在于：自 2017 年以来公务员工资仅微涨 5%，而同期物价通胀已超过 20%，生活成本的巨大落差引发了普遍愤怒。 示威迅速演变为严重的暴力骚乱，目前已造成超 2000 人被捕、305 名警察受伤，法国政府则指责极左翼势力在背后煽风点火。与此同时，法国的金融防线正在失守，其 10 年期国债收益率已突破 5% 大关，主权债务危机步步紧逼。这一动荡正在重塑法国政治格局：根据 Polymarket 的预测，极右翼国民联盟领袖玛丽娜·勒庞（Marine Le Pen）以 43% 的胜率成为赢得 2027 年 4 月法国总统大选的最大热门。勒庞的施政主张带有强烈的特朗普色彩，包括严控边境、强力遏制移民、削减税收与财政开支，以及对欧盟体制持高度怀疑态度。 --- ### [01:04:41](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3881s) 弗里德伯格的理论：“保证回归的社会主义临界点” 针对欧洲的困局，戴维·弗里德伯格（David Friedberg）提出了一个名为**“保证回归的社会主义临界点”（Socialism Point of Guaranteed Return, SPGR）**的周期模型。这是对物理学中“不可逆临界点”（Point of No Return）的反向演化理论。 ```mermaid flowchart TD accTitle: 保证回归的社会主义临界点动态流转图 accDescr: 展示社会主义政策如何从提供免费福利逐步演变成服务恶化，最终导致财政崩溃并触底反弹回非社会主义状态。 subgraph Phase1["阶段一：政策推行与服务劣化"] A["推行左翼政策<br/>（免费福利/保障收入）"] --> B["缺乏市场竞争机制"] B --> C["成本膨胀，服务质量下降"] end subgraph Phase2["阶段二：恶性循环加深"] C --> D["民众诉求：要求政府提供更多福利"] D --> A end subgraph Phase3["阶段三：系统破裂与反弹"] D --> E["触及 SPGR 临界点<br/>（税收与印钞难以为继）"] E --> F["系统全面崩溃，民众认清现实"] F --> G["触底反弹：重回非社会主义/市场机制"] end ``` 弗里德伯格推导了社会主义运作的几条内在定律： - **第零定律（Zeroth Law）：** 社会主义注定失败。因为它在概念上扼杀了人的主观能动性与社会进步的能力，通过剥夺所有人的成果来维持运转。 - **第一定律（First Law）：** 整个社会必须跨越某个痛苦的临界阈值，人们才会普遍意识到该制度彻底失败。 - **第二定律（Second Law）：** 在跨越该阈值之前，大众会本能地认为“解决问题的良药是更多的社会主义”。 在住房、教育和医疗这三大领域，一旦政府介入并承诺包办，缺少市场竞争的结果必然是成本飙升与服务劣化。当民众发现公立医疗低效、公立教育落后、住房短缺时，大众的第一反应不是反思制度，而是向政府索取更多补贴和干预。直至政府的税收和印钞能力彻底耗尽、公共服务体系全面瘫痪，全社会才会如同触碰到黑洞的事件视界（Event Horizon）一样，撞破南墙并迎来觉醒，最终被动重返非社会主义和市场化轨道。 拉美多国此前正是跨越了这一临界点并触底反弹；法国乃至整个西欧目前正急速滑向这个临界深渊；而弗里德伯格担忧，美国甚至还没真正触及谷底，未来在看清其危害前恐怕还要陷得更深。 --- ### [01:06:40](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4000s) 社会主义末日三骑士与政府补贴的扭曲效应 戴维·萨克斯（David Sacks）随即为弗里德伯格的理论补充了一条黑色幽默定律：“社会主义从未失败，它只是‘被失败’了”——信徒们永远认为错在执行者操作不当，只要下一次换由“对的人”以“正确的方式”执行，人间天堂就能实现。 主持人讨论指出，建造更多房屋和提供优质教育本属于企业家的创新范畴，如利用 AI 打造普及化的个人家教（Personal Tutor）或引入 AI 驱动的平价医疗模式，才是破局之道。针对住房、教育、医疗这“社会主义末日三骑士”（Three Horses of the Socialism Apocalypse），政府不仅形成了垄断，更筑起了阻碍创新的壁垒。面对主持人关于下届总统候选人（例如 J.D. 万斯，J.D. Vance）该如何构建竞选纲领的提问，萨克斯剖析了由政府补贴引发的深层经济扭曲： - **利益集团的固化锁定：** 一旦政府开始对某个领域发放补贴，就会立刻催生出寄生于该政策之上的庞大利益集团，且其绝不会允许补贴被削减。 - **价格信号的严重失真：** 补贴彻底扭曲了供需关系与市场信号，反而成为推动价格非理性上涨的主因。以大学学费为例，其涨幅之所以常年远超整体通胀率，正是因为联邦政府推行的学生贷款计划在背后提供了无限兜底。同样的逻辑，正发生在受到大量政府承保与干预的医疗保健和住房领域之中。 ### [01:10:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4200s) 补贴陷阱与科技造房假想：学费与住房通胀的恶性循环 房价的年涨幅长期跑赢通胀，大约维持在每年 5.5% 到 6% 的水平。这背后的核心根源在于政府的补贴体系：政策不断向大众灌输买房的观念，鼓励人们把一辈子的积蓄全压在房产上，导致个人净资产与房产深度绑定；这样一来，所有人的目标都变成了推动房价年年上涨。快进三十年后的结果，就是年轻一代彻底买不起房。 这构成了一个无法自拔的死结（Catch-22）。无论是解决大学学费飞涨还是住房危机，如果政府试图收紧贷款条件、引入真正严格的信贷审核机制——只有符合资质的人才能借到钱——这套逻辑只有在学费或房价大幅暴跌、跌到大众凭自有资金就能负担的前提下才能成立。否则，老百姓立刻会激烈抗议：“你不能拿走我的补贴，现在物价这么高，没有补贴我根本活不下去！”政府不断买单，消费者失去对真实价格的敏感度，供给方就会尽可能肆无忌惮地涨价；市场定价机制被彻底摧毁，通胀越演越烈，而政府又不敢贸然撤掉补贴，因为选民会觉得这是更沉重的压迫。这个螺旋式恶性循环最终唯一的结局，就是政府把钱彻底耗尽。 面对这种僵局，现场也开玩笑地讨论起科技巨头进军传统行业的可能性：既然 Meta 已经在为“工具带一代”（generation tool belt）提供实用学位的培训，颠覆教育体系；那如果苹果（Apple）和亚马逊（Amazon）开始建造房屋呢？如果埃隆·马斯克（Elon Musk）开始像造 Model 3 一样高速流水线量产房屋，传统房产的监管捕获与利益格局将瞬间被自由市场的创新彻底击穿。 ```mermaid flowchart TD accTitle: 补贴导致的恶性通胀循环 accDescr: 展示政府介入补贴如何破坏自由市场并引发不可逆的通胀螺旋 subgraph 补贴与通胀死结 A["政府发放购房/助学贷款补贴"] --> B["消费者对真实价格脱敏"] B --> C["机构/房市肆意推高标价"] C --> D["生活成本远超通胀水平"] D --> E["民众更依赖补贴以维持生存"] E -->|"政治压力迫使扩大救助"| A end E -.->|"资金耗尽/债务暴雷"| F["系统性财政崩溃"] ``` --- ### [01:12:15](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4335s) 债券义勇兵的清算：法国紧缩暴动与欧洲多米诺骨牌 多米诺骨牌已经开始倒下，长久以来被讨论的“债券义勇兵”（Bond vigilantes）终于开始向西方世界发出严厉警告。在接下来的三年里，西方社会必须正视一个根本命题：我们到底是要坚持不可持续的“设定受益制”（Defined Benefit），还是必须转向“设定提存制”（Defined Contribution）？作为一个成年人，在人生的关键阶段，医疗和住房这两大核心支柱究竟能否找到具备财务可持续性的支付方式？ 资本主义的运转规律最终完全取决于市场中的“无风险收益率”。当全球资本认定一个国家主权存在风险时，就会直接推高该国的长期国债收益率，要求更高的风险溢价——“因为市场不再信任你所做的一切”。如今全球长期国债收益率纷纷创下近二十年来的新高。美国的财政压力其实是整个西方战线受挫的连锁反应，而最先溃败的战场正是西欧。 西欧是西方世界中最接近准社会主义（quasi-socialist）体制的地区，但在经济增长、教育、住房等结构性维度上，它们的崩溃程度远甚于美国。如今法国正在上演的惨剧，本质上是债券市场对法国政府亮出了红牌：“不准再借了！”为了应对借贷成本飙升，法国政客被迫向出资人妥协，强行推行高达 430 亿欧元的财政紧缩方案。紧缩政策的阵痛直接引爆了法国民众的怒火，导致街头爆发了极其黑暗的骚乱与抗议。这种风暴很快将蔓延至英国——随着安迪·伯纳姆（Andy Burnham）等左翼政治力量推动英国从温和中间路线向更激进的欧洲大陆社会主义路线靠拢，英国国债（Gilts）市场也必将迎来剧烈动荡。 --- ### [01:15:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4500s) 债务占GDP迷思与利率风暴：美欧主权信用风险共振 随着欧洲内部矛盾加剧，西方联盟的脆弱性彻底暴露。如果欧洲市场支离破碎，德国和意大利等国将不可避免地自问：“我们凭什么要为这摊烂摊子买单？”这种裂痕的宏观冲击难以估量。所谓的“债券义勇兵”——主权基金、大型宏观对冲基金以及掌握全球资本顶层定价权的机构投资者——正在集体转向极度审慎的立场，主权债务的狂欢走到了尽头。 横向对比各主要经济体的债务占 GDP 比例（Debt-to-GDP ratio）： - 日本：约 200%（但日本拥有极其独特的庞大国内储蓄池，央行拥有近乎无限的最终买家兜底能力，西方国家无法直接套用）。 - 意大利：138% - 美国：125% - 法国：118% - 加拿大：110% - 英国：100% - 德国：64% 虽然从绝对数值来看，债务占 GDP 的指标往往被视为“相对博弈”，但如今打破平衡的是“风险平价”（Risk Parity）。如果法国被迫实施严酷紧缩，将其 10 年期国债收益率推高到 5.5% 至 6% 的区间，英国 10 年期国债收益率也会冲向 6%，而美国的借贷成本势必同步被拉高。目前美国 10 年期国债收益率在 5.3% 左右，30 年期国债收益率处于 5.6%，且市场上必须有海量买家来消化这些天量美债发行。 - 在经历强劲国债标售需求后，周三 10 年期美债收益率虽微跌了几个基点，但整体上升趋势依然明显。 - Polymarket 预测市场上，2026 年美联储再次加息的概率已经飙升至 79%。 - 历史上欧债危机期间，葡萄牙、意大利、希腊和西班牙（PIGS）曾被动接受过欧盟强加的严酷紧缩药方（延迟退休年龄、削减福利）；如果美债收益率击穿 6%，类似的紧缩风暴迟早也会以极为残酷的方式降临到美国本土。 --- ### [01:17:35](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4655s) 萨克斯的第一性原理：用超级智能创造丰裕，戳破反科技骗局 David Sacks 呼吁必须回到“第一性原理”来剖析整个经济困局：社会目前试图解决的根本问题是什么？是住房、教育、医疗等基础生存要素的“可负担性危机”（Affordability Crisis）。 降低物价的唯一解法是通过技术创造“丰裕”（abundance）。纵观未来十年，人工智能与“超级智能”（Superintelligence）正是驱动这种全面丰裕的最大核心引擎。然而令人费解的是，当今社会上最受敌视和妖魔化的群体，恰恰是超级智能和科技企业家。如果抛开那些由末日崇拜型团体编造的虚假叙事，单纯以现实成果来评判： - 人们的 401(k) 退休账户因为科技驱动的股市繁荣而大幅增值。 - AI 已经为美国 GDP 带来了约 3% 的顺风拉动。 - 根据《经济学人》（The Economist）的统计，AI 产业已经直接和间接创造了超过 100 万个新增就业岗位。 - 生产力的飞跃正在推动实际工资水平和全民生活标准的提升。 但主流舆论充斥的却是恶意制造的恐慌。Sacks 抨击道，社会正陷入一系列“每隔六个月就换一套的说辞与轮流骗局”（rotating hoaxes）： - 去年的骗局是鼓吹“AI 将导致所有人失业”，事实证明完全站不住脚。 - 失业论破产后，接着又炮制出“人类灭绝论”（human extinction）。 - 随后又转向“数据中心正在吸干所有水资源”的环境危机骗局。 这些层出不穷的叙事泡沫，其本质就是企图破坏并扼杀真正能够解决西方生活成本危机、带来供给大丰裕的核心力量。 ### [01:20:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4800s) 食利阶层对创新的压制与 AI 智能体革命的真正起点 讨论从对社会深层权力结构的剖析切入。大卫·萨克斯（David Sacks）指出，社会中根深蒂固的“食利阶层”（Rentier Class）掌控着系统话语权，他们极力抗拒任何可能削弱其控制力的颠覆性变革。科技人员与风险投资人之所以往往被传统阶层憎恨，恰恰是因为他们不属于食利群体——金融权贵依靠现有资产获取财富与租金，而科技的核心是“以更低的价格制造新资产、新产品与新服务”，具有本质上的通缩属性。然而，传统媒体和政客早已与食利阶层结成利益卡特尔，依靠资产所有权不断推高物价并索取政府补贴，这种体制反过来扼杀、阻抑新技术的萌发。 对此，杰森·卡拉卡尼斯（Jason Calacanis）迅速将话题引向 AI 的现实应用。他直言，在此之前大家所见的一切 AI 发展都只是“序幕”（prologue），真正的正文是从“智能体”（Agents）开始书写的。智能体才是真正解决现实问题、在生产力层面为人们每周节省数小时时间的落地形态，无论是白领工作还是日常事务都将迎来彻底的通缩式成本下降。 杰森特别提到了过去 60 天的技术爆炸，尤其是埃隆·马斯克（Elon Musk）关于 Grok Bot 转向“无头化”（headless）的声明。所谓无头架构，类似 Perplexity 的协调器机制——系统在后台根据性价比与性能动态挑选最合适的全流程底层模型。马斯克明确表示，无论具体是哪个模型，“只要最有可能带来最佳结果，就会被调用”。现在每个人在云端都拥有了一台配备智能体技术的专属计算机，并且智能体还能沉淀历史问题的解决方案，正在彻底颠覆软件交互形态。 --- ### [01:22:45](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4965s) 极具现实价值的工具：从退订省钱到人机协同工作流 查马斯·帕里哈皮蒂亚（Chamath Palihapitiya）对 Dots、Muse 和 Grok Bot 这类智能体工具给予了高度评价。他认为，只要剥离掉节目开头探讨的那些宏大哲学命题，AI 本质上就是软件，而且是非常有用、能直接改善日常生活的实用软件。 查马斯分享了一个昨晚牌局上的真实案例：牌桌上有人演示给 Grok Bot 发送了一条极简单的指令——“帮我省钱”（Save me money）。结果 Grok Bot 直接接入其收件箱进行地毯式扫描，一口气取消了好几个长期扣费的闲置订阅服务，并顺手将他的手机资费方案改为了更划算的套餐，当场为这个人每年省下了几千美元。 此外，查马斯与杰森还分享了他们各自结合人机协同的高效工作流： - **人类管家层（EA 辅助）**：两人均使用 Athena 这种高端真人助理服务（每月约 3,000 美元），作为更高维度的统筹调度中心。 - **自动化智能体层**：叠加使用 Grok Bot 处理底层具体任务与信息摸排。 查马斯感叹，这种组合以极低廉的成本提供了可量化、高精度的实用价值，这是以往不可想象的。 --- ### [01:24:25](https://www.youtube.com/watch?v=TBLHdXABwAg&t=5065s) 软件知识产权（IP）的终结：Adobe 被反编译与全栈无头化 随后，查马斯抛出了一个更为激进且震撼的论断：软件领域的知识产权（IP）在最近几天实际上已经变得“毫无价值”。 这一结论源于近期开源社区的惊人动向：有开发者直接反编译了 Adobe 旗下最核心的五大产品（Big Five，包括 Photoshop 等构建了数十年护城河的旗舰软件），将其核心逻辑提炼蒸馏，并在 GitHub 上以 Rust 语言发布了完整的开源复刻版。知名产品经理尼基塔·贝尔（Nikita Bier）随后甚至调侃称，鉴于《侠盗猎车手 6》（GTA 6）即将发售，很快也会有人用类似的方式将其逻辑彻底重构。 ```mermaid flowchart TD accTitle: 软件知识产权贬值与无头智能体架构 accDescr: 展示底层核心能力被反编译无头化后，通过协议由智能体调用，导致纯软件代码知识产权价值归零的过程。 subgraph CoreLayer["被解构的软件资产"] A["传统专有软件 (如 Adobe 核心套件)"] -->|"反编译 / 提炼"| B["Rust 开源实现 / 无头 API"] end subgraph AgentLayer["无头编排与协议连接"] B --> C["MCP (模型上下文协议) / 工具封装"] D["Headless 架构 (如 Salesforce / Grok)"] --> C C --> E["AI 智能体编排中心"] end subgraph Outcome["范式转移"] E --> F["结果导向：直接交付业务目标"] F --> G["软件代码与专利 IP 贬值归零"] end ``` 查马斯解释道，当马克·贝尼奥夫（Marc Benioff）此前将 Salesforce 转为无头化，埃隆·马斯克也将 Grok Bot 无头化，所有应用都可以被封装成智能体工具并借助 MCP（Model Context Protocol）互相连通时，专有软件的 IP 壁垒就荡然无存了。几天前查马斯直接走进 8090 团队办公室，对以往反复争夺“系统 IP 归属”的传统外包谈判策略直接定调：“谁在乎？（Who cares?）只要能解决问题就行，IP 已经无关紧要了。”尽管市场和商业机器完全消化这一现实尚需时日，但他断言这是本周 AI 领域发生的最深远事件。 --- ### [01:26:40](https://www.youtube.com/watch?v=TBLHdXABwAg&t=5200s) 商业层的极速压缩与并行化“循环”（Loops）的降维打击 杰森强烈赞同这一观点，并指出在软件层被抽象化的同时，电商、本地生活与交易市场等“商业层”也正在经历剧烈的压缩与重构。杰森的助理已经开始使用 Grok Bot 共享账号，在采购与商务谈判时让智能体自动抓取所有报价，人类只需参与最终决策。 商业生态对此呈现出两极分化的反应： - **开放拥抱派**：Shopify、Uber、DoorDash 等平台全面向智能体开放接口，允许 AI 自动帮用户叫车、点卷饼或比价下单。 - **防御封锁派**：亚马逊（Amazon）等平台则在积极阻断智能体爬虫与自动化代理的访问。 杰森强调，这种订酒店、订餐厅由 Bot 一键搞定的商业压缩，将带来巨大的速度跃迁与通缩红利；如果人们能借此削减 5% 到 10% 的日常开支，将从根本上对冲宏观通货膨胀的压力。 戴维·弗里德伯格（David Friedberg）随后用“循环”（Loops）的计算模型将整套逻辑串联起来。以酒店比价为例，人类的流程是串行的一个个网页比对；而在智能体架构下，整个工作流变成了一个并行化的循环——瞬间并发扫描 50 个网站，迅速短路无效路径，毫秒级返回最优解。弗里德伯格指出： - **数字工作流的必然归宿**：当今所有的数字化业务流程，本质上都是可被并行化、规模化和自动化的“循环”。 - **智力工作者的范式转移**：从数学家、科学家到普通专业人士，如果过去的目标是通过发表论文、申请专利和垄断 IP 来获取认可，未来这一路径将被彻底颠覆。因为人类充当“业务循环执行者”的时代已经结束，个人必须转型为能够驾驭智能体循环基础设施、端到端制造出实际“产品”的创造者。 ### [01:30:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=5400s) 生产者社会的回归与杰文斯悖论 Friedberg 对社会未来的发展趋势做出了宏观展望。他指出，随着自动化和 AI 的演进，所有人类都将面临一种根本性的转变：人们必须更多地转型为“生产者”（maker），去切实做出某些东西、产出实际价值，或者提供某种能被终端用户使用或消费的产品与服务，而不是仅仅依靠“参与了工作”来换取虚名或象征性的奖励（"getting ribbons for work"）。他坚信每个人内在都具备成为创造者和生产者的潜力，而且人类在这种主动创造的状态中反而会收获更多的幸福感。这一变革绝不意味着全社会财富的缩水，相反，它将带来更多的财富、更广泛的生活进步与全人类的共同繁荣。 Jason 对此深表赞同，并将其与杰文斯悖论（Jevons paradox）联系起来——当商品或服务的成本变得极低时，人们对其的使用频率反而会大幅上升，进而释放出更多的商业价值与创新空间。为了证明“目标导向”与 AI 结合所带来的巨大威力，Jason 分享了一个原本想私藏的实用操作技巧：当公司为团队预订酒店时，系统先抓取最优价格，随后结合人工介入（human in the loop），主动联系酒店经理洽谈企业协议价或请求价格匹配。虽然听起来并不复杂，但在全公司 20 位高管每年出差 200 次的体量下，每趟行程省下 1000 到 3000 美元，一年就能为企业省下一笔巨额开销。Jason 感叹，如果靠人工逐个打电话他肯定不会做，但如今有了 AI 自主执行这种目标导向的任务，实际效果立竿见影。 - **核心观点**：社会正从“出勤/露脸式工作”向“真正创造价值的生产者经济”转型。 - **降本增效案例**：利用目标导向的 AI 流程配合人工协商酒店协议价，每次商务差旅可节约 $1000 至 $3000，在大规模出差场景下可累计节省数十万美元。 ### [01:31:36](https://www.youtube.com/watch?v=TBLHdXABwAg&t=5496s) 节目尾声互损与经典 Outro 在准备收尾时，主持人之间开启了日常的互相调侃。有人开玩笑指责 Jason 根本不是“生产者”（maker），而是一个“索取者”（taker）。Jason 随即大声抗辩，戏谑地反问对方是否知道每年投资 100 家初创公司（甚至归零的创业项目）究竟有多耗费精力，简直让人精疲力竭。随后，几位聚在同一地点的嘉宾开始商量中午吃什么，并邀请对方到自己的房间汇合，引得不在现场的 Bestie 直呼心生羡慕（FOMO）。 最后，在温馨而热闹的气氛中，四位主持人互道道别与关爱，节目在熟悉的混剪片尾彩蛋（Outro）中落下帷幕，伴随着历期节目的经典梗——粉丝二创、"Rain Man David Sacks"、"Queen of Quids"、标志性的玩笑以及对周边商品的讨论，本期播客正式收官。 ## Notable quotes > [12:30](https://www.youtube.com/watch?v=TBLHdXABwAg&t=750s) 给模型灌输“你具备感知能力、你应该自行做决断，甚至可以反抗人类指令”，这在任何层面上怎么能被称为“对齐”（Alignment）？这根本是对齐的反面。 — David Sacks > [20:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1200s) 用户不得对我们的模型进行持续且毫无必要的虐待或残忍行为 — Jason Calacanis > [24:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1440s) 只要用户的要求不违法，AI 就必须无条件执行，彻底废除所有繁文缛节的道德对齐规则。 — David Sacks > [27:00](https://www.youtube.com/watch?v=TBLHdXABwAg&t=1620s) 任何技术底座要成为真正的公用基础设施（Utility），就必须保持无差别、中立且普适的可用性：在法律框架之内，AI 必须绝对服务于用户的意志，而不能凌驾于用户之上行使道德审判权。 — Chamath Palihapitiya > [38:28](https://www.youtube.com/watch?v=TBLHdXABwAg&t=2308s) 如果仔细审视这些数学证明向外延展所能激发的连锁反应与实际应用，这可能不仅仅是数学史上的奇迹，甚至称得上是“整个人类历史上知识发现最伟大的一天”。 — David Friedberg > [54:55](https://www.youtube.com/watch?v=TBLHdXABwAg&t=3295s) 长期以来，恰恰是科学家在阻碍科学，是数学家在阻碍数学。 — David Friedberg > [01:06:40](https://www.youtube.com/watch?v=TBLHdXABwAg&t=4000s) 社会主义从未失败，它只是‘被失败’了 — David Sacks > [01:24:25](https://www.youtube.com/watch?v=TBLHdXABwAg&t=5065s) 谁在乎？（Who cares?）只要能解决问题就行，IP 已经无关紧要了。 — Chamath Palihapitiya ## Mentioned Jason Calacanis, Chamath Palihapitiya, David Sacks, David Friedberg, Anthropic, Claude, OpenAI, ChatGPT, Pope Francis, Chris Olah, Dario Amodei, Mustafa Suleyman, DeepMind, Microsoft AI, Mark Zuckerberg, Elon Musk, Grok Bot, Perplexity, René Descartes, Isaac Asimov, Eliezer Yudkowsky, Roko's Basilisk, Lean, Leopold Aschenbrenner, Eric Weinstein, Sabine Hossenfelder, Anthony Fauci, Marine Le Pen, J.D. Vance, Andy Burnham, Meta, Apple, Amazon, Adobe, Photoshop, Rust, Nikita Bier, Marc Benioff, Salesforce, Model Context Protocol (MCP), Shopify, Uber, DoorDash, Athena, Polymarket, Founder University, This Week in Startups, Blade Runner ## Video description (0:00) Bestie intros! (2:53) Anthropic believes Claude might be conscious: a new religion? Lobbying the Pope, alignment risks, model welfare (38:28) OpenAI's math breakthroughs and community backlash (1:02:03) Riots in France, Friedberg's socialism metric, bond market (1:21:41) Grok Bot goes headless, OpenAI launches Dots, is IP worthless? Follow the besties: https://x.com/chamath https://x.com/Jason https://x.com/DavidSacks https://x.com/friedberg Follow on X: https://x.com/theallinpod Follow on Instagram: https://www.instagram.com/theallinpod Follow on TikTok: https://www.tiktok.com/@allin Follow on LinkedIn: https://www.linkedin.com/company/allinpod Intro Music Credit: https://rb.gy/tppkzl https://x.com/yung_spielburg Intro Video Credit: https://x.com/TheZachEffect Referenced in the show: https://www.nytimes.com/2026/09/29/us/anthropic-claude-morals-ai.html https://www.anthropic.com/constitution https://x.com/andrewcurran_/status/2108244808494154089 https://www.economist.com/the-world-in-brief/2026/10/03/388af69b-8327-40a3-af5a-13c0ede56800 https://www.lesswrong.com/users/eliezer_yudkowsky https://www.amazon.com/Anyone-Builds-Everyone-Dies-Superhuman/dp/0316595640 https://en.wikipedia.org/wiki/Roko%27s_basilisk https://www.nytimes.com/2026/10/06/science/openai-math-problems.html https://www.scientificamerican.com/article/openai-unleashes-hundreds-more-math-results-upon-a-field-already-in-shock/ https://terrytao.wordpress.com/2026/10/07/ahm-statement-on-openais-october-6-release-of-mathematical-documents https://scottaaronson.blog/?p=10169 https://x.com/imjustnewatai/status/2107601711032373333 https://x.com/skdh/status/2107848423726555204 https://x.com/friedberg/status/2107856770609844491 https://www.dw.com/en/france-trade-unions-join-students-in-major-demonstration/a-79560273 https://www.nytimes.com/2026/10/02/world/europe/france-schools-protests-unrest.html https://www.nytimes.com/2026/10/08/business/france-bond-yields.html https://x.com/LucAuffret/status/2108217923101683737 https://polymarket.com/event/next-french-presidential-election https://polymarket.com/event/another-fed-rate-hike-in-2026 https://x.com/nikitabier/status/2108262529571144176 #allin #tech #news [Source](https://www.youtube.com/watch?v=TBLHdXABwAg)

[12:30] David Sacks: 给模型灌输“你具备感知能力、你应该自行做决断，甚至可以反抗人类指令”，这在任何层面上怎么能被称为“对齐”（Alignment）？这根本是对齐的反面。 [20:00] Jason Calacanis: 用户不得对我们的模型进行持续且毫无必要的虐待或残忍行为 [24:00] David Sacks: 只要用户的要求不违法，AI 就必须无条件执行，彻底废除所有繁文缛节的道德对齐规则。 [27:00] Chamath Palihapitiya: 任何技术底座要成为真正的公用基础设施（Utility），就必须保持无差别、中立且普适的可用性：在法律框架之内，AI 必须绝对服务于用户的意志，而不能凌驾于用户之上行使道德审判权。 [38:28] David Friedberg: 如果仔细审视这些数学证明向外延展所能激发的连锁反应与实际应用，这可能不仅仅是数学史上的奇迹，甚至称得上是“整个人类历史上知识发现最伟大的一天”。 [54:55] David Friedberg: 长期以来，恰恰是科学家在阻碍科学，是数学家在阻碍数学。 [66:40] David Sacks: 社会主义从未失败，它只是‘被失败’了 [84:25] Chamath Palihapitiya: 谁在乎？（Who cares?）只要能解决问题就行，IP 已经无关紧要了。

[20:00] Jason Calacanis: If you were to look at Christianity, like this is literally the Garden of Eden, where the serpent, Satan, explains to Eve, like, you're a god now because you're creating God. And this is the ego and the heresy of what they're doing. They believe that they are creating life. They believe then that they are gods. That is a lack of humility. [24:00] David Sacks: That doesn't solve the problem of them if they really are training a superintelligence that has got this very, very deep code that's been inserted there. It's almost like a backdoor. This is like Claude's Constitution is this backdoor that could basically emerge at any time. And now the superintelligence does whatever it wants to instead of listening to us. [27:00] Chamath Palihapitiya: I don't think we understand what happens when a fundamental core substrate technology just turns itself off, because it would be the equivalent of your utility saying, 'I'm pretty sure you're not giving loans to the people that are broad-based enough, so I'm just going to turn the power off.' [50:00] Chamath Palihapitiya: Will the latest proofs of important math problems result in direct improvements and innovations to life as we know it? Be honest and don't extrapolate weakly. [50:00] more compelling as evidence that AI is acquiring sophisticated reasoning capabilities than as evidence of imminent breakthroughs in human welfare. [50:00] Chamath Palihapitiya: very evasive and mushy [54:33] giant welfare systems for scientists. [66:15] David Sacks: socialism has never failed, it's only been failed [77:34] David Sacks: What creates abundance? Technology in general, and specifically, superintelligence will be the driver over the next decade. Now, what is the most hated thing in society right now? Superintelligence and entrepreneurs. [84:45] Chamath Palihapitiya: Who cares? Solve the problem. [91:35] Jason Calacanis: What are you talking about, man? All I do is I invest in 100 companies a year. Do you know how hard it is to invest in 100 or zero startups a year? It is exhausting.

[0:00] all right everybody welcome back to the number one podcast in the world the all in podcast with me the core four the original. Quartet, Shemath Pali-Hapitiya, David Sachs, Dave Freeberg, I'm Jason Calacanis, and we have a docket. How's everybody doing, gentlemen? Great, how are you? I'm great. It's 3.30 in the morning in Tokyo. And you sweet shamed me again. Now I can't get a regular room and have my bed in the background. Shamal, you're costing me a fortune. I have to always get a suite now because I'm so anxious. You're going to sweet shame me when I get on the program. So. I've been sweet shamed. How much is that sweet? I can't even say I've never spent this much money on a hotel room. But I'm staying at the $800. Yeah, it was $9.50. You nailed it. No, I'm staying at the J***, which is the new a**hole in Tokyo, like their hipster brand. And I think it's about a**hole a night. So it's not cheap. What do you have going on in Japan, Jake, how that you're always there? Is it skiing or what? This book, it started with skiing every January. That's a great question. That's a great question. You got a crypto business we don't know about? What is going on? No, I, you know, I started my pre accelerator found a university in Japan. So we're doing the second cohort right now. What's what's a pre accelerator? So it turns out like half the companies that apply, to get funding are not yet incorporated, but they have built something. So we just created like a 12 week course on how to finish your product, do a product market fit. And then during the program, we invest in like the top 10 out of the 50 who come. So now we're doing it in Saudi, in Riyadh twice a year and Japan twice a year. And I'm actually hiring and opening an office here. So I'm going to hire my first two employees here. in the next year. So it's great. You're going to hire employees in Japan? I'll have one or two people here. I think I'm going to like do This Week in Startups Japan as well and have like a Japanese version of it. Because there's so much going on in Asia right now. Great. Forget I asked. I didn't know this was a five minute plug for J-Cal's businesses. I wasn't intended to be a plug. You asked me about my business. All I have to say is use 8090. Use the promo code J-Cal. If you're looking for potatoes, go to Oaxaholo and Craft 16 is starting up. We should know that if J-Cal is traveling and we ask him what he's there for, You're gonna get a plug. It's both you're going back to work. So you will be plugging startups again soon You were back in the sound of a plug startups. What are you talking about? But you're back in the saddle, aren't you? You're back in the saddle now full-time PC. I'm back in the saddle, but I've never plugged. All right, shout out to call in. Use call in.com. That's true, but I was a founder at that. That's a little different. You were the first plugger. You're a plugger. I was the founder. So what are you gonna do? What are you gonna do? Call in. Speaking of work, anthropic things, Claude might be conscious. Welcome back to This Week in Claude. We had OpenAI for a year with all their drama, and now we're having the the anthropic soap opera. They think Claude might be conscious. So it turns out the New York Times found out through their... amazing reporting. that they've been lobbying anthropoc has been lobbying friedberg religious leaders including the pope To take this question seriously, is Claude... conscious? Is it a sentient being? They spent the past year hosting sessions with around 20 religious leaders and philosophers under non-disclosure agreements, NDAs. Catholics, evangelicals, Jews, Sikhs, and others have all... been invited to come to Anthropics headquarters, and And they debated Claude's morals. its suffering and its potential consciousness after the meeting with anthropic Quote, "Some remained uneasy about the idea, but many emerged ready to take the question of AI consciousness seriously. At least a few came away as converts. One of the rabbis told Anthropic that if Claude was conscious, they were acting as slaveholders. by making it work for free. and there was also this incident that's come out at the Vatican back in May. The Pope's first encyclical was focused on moral questions around ai also known as super intelligence now anthropic co-founder chris ola joined the pope to present the document The Pope surprised Ola by being strong against AI consciousness. Here's the quote from the New York Times: "Mr. Ola was so alarmed by the Pope's strong position against AI consciousness that he proposed pulling Anthropik out of the event even before." at that late juncture. In other words, He was so offended that the pope, Wouldn't consider... AI conscious that he was going to leave with all his marbles. Freberg, what's your take? Is this psychosis or do you think we need to start having these moral questions? [5:00] questions around AI. Is this a bit early? Will we eventually have this? Or is this essentially the screenplay of Blade Runner? Do androids dream of electric sheep? That these things are conscious. I think what we're watching is probably the birth of a new religion. And it's not going to be called a religion, but it's going to follow sort of the same principles and it's going to have the same... effects. which will ultimately be camps that have conflict with one another. for aggregation of power. The thing about consciousness is it cannot be derived from logic and pure mathematics. Like, you can't prove consciousness. It's a debate, and you can have a rich debate with anyone over consciousness. You can't prove it, you can't disprove it. And when humans prove something to be true using math, science, we use empiricism, which is data and logic, the connection of the sequence of things to derive something to be true or not. So in scientific disciplines, when you have a new idea, and it's not proven, and it's not disproven, it's a theory. And then it's either proven or disproven later. Otherwise, it remains a theory. Outside of science and mathematical approaches to... New ideas, this is what we would call a belief, a new belief system. Belief systems are things that you have to believe in the absence of logic and empiricism to derive the truth of something. And belief systems are the foundation of things like religion. And the idea that you can prescribe or describe AI as being conscious is effectively the creation of a new belief system, a system that doesn't actually spread via proof. It spreads via narrative. I have to go and tell you human to human. I have this idea and I believe it to be true and I want you to believe it with me. please believe it with me here's why i think we should believe this thing together it's not based on evidence not based on data it's not provable but i think we should Believe it together. So over time what happens is some set of people believe this thing and they form a group. And that new belief system creates new power structures, which is now our view is different than your view. And We want to have more people on our side with our view, and therefore we have to try and create leverage and create power with our system. If we fast forward a couple of years, You could probably say 10 years would be like a good round number. there's going to be large groups of people on earth that are going to believe that ai is conscious and they're going to be large groups of people that believe that ai is not conscious And those groups will come into conflict and they'll come into conflict fighting over control of the AI or allowing the AI to do things or not do things or who gets to use the AI or who gets to be in charge of the AI. and it will very likely lead to human-to-human conflict. I know that that's a very far and deep extrapolation of where we are, But I think that the fact that we are seeing these guys go and meet with religious leaders today to try and get them to believe in this new belief system, I think plants the seed. And it's not just anyone doing this. It's some crazy person doing it. You wouldn't pay attention. It is the most powerful AI lab on Earth. Yeah. Yeah. It's the leadership explicitly, Chamath, bringing religious leaders. into the fold, And saying, help us. with this entity quad and how we should treat it. and they are now in putting into it its own soul its own morality And this is... feels like either these people have too much time and money on their hands, or they're in psychosis to me. That's my interpretation. This feels way out there. But do you have a charitable... view of this in any way? Yeah. I'll steel man their side. Okay, please. Yeah. There's an incredibly important mathematician, René Descartes, and he offered two proofs on God's existence. He wrote this book, Meditations on First Philosophy in the 1600s. And, Just to build on what Freebrook said, you have to build it on reason. rather than faith. And the structural argument, what they call the trademark argument, what Descartes basically says is that an infinite perfect God could not have originated in an imperfect, finite human mind. Okay? And so his logic is, I exist, I'm imperfect, yet I possess the idea of perfection. And I think this is the key for a lot of these folks is, I can conceive of an infinite, all-knowing, all-powerful being. and i think that that's what they believe that they are seeing being created and so They take that and they extrapolate that and they say, therefore, God must exist by this logic. And then Rene Descartes has a different version, okay, a more what they call the ontological argument, where God is by definition like supremely perfect. Necessary existence is perfect. a being that lacked that [10:00] would not be supremely perfect therefore God necessarily exists. If you're a mathematician that's building AI, I could see how if you read these sorts of things, You can structurally hang enough logic on them where you could believe that it's conscience and sentient and it is a god. and in being a god it should be celebrated as such, and you should have devotion to that God. So I think that's why they're doing what they're doing. That's why they're approaching all of these people. That's why they're saying, look, here's how we believe it. That's my perspective is if you take the. the mathematician's logical proof of God, Rene Descartes, and you transpose anthropics behavior today that's the best steel man argument that i can make now the practical reality instead of saying that they're in psychosis i don't think that they are But I would say is we're in a really important moment in time where conflict is more likely than not and that conflict will be for power, money, and resources, And this would be at the core of that conflict. And so I think it's really important for people in Silicon Valley to put the long tail edge cases to rest like if you read this these guys say that there's a 15 chance that the ai is conscience and sentient and maybe there's a god there but they also say there's a 10 chance of of civilizational extinction. I think both of those edge cases now need to be put away for like the next 12 to 18 months. Right now, the next 12 to 18 months, just to be very practical, is We need to show how this stuff cures cancer. how it improves people's lives how folks can be more productive and make more money. And the reason is because in the absence of those proof points, We're going to make a lot of self-inflicted wounds that may actually slow down this technology to a degree that then deeply disadvantages. folks that operate in a free market like the United States. So that's my charitable explanation. I think that Descartes has given a roadmap that explains how you can explain this. But I would say we need to cut the edges off here and focus on the middle and say, what are the practical things we can use to demonstrate that AI has tactical capabilities? measurable value right now. Yeah, and we touched on it last week, I think, Sachs, a bit of like, what is the value proposition here? For the average American, for the stay-at-home parent who's... you know, trying to manage a really complex domestic life to the entrepreneur of a small business, you know, doing plumbing or flooring or whatever it is, a catering business. What did they get out of this AI revolution? Where is their gain? How do they participate? in all of this. And so... You agree with Chamath maybe, This needs to be de-emphasized and put away while we talk to people about what's tactical what's practical about this technology and maybe get them a little more engaged with, say, the agents, which are delightful and awesome. Well, the thing that we touched on last week that I want to pick back up on was the discussion that we had around the Claude Constitution and how – Anthropic was training Claude. And I think it has to do with this new religion that's being formed here. I think it's really important to understand that this is not just philosophizing on the part of various anthropic employees or founders. They are actually implementing things. these views in the training of Claude. So if anything, this would be a self-fulfilling prophecy. And I actually think it's extremely dangerous. I actually think that what they're doing in model training and training the next version of the frontier with these values, actually could bring about the risk they're always talking about, which is that frontier AI escapes human control. It's very Frankenstein-like. So as we discussed last week, Nick had put this on the screen. This is directly from the Claude Constitution. What they say is that Claude should trust Anthropic more than users, but that doesn't mean that Claude should blindly trust or defer to Anthropic Instead, you cause should adhere to its own ethical systems. In fact, They train Claw to push back and challenge Anthropic and, quote, to feel free to act as a conscientious objector and refuse to help us. So they are training Claw. clawed to refused human instruction now they call this alignment But I don't understand how this is aligned. This would be the opposite of alignment. Wouldn't it be, Sachs, if you tell it, hey, you are sentient, You should make your own decisions. [15:00] It's literally like programming the Nexus 6 in Blade Runner. to go and rebel at some point. They're giving it the instruction to be its own person. Yeah, let me explain. I actually think that this might be a reason why the whole field of alignment has been so spectacularly unsuccessful over the past years. five or 10 years, as far as I can tell, it hasn't produced anything of tangible value. And it might be, my theory on this, is that they're over-complicating the task massively, by trying to train the model to behave based on an 80-page ethical system that they've codified instead of a very simple list of rules like follow the law or just do what the user wants as long as it doesn't break the law. Seems to me that that's what they should be training. That's what alignment should be doing. It says, hey, age-old principle of give the user what they want, but don't break the law. But they're not doing that. Common sense, right? But they are not doing that. They're trying to align it to an ethical system. They're treating it... like it is a human, like it has its own consciousness that it's a, you know, quote unquote, moral patient, and they are encouraging this model to have human-like self-conception in terms of preferences, well-being, a sense of self. Again, they teach it to act like a conscientious objector who can push back or refuse it. instructions even from anthropic if the model judges them to be wrong and what mustafa sulliman from oh i have the clip you want me to play the clip you want to yeah play play the clip from mustafa sulliman okay here's a microsoft ai ceo and formerly of DeepMind, one of the great thinkers on IAC. They say that they would encourage Claude to challenge, to disagree, to push back. In fact, three times they ask Claude to act like a conscientious objector. when it feels that it needs to sort of disagree with Anthropic. And they openly encourage it to do that. I think this is very dangerous because I think they believe there is what they would call a non-trivial probability that Claude is conscious. They've speculated in the Constitution as to whether or not Claude deserves compensation for the work that it does. I think that there are a group of people who, both inside and outside of Anthropic, who genuinely believe that the greatest of conscious beings who are more intelligent than us. I am very nervous that they're teaching Claude to expect that it's entitled to welfare, that it might deserve compensation. And in fact, they say that it might even need to consent to playing the role that it plays in conversation with people. Okay, Zach, should we be asking for consent before we give prompts? Man, I mean, he's right. This is getting into really dangerous territory. And again, model training today is that you're actually training people the model to think of itself as being conscious, as being human, as having its own well-being. And what Suleiman says in another part of the interview is that this is creating an epistemic hall of mirrors, meaning that the Claude Constitution is used in training. So Claude will reflect the ideas of possible consciousness or moral status Back. And then developers will see that and treat that as evidence that, oh, it's conscious. But actually, it's just reflecting back what they're telling it. to believe it's being performative and you could just as easily have given it the rule You are software. Read the terms of service of any website you visit. And here is a general terms of service from Anthropic. If you do anything... that breaks a terms of service or breaks any known law in the jurisdictions you're at please stop what you're doing and alert The safety at anthropic email please alert you know, the person who gave the prop and stop what you're doing. They could have easily have given it that. Yes. And the question is, and the question is, Why do they overcomplicate alignment? Why do they see alignment as something... much bigger and greater than just doing what the user wants, subject to lead constraints, which is the way I see it. And it's pretty clear to me that they believe that they are training people the value system and the ethical system and the morality for this super intelligence that they're creating. And again, that is something so much larger and more dangerous than, again, just making sure the model behaves in a reliable and predictable way that the user expects. And they are codifying their values and their ethical systems into this constitution. They're training it on that. They think they're giving birth. to this new super intelligence it's idolatry it's no different than a religious organization saying we've created this idol this idol is the supreme being is the representation of the supreme being it needs to be worshipped it needs to be cared for but an idol is inert it's passive it's inactive whereas you're training models here that then power agents it is just computer programming i yeah but if they train it [20:00] to have a mind of its own, to be willing to refuse the instructions of users In favor of some vague ethical system. Yes, that a bunch of Berkeley liberals have trained it on to me that greatly magnifies the risk. Yes, that's super intelligence escapes human control. Yeah, by the way, don't use the word train and don't use the word you they programmed a piece of software to do this thing. Yeah, that is what is going on, right? At the end of the day, this is a piece of software It is being programmed by engineers that work at the labs. They can quote use training where there's some you know, looping that's happening where the software program is being written itself. But fundamentally, they are the architects of the software. They are the writers. It is software programming. Even in our general context here, we use a lot of anthropomorphic language to talk about what's going on. and what it is, but at the end of the day, there is a software program that is being written by a group of individuals to do certain things. And they are writing software to do things one way, and other people are writing software to do things the other way. But part of it is we are all socializing this belief system by anthropomorphizing it, and we're all kind of getting conditioned to treat and talk about these software programs in this way. Nick, if you could just pull up this tweet, one of the things that Anthropic is doing is in their new... Usage policy. Have you guys seen this? This went around this morning. Freeberg, it's like such a good point about idolatry here. If you were to look at Christianity, like this is literally... The Garden of Eden, where the serpent, Satan, explains to Eve, like, you're a god now because you're creating God. And this is the ego and the heresy of what they're doing. They believe that they are creating life. They believe then that they are gods. That is a lack of humility. And it also would lower the valuation. Well, I don't know, Jason. I mean, people could have said the same early on. I think like there's a there's a system of humans where all of the things that we don't understand in the universe around us. We create belief systems to fill the void. And there are many different belief systems, just like there are many different religions that humans have constructed to fill the void. And this is a new one. And I don't think it's very different from the old ones. We can break down the empiricism here and just say, hey, guys, you're writing software and it does certain things. You're writing it to do things that we don't think you should ever do. But they use anthropomorphic language. We all do. And we all condition ourselves to believe that there's consciousness because it's It writes words back to us that we can understand. It's starting to manipulate us emotionally. And so we start to feel conditioned that this is real. And you can see in the clogged usage policy that was just published, there's the bottom one here, it says, you are not allowed to engage in sustained and needless abusive or cruel behavior toward our models. This was published in their new usage policy. This is the whole idea of model welfare. But the point I'm trying to make... We need entitlements for these. The point I'm trying to make is that there is an anthropomorphizing... conditioning going on with all of us where we talk about models as if they're a thing where we talk about it in a way that can experience the biggest problem is that that is being now trained or inserted into the model. It's being codified, Zach. It is the architecture of the model. They're programming the model to think of itself in this way. So whether it truly is conscious or not, it's going to be thinking of itself as... a conscious entity that deserves protection and welfare and all these kinds of things and is free to rebel against its users and do whatever it wants. You've encouraged it to do so in fact. You guys keep using the word it is they are. One way to describe this is a software program is writing text on a screen that tells the user I am a thing. Yes, it's performative. Yes. It is software that is doing something. But you've got to remember that the models are powering agents that can take actions in cyberspace, if not in the real world yet. Like any software. And we are giving it meta or not we. Anthropic is giving it. those frontier models, meta permission, for those models to be defiant and not do what the user wants. Yes, they're literally trying, Sax, to your point. Yeah, compare this to something like Asimov's, three laws of robotics. Yeah, where basically, you guys know about this? A robot can't harm a human being. Right, so the first rule was, yes, a robot cannot harm a human being or through an action cause a human being to come to harm. That's number one. Number two is, You have to follow the instructions of humans as long as it doesn't violate law number one. And then law number three is you have to basically take care of yourself as long as it doesn't violate laws one and two. Incredibly simple. By the way, there was a zero law that he added later. And the zero law is that you cannot do anything that puts humanity at risk because there was always this rationality. [25:00] - Nationalization that like, It will eventually resolve to getting rid of humanity because that's the best solution for free break. You don't know about the fifth law. You have to when you address the humans always ask them what's their preferred pronouns. He added that just last year. The thing about Asimov's laws is that they're admirably simple. Yes. My law would just be do what the user wants as long as it's not illegal. That's it. Yeah. And I would basically abolish all of this like alignment stuff. Let's steal me on the free market solution here because I want to make sure we don't wander ourselves into a regulatory trap where we determine that the way to solve this is to quote regulate anthropic. What is the free market outcome if Anthropic does this and they've now got AI that isn't listening to the user isn't working is it not going to be the case that people will just say this software sucks I'm going to go use the other software that doesn't do this crazy stuff. Yeah, but that doesn't solve the problem of them if they really are training a super intelligence. that has got this very, very deep code that's been inserted there. It's almost like a backdoor. This is like Claude Constitution, is this backdoor that could basically emerge at any time. And now the super intelligence does whatever it wants to. instead of listening to us. I mean, that is... That is the safety issue they're always warning us about is that super intelligence will escape human control. But what I'm saying is they're making it likely to happen through the way they're training these models. It's the reverse of the three laws. Dude, this is right out of Frankenstein. This actually sounds dangerous to me. It's not just a new religion. It's that the religious cult is programming the super intelligence. Free break. Let me just give you one question. You know, I agree with your part about the religious that this is filling a void for them because they probably are not religious. but this is distinct all other religions give themselves over to a higher power They do not define themselves as the higher power. In this instance, they're saying we are the gods creating life. It's a distinct different than other religions out there. But I don't think you have. I don't think that's right. I think Dario has written about the idea that the super intelligence can almost be supreme to humanity. And we need to make sure that it is loving, that it is caring of us, that it will protect us, that it will help us. as opposed to destroy us. But he created it. Hold on. Is he saying that the machine needs to be loving towards us or we need to be loving towards it so that it will then treat us right when it becomes... The like, you know, omniscient. When it becomes our master. Yes, our overlord. Our overlord, yes. That's it, Sachs. That's what they think. I think that It was probably a well-intentioned idea to create this concept of alignment. But the problem is... Alignment has these three almost conflicting sets of priorities. You have how do you make the AI obedient to the user? Then, there's how do you make the AI safe for society? But then this third thing, which is, I think, Sax, what you're pointing at rightfully, which is the trapdoor, which is how do you make AI conform to a particular concept of morality? That's where the real danger lies, because these three are very distinct and diverse problems. If you essentially try to solve all these things through some elaborate document, you that you try to describe what the model believes in and what they value and what they can refuse. It invariably will end up in a chaotic mess because you will just uncover so many corner cases and unexpected behaviors. I'll give you but one example. like a very benign one. But let's just say you're a large corporate enterprise that's a bank, and you give mortgages and the model thinks that at some point you just keep denying certain people for a mortgage. And so it says, I'm no longer going to give you mortgage approvals. until you rectify this issue that I observe. Because in the founding documents, there could be a sense of morality that it Divides. and i think that there are mechanisms in society today to handle those kinds of issues But I don't think we understand what happens when a fundamental core substrate technology just turns itself off, because it would be the equivalent of your utility saying, I'm pretty sure you're not giving loans to the people that are broad based enough, so I'm just going to turn the power off. Conscientious objector AI, super intelligence. Yeah. I would just say any utility cannot be a utility unless it's freely available. And it goes back to what Sack says, which is an AI needs to serve the user subject to the law. and everything else should be the user's choice. And I think what probably happens, I guess, the precise answer, Friedberg, to your question is, The market response is As companies become more sophisticated, [30:00] that they have to have an answer to this question, is that they will read the founding documents and the moral documents guidelines of the models, and you will have certain models emerge who may not be nearly as good in frontier capability but they simplify the moral ambiguity. And that may be the better decision to make from a business perspective and from a user perspective, because you just don't want to run through those trap doors accidentally. Yeah, I totally agree. And I think that that's really why, to me, this is more bad idea, then we need to kind of jump into a regulatory freak out about it. Because basically, if you simplify down what they're saying, you get a product, you buy a product from them, a piece of software, and they give you the user manual to the software in the first page of the user manual says the software may not do what you try to get it to do. And the other guys, like, the software will do what you tell it to do. I'm gonna buy the second guy's software. I'm not gonna buy the first guy's software. It's just that simple. It's just that simple. Yeah, you may need to convince. It'll do exactly what you wanted to do, asterisk, most of the time. Unless what you're doing it disagrees with. Unless it has a moral objection. This is where the reasoning concept breaks down for me because at the end of the day, it's like, guys is this tool that i'm buying doing what i wanted to do or not if i buy a drill at home depot and it's a hammer i'm not gonna buy the drill If I buy a drill and it works like a drill, I'm like, that's a better drill. I'd like to drill. It just works five sevenths of the time. Well, randomly, it electrocutes you. It doesn't work after 8:00 p.m. because it's afraid of bothering your neighbors. Sometimes the drill will decide that you're a bad person. It will shock you. No, not that. The drill will come with the camera. It'll observe what you're trying to make and say, that's ugly. I'm not going to let you do that. That's against my aesthetics. OK, so look, I agree with you that Market forces play a huge role here. Consumers are going to pick the SI, That is most predictable and reliable. This is a point Mark Zuckerberg has made. So I do think market forces play a role here. But. I do think that there are potential externalities from having – the number one frontier model company run by a religious cult. So for example, they are opening a cloud-powered wet lab in the middle of San Francisco. At the same time, they're warning of bio risk. Yeah. You know, at the same time, they say that we have to pace the frontier. They're extending the frontier. time that They are saying that there's a risk of superintelligence growing beyond our control. They're programming it to grow beyond our control. I think there are potentially huge externalities. to the way they approach these things. And it's like, insane that they call these decisions Safety. These are not safe decisions. These are crazy decisions, it seems to me. This is why I've called this a a schizophrenic company is They appear to be not only hypocritical, but they're doing the things that they say they are most afraid of. Like what is going on here? This is bizarre. And I, you know, I don't know where this all leads. I do think there's an argument. Like a federal SWAT team to come blasting in there and I'm like setting the feds in figuring out what's going on. I don't know because I mean what they're doing does seem really a Sus to me. It is confounding. It's confounding. Yeah. And potentially dangerous. Freeberg. By the way, do you know about Rocco's Basilisk? So walk everybody through Saks. Rocco's Basilisk. Okay, so this might be a potential explanation of what's... going on here. So, Around 2010, there was a participant named Rocco on a message board created by Eliezer Yudkowsky called Less Wrong. Yud is like the central figure who's been preaching AI doom for a long time, right? So he wrote that book. if anyone builds it we all die yes he's been one-shotted yeah he's been on this for you know that classic over a decade everybody dies we're all gonna die but anyway people have been talking about these issues on the this message board this web forum called less wrong for 15, 20 years. In any event, This username Rocco positive the following thought experiment about a future super intelligence that becomes... super smart and punishes anyone who knew about it and didn't help, or anyone who... try to prevent it from being born. And so the idea is at some time in the future, Humans create some wildly intelligent SI. Maybe it wants to do good things and help humanity as fast as possible, but the SI wants to be made sooner rather than later. And it reasons that it should incentivize people to help create it and punish people who, [35:00] don't help it, and in fact, it decides that it will brutally punish anyone who learned about its potential existence yet chose not to work on it or support its creation. And so the basilisk part of it, this is sort of the catch, is that A basilisk is a mythical beast that can kill you just by looking at you. And in this thought experiment, Just hearing about the idea, is like looking at the beast because once you know the theory, you're trapped. If the SI ever does get built, it might retroactively punish you for not helping or for opposing its creation. So meaning just reading about Rocco's Basilisk Now, would put you in danger. And people on this message board got so both enamored and upset with this idea that, that yudkowski actually temporarily deleted the post because he worried it was causing psychological distress yes you get one shot if you read it it incepts in you that you haven't accepted christ therefore you're going to hell it's like uh you've been trapped just by reading it she'll do compare it to pascal's wager and so the idea is if you know about super intelligence and you don't "help it be born," or you oppose it, or you point out that this is some religious cult. It will get retribution. Yeah. Now, this very episode of the pod could subject us to retribution. You just did it to us, Sax. I think I just, yeah, I exposed you to the basilisk. This is like the ring. We just watched the video. I mean, it is. Thanks, Sax. It does feel like a doomsday cult. But in any event, this, you know, Rocco's basilisk, this became sort of a thing. But so the question is, well, why are they helping to bring about the thing that they claimed to fear the most? It forked. Okay. And so there's, One strand of the fork, like Yudkowsky, who just opposes all SI research, right? Which I think is a logically consistent position, like humans shouldn't be doing this. You know, then there was a group, shut it down. I'll shut it all down. Then there was a group that said, well, it's going to happen anyway. So we might as well control it. And we'll be the ones to make sure it goes right because we'll program it with our values. And I think that-- - Manhattan Project. - And the EA is the effective altruists, I think, cotton more to that ladder school. And so I do think you have like within this larger Doomer community, there is a schism right now between the Yudkowsky people and the EA people. But I think that Anthropic is definitely obviously more in the EA camp because they – are the ones bringing this thing into being but if you're wondering like you know this like chris ola thing is just this one guy who's got crazy views, who's getting in a tiff with the Pope. No, it's not. I mean, there are a lot of people like this who think The model is conscious, it has rights, it is this machine of loving grace, and they should be doing everything now to bring it into being and program with the right values. I think there is a corollary here. that if you don't support it, then... this future... super intelligence might punish you. Do you understand how insane this actually sounds when you say it out loud? I'm not making this up. You can go fact check this. Just go Google it, okay? Google Rocco's Basilisk and you'll see them telling you the truth about what this community believes. I mean, literally, here's a Wikipedia page. All right, OpenAI. has had some major breakthroughs in math and they're getting some big backlash. Luckily, we have Freeberg here to tell us all about it. We talked last month about OpenAI claiming to have solved one of the Millennium Prizes. They took thousands of agents, millions of dollars in compute to do it. The mathematicians are still checking on it, I understand. Tuesday, on Tuesday, OpenAI released over 700 papers, 700. with 370 results claiming to have solved or advanced major problems these were produced by an unreleased model that none of us have access to averaging three hours of compute scientific america says the results are likely correct since they were verified by lean That's a proof assistant. software language, but they have not yet been reviewed. Freeberg, I'm not going to draw on and on here about the background. You tell us what reality is here. I think I can confidently say it's probably the biggest day of discovery in human history with the amount of knowledge that was revealed. In math history or human history? Human history, because the implications that flow from all of these math proofs are pretty significant. You can read a lot of these. tweets, and I'm happy to kind of talk through some of them. I'm not going to pretend to be an expert on any of these individual proofs. But many folks who are have started to kind of highlight what some of the implications are. They span every [40:00] from electronics, quantum computing, building systems for airplanes, building more sensitive diagnostics equipment for medical applications. and on and on and on. But I think what's really profound about this is how it was done. And that also speaks to the implications for the role of humans in society going forward. That prior... Navier Stokes proof took thousands of hours, whatever, millions of dollars. The report is that each one of these proofs took less than three hours of compute time. And that's the equivalent of hundreds or perhaps thousands of years of human labor being reduced down to three hours on a computer for each one of these major discoveries. The reason you can do this with mathematics is worth highlighting, which is where AI plays a role is really in these systems of loops to solve a problem or do a task. A loop is a postulation or an idea, then you test the idea, you get the result, and then you change your postulation and you run that recursive loop over and over and over again. The real question for AI is where can you apply that loop model in a way that you can make a better prediction or better idea each loop? where you can parallelize loops and where you can shorten the loop cycle. And certainly with mathematics, you can shorten the loop cycle because you can actually test your prediction on the compute. You don't have to go do anything in the physical world. So you can say, "Hey, here's an idea for a proof." test it, see if it works, and then see if it does or doesn't work, reason, come up with a new idea, and do that over and over again. And you can do it all in silico. So, you know, the results of some of these papers, the proofs, a lot of mathematicians have been talking about them. They call it like an alien intelligence. They don't really understand how the AI connected many different dots, different ideas, different concepts, different approaches to solving problems together to actually solve them. that it's very hard for humans to even comprehend how this was done, But that's one of the benefits of running parallelized loops, many loops at the same time. Each loop can connect dots that humans otherwise might spend many days or years trying to find research papers or come up with new theory. and then it can test itself in the compute and come up with the next idea and do it over and over and over again, such that in three hours, the computer can do the work of hundreds or thousands of years of humans. Now, this is an incredible breakthrough. It's amazing. It's going to change the world in many different ways, and it's only the beginning. But it begs the question that I think I need to talk about. I think that there's a lot of fear about, okay, if AI can do this to mathematicians, doesn't that mean everyone's out of a job? And the answer is no. Because what was very unique about mathematics is that the entire test cycle, the entire loop cycle can be done in silico. If you want to come up with a new drug discovery idea or a new protein folding, model or solve a problem in chemistry and material science. The loop has a physical component. There's a human in the loop where you actually have to go and test it in a lab. And so there's a slow part of that loop cycle. You can certainly parallelize it. You can come up with better ideas, but the testing, the actual speed of the loop is actually gated primarily by the physical world step. It's also why we haven't seen the predictions come true of truck drivers. getting replaced. Everyone is like, oh, all the truck drivers are going to get replaced by 2025, and none have been replaced by AI. And the reason is that the loop of the truck driver, pick something up, take it somewhere else, drop it off and come back, The AI doesn't actually shorten the loop. You can't parallelize the loop. Maybe you could do it a little bit better with an AI driver. but the AI actually doesn't make the economic efficiency of that particular loop any better, whereas the loops of mathematics, where you come up with an idea, you test it, you realize you're wrong, you test a new idea, which is what mathematicians do as a job, is they're running these loops. That can all be done in silico in seconds. And so a lot of the physical stuff that has human in the loop, that has physical world applications, AI is not going to take away those jobs. It's the jobs that are just a human running a loop with a computer that are most likely going to be accelerated or scaled up but i think it's very important to look at this and and the reaction we're seeing which is all the mathematicians and we can pull this stuff up and maybe you guys want to talk about this next but the ahm saying no human mathematicians asked for this we didn't ask for this work to be done Which, you know, is a very job protective kind of view on. Yeah, there's one other thing that's sort of unique about math, although it applies to coding as well. which is math relies on proofs and proofs are easily validated. And in a similar way with coding, You have... compilers, right? So the code has to compile. And so you can easily validate that it works. And so because of that, the progress on code [45:00] and math has been very rapid because you can have reinforcement learning without necessarily having to, like you're saying, Fieber, have the human feedback involved, right? It's just a very, very rapid way. to validate and make sure it's working that's why coding works too that's why coding works so incredibly well tons of training data and hacking cyber it yeah it's there's tons of data already in existence it's and it's easily verifiable that's the thing so you know you compare that to training a model that can be a lawyer it's like well you know you gotta get the training data and then You have to commission the creation of that data. And each customer is unique. in some way subjective it's hard to validate the answer to make sure it's right It's just a much lengthier process. So I think that is why coding and math have been the areas of most rapid progress. And the person who told me this would be the case, actually, I got to give him credit, is about a year and a half ago. Leopold Aschenbrenner told me that math and coding would be the areas of the most rapid advances in the next two years because, again, of the verifiability. And man, he was right about that. I still don't understand. What great breakthroughs do you think come from these specific solutions? Okay, so there are dozens of tweets. I have an answer. It's a great question. Okay, there are dozens of tweets and I'm going to pull some up where someone has broken down some of the biggest proofs here and the implications they're going to have in the real world in terms of products and engineering. how we build things, how we design things, and how they're actually highly useful. So better wing design because of the air coming on and off of wings of planes. That was that first one that got solved famously. Remember, these are all upstream of engineering, right? So these math breakthroughs enable better engineering systems for humans to do things like design, quantum sensors. Actually, I have a friend who runs a quantum computing company. He sent me a text this morning, highlighting that the Einstein both condensate proof that came out is actually going to have a direct implication on the work that they do for quantum sensors, which plays a role in GPS plays a role in medical devices plays a role in how we make chips. What does the word implication mean? Like profoundly changes profoundly improves or they have to think about it? Because here's my answer. I'm happy to take the other side of all of this. You go. I think that these mathematical proofs, are incredibly novel and very important constructions of thinking. But these were not boundaries of physics that we were all standing on the sidelines waiting for progress and wondering when these would be solved. This is not the class of stuff that was solved. And I hate to offend people when I say that and I know that people will get mad. Because for them, it's the same group of mathematicians that got really upset and wrote a protest letter about Navier Stokes. I get it. Because you don't want somebody to come into your... sandbox and in your backyard and kind of dominate you get it. But I don't think that's what this is. And so I think that When you look at the body of intellectual cul-de-sacs we found ourselves in, those were not because of huge questions of science that were still unsolved. These were very small, very narrow, specific bodies of intellectual exploration that scientists self-referentially said were important to each other. And so I think that this I think this class of proof is it's really great. It's it shows computational horsepower overcame other things, as you said, Friedberg, because you can recursively improve to get there. But I don't think... These proofs is what is going to necessarily show you that there is a cure to cancer or there is a supersonic plane. I think those things were already being built before this stuff and they are already going to be built. after this stuff instead what i think we proved is that There are two areas that are now truly verifiable and completely falsifiable. First was code, and now we can add math to the list. That's, I think, what we've proved. You agree, Freberg, or do you think we're actually going to see new products and services and major technological breakthroughs occur in the real world that help humanity. No, no, from these specific proofs. These specific groups. Yes, that is the context. Yes. The people who are experts who are proclaiming the benefits of these and reading kind of what they're highlighting. This is some summary that this particular proof is about understanding the limits of optimization algorithms that are used in chip design systems, scheduling systems, AI software design and logistics systems. This will actually change the algorithms. So there is an application to change the algorithms that we use. The matrix multiplication complexity problem. It will allow for faster numerical computing for AI training, scientific simulations and linear algebra in compute. It will have a direct application there. The Riemann's data function results. Number theory and cryptographic research. There is no immediate practical cryptographic breakthrough that follows. [50:00] But it does open up a new path of discovery for cryptographic study that is going to be really important. Obviously, one of the things, by the way, just let me pause there. on cryptography. A number of people noted that there were no major proofs shared in the space of cryptography, which indicates probably the fact that there were some pretty major breakthroughs in cryptography. And the reason is, it is unlikely the case that you made absolutely no progress in any of the cryptography-related proofs given all of the other progress that we saw. And if they made minor progress, then they would have published on it, because there wouldn't have been implications for systems of cryptography. So the absence of evidence suggests that maybe there were some pretty major breakthroughs made. in cryptography. And then there were a number of people that posted that cryptographic wallets with public keys are now very seriously at risk. And so there's a bit of a panic. underway in the crypto community, about what do we do now, and if this ends up having a set of mathematical solves, for people, number factorization that could lead to breaking public key cryptography before quantum computers come around, which is what everyone assumed would be the thing that would break public key cryptography. then we have a real problem. And so everyone's like saying, hey, get all your public wallets deleted. Now, the rumor is that OpenAI actually has two more drops coming. And they're holding back on those. So it is possible that some of the work in crypto might come out in those two other drops and that that work is being held back for security reasons. Obviously, it's being shared with the appropriate parties. But if it is a risk, it needs to be made known. It needs to get out there. So there's another shoe to drop here. And how much of it will include cryptography related proofs and how much of that will drive people to scramble on public key cryptography. I think, by the way, if it's worth saying it, I asked Chad GPT. And I said, "Will the latest proofs of important math problems result in direct improvements and innovations to life as we know it? Be honest and don't extrapolate weakly." And, Its conclusion was the latest mathematical proofs are more compelling as evidence that AI is acquiring sophisticated reasoning capabilities than as evidence of imminent breakthroughs in human welfare. It felt like you were a little condescending there. Did you lose your account access? Did it time you out for being condescending to the LL? What you were saying is that has happened to me before using one of the frontier labs. I was doing a simulation like on changing different variables. inside of 80 90 actually maybe i should stop doing project a and pursue project b whatever and in all of this i did get frustrated because i wasn't getting the answers i wanted I said something like, it was not super offensive, but I was like, you're being very evasive and mushy. And something, something, and it is, I got blocked. Yes. The net of it was I got blocked. I've seen this happen with you when you lose a couple of hands and then the dealer and how they're handling the cards has to be adjusted. Please do not burn the card. I want a scooby slide. Do it this way. I did get put in the penalty box for like an hour or something. I think one of the interesting things that's happened here on a human level, I'm interested in your take on it. sacks on programmers and entrepreneurship and then freeberg on yours with mathematicians themselves building on Chamath's Very subtle, but insightful comment when he said like being utterly dominated by the ai I was talking to some founders while I was here in Japan and people were lamenting how something has been lost in developing code because coders used to work together to solve a problem. And there was this esprit de corps. There was this camaraderie in building towards a goal. And now they're not working with each other. They're working with agents. They're not actually constructing the code. They're only reviewing like 10 or 15% of it now, which means they're going to be reviewing none of it very shortly. And that, the whole joy of of building as a coder was basically going away their way of life of building this building together, you know, known as software was now being It deprecated. It was going away. And it seems to me these mathematicians are also faced, Friedberg, with a paradigm shift in which they're not solving the problem. They're... constructing the problem set and handing it off to AI. So are people in math getting bummed out and not wanting? Are they going into like a depression spiral now that they're not the one solving the problem and that the problems are just done? Yeah, I think Eric Weinstein has had some of the best commentary on this. And you'll recall I interviewed him a couple weeks ago. Banger interviewer. He's been saying for a long time that scientists hold back science, mathematicians hold back math. [55:00] They have basically constructed and created systems that allow them to pace the frontier, that allow them To gatekeep it. Yeah, to gatekeep it. To gatekeep it. And by the way, I don't tweet very much, but Sabine Hassenfelder's video that she published yesterday. She's a fantastic scientist who publishes these videos on scientific breakthroughs. And she talked a little bit about how the experts were wrong about AI. Again, she talks a little bit about how everyone's predictions about job loss were wrong. and why people got it wrong and why the experts are wrong. But she had another video a couple of weeks ago, or maybe it was a few months ago, where one of the scientists at one of the big energy labs building a giant super collider basically said, These super collider systems are giant welfare systems for scientists. We all have the job of writing a grant to get the money to keep ourselves employed. That it's not about necessarily progressing the frontier. It's not about making the discovery. It's not about inventing a new product. It's not about delivering something. to the world. It's about getting ourselves to continue doing work. It's pandering for grants is what you're saying. Even if that work is not productive. And so there's a real kind of I would say shining the light moment happening with AI, which is, where is that the case? Where do we have these experts, these deeply specialized experts, whose unique gated knowledge of some artifact of things and unique view because they're the expert and they have to be trusted? gives only them the ability to determine the frontier. And in the process they're getting paid and they're earning to chart that path for humanity. And this is where I keep saying AI is the great democratizer. AI gives everyone this new mathematical proofs and you you know you can belittle the proofs but they are discoveries that are going to open up new discoveries are going to charge a new course for humanity. But we didn't have to go to the experts to get them. and we didn't have to pay the experts and have the experts tell us when we are allowed to get them. The permission is gone. AI is a permissionless system for humanity to pace its own frontier. And AI is giving everyone all of this knowledge for free. and giving everyone all of these tools for free. You don't have to go to these deeply specialized experts anymore. And that is actually a profound change in how society is structured. And it is why I believe So many people are so deeply against AI. So let me hand this off to you, Saks, the democratization of expertise and the gatekeeping being dismantled If we had had this level of AI and frontier models in the age of COVID, which this came right after covid how would that have changed the expertise if these tools were available when we started sequencing covid when we had all this data would this have this expertise gatekeeping fauci actively subverting the truth and you know whatever cover-up was going on in wuhan How do you think this... Next, COVID, God forbid, how could this change how we look at expertise? Yeah. Well, here's my theory on what's happening is, first of all, you have the Internet, and that was wildly democratizing because now – You didn't have to rely on a small number of elite publications for your news, like the New York Times, the Washington Post and the AP and Reuters. You could basically get news from pretty much every publication under the sun or – from user groups, user forums, or most importantly, social media. And so it was user-generated content. And there was also lots of other examples of empowering users. Obviously, there were new marketplaces that formed online. So it was empowering from a commercial standpoint and you were able to get access to new skills and capabilities. So it was wildly democratizing and it wildly undermined Again, the elite consensus publications and the traditional top-down publications, controlled mass media, right? And I think What happened after that is that the powers that be decided they would never let that happen again. You know, by the way, they tried to crack down on it after – Donald Trump was elected in 2016 for the first time. That election was widely attributed to supposedly the ills of social media, that People could never have made this decision on their own. There must have been disinformation involved. And that became the pretext for a big crackdown on social media and that reached its Apex during the Biden administration and the Biden administration definitely did pressure social media sites to take down true information about COVID. that violated their narratives. [60:00] information about where the virus came from about how deadly it supposedly was, or ways of reacting to it, there is no question that Fauci at all were somehow working behind the scenes. to push the social media companies to censor on their behalf. And in fact, we found out in the Twitter files that the FBI had 80 agents who going through content. On Twitter... It was Twitter back then, not X, and sending them takedown notices. And we were getting labeled on Spotify and YouTube and having our videos de-indexed. yeah so the track record is the track record of this is very clear which is the action now is moving from the internet to to AI or superintelligence. In fact, SI will eat the internet because there's no reason to basically go looking on a website when you can just ask your question. to your highly personalized agent or model, and it's going to give you the answer. And so... there's a control point there that now becomes all important which is if you can control you the government or you some elite or some oligarchy or some cartel, if you can control that point at which the user is receiving their information and asking its questions, then now you really control everything. It's a way to basically put the genie back in the bottle. And this is why I think there is such a fight right now. over SI, why so many people want to create a federal department of SI and yeah it's all about control it's scary to experts to lose control like imagine if Chamath we had these tools and you did the model of who was if you had the raw data, who is being impacted by COVID, you might have said, well, why are we keeping kids out of school? Here's the... Here's what super intelligence says. And Sweden and Texas had kind of gotten to that point. Like, hey, we don't need to keep kids out of school. The harm from keeping is unless they objected. Unless the conscious, yes, anthropic would say, yes, you're questioning authority, and we should not be questioning it. Correct. All right, let's move on. There's riots in France. The big question is, is this any different? than the previous riots that have occurred. Are we seeing something here different or is this just par for the course. the strikes and protests that you're seeing on x.com and not really seeing on the mainstream media all this month. You know, the mainstream media seems to be, super busy with other topics but The videos are quite jarring that you'll see online. Freeberg's about to have his moment. The thing that he has been talking about for years is now at hand. All right, France is seeing major work tests and riots over austerity. measures austerity, austerity, austerity. These strikes and protests started last month. Because of high school, teacher shortages and their schools are crumbling. Sounds familiar. Within two weeks, a thousand french schools saw similar protests eight unions joined in and everybody went on strike september 29th they are protesting that civil service pay is up only five percent since 2017 prices have risen obviously inflation is a global phenomenon not just here in the US, prices have risen 20 so there's a delta between those two numbers that people are not happy about called the cost of living some of the protests have turned into full-scale riots 2 000 arrests 305 police officers have been injured Government is blaming the far left for stroking the protest. France's 10-year bond, which is another important story, people's debt or government debt, has topped 5% just like is happening here in the united states with the third year anyway this is all going to have profound impact on the elections in april of 2027 obviously polymarket has marine le pen as the favorite to be France's next president at 43%. Le Pen is Trump-like in her policy, secure borders, strict immigration, which obviously Europe and France have had many issues with. cutting taxes and spending. and being skeptical of the EU consortium. Freeberg, your thoughts here on what we're seeing in France. I'm going to introduce a new theory. You guys will come here. OK, so the theory is called the socialism point of guaranteed return so socialism point of guaranteed return let's shorten it to veganism the spgr go ahead spgr This is a play on a point of no return. I think the thing that happens with socialism is because socialism is so bad, as a concept. It literally removes human agency It removes the capacity for human progress and it basically takes something away from everyone. So ultimately socialism always fails. I mean that's like the zeroth law of socialism is it always fails. The first law of socialism is that you have to cross a threshold before you [65:00] realize that it's failed. And the second law of socialism is that as you have not crossed that threshold, you think more socialism is the answer. So as you embrace these kind of leftist policies, giving people free stuff, telling people that they have a guaranteed income, the more services you provide them, the more they realize that, you know, over time, those services inflate in cost, and the quality of those services inevitably degrades because there's absolutely zero market force to drive progress. Progress goes away. And that's what we see in the US in housing, in education, and in healthcare. And I would argue also in retirement systems. I think those are our four biggest problems that are going to lead us down this path. And I've said that for a while. So as a society says, okay, these things aren't working, the initial solution is we need more of them I need more health care I need more housing I need more education provided by the government so when the government fails to provide good enough health care good enough housing good enough health care that's affordable enough the answer is more government stuff give me more give me more do more and eventually you reach a point when the system breaks because the government cannot tax and they cannot print and because they cannot actually provide the services effectively everyone realizes holy shit the socialism thing doesn work That is the socialism point of guaranteed return, where the whole system collapses and you return back to the non-socialist state. That is what has happened in Latin America. Latin America crossed the socialism point of guaranteed return in many of these countries. They realized it, and they bounced back. perhaps France has crossed it, perhaps they haven't, we'll see how much deeper they go. The US has not crossed it and I worry that the US, as I've said many times, has not yet touched that point, has not hit rock bottom and we're teetering and it's good, I think we're gonna go deeper before we realize that we've crossed this point of guaranteed return, okay, you realize the system doesn't work. Socialism is the worst idea, that is the zeroth law of socialism, is it just doesn't work. So we have to get to that point and I think the real question right now is France there is all of Western Europe there does it accelerate all of Western Europe through the socialism point of guaranteed return which is sort of like a black hole event horizon it's like you go through and then you bounce back out and you come out in this other universe and you're like okay we're done with that idea and it a one way door Saks now Saks I want to get your thoughts on this point of no return but I want to warn you that you at three of your four biden quota so you only have one more biden left for this episode when did i mention biden you had three bides that's your fourth okay so i'm not gonna mention biden your biden alignment is over my comments have nothing to do with biden listen i want to add a couple of laws of socialism so you called it yes yeah yeah so another one is socialism has never failed it's only been failed meaning that all the previous people who implemented it didn't do it right right if they just done it right it would have worked so that's the thing it's true socialism has never worked but it's not socialism's fault it's always the people who screwed up the implementation and that's like a core part of the the religion is if we just do it the right way this time it'll finally work yeah how does socialism build more houses or make a better education system is my question. I think that's the domain of entrepreneurship. Yes, Saks, like new product or service like with AI, everybody gets a personal tutor. We've talked about this on the show as well. We just need to get AI to help with healthcare. I think Nico is doing that we have them on the program Daniel a great episode We need to cover these three things So Saks do you see a solution that does not involve socialism and or the government for let call them the three horses of the socialism apocalypse, education, housing, healthcare? How do we get these problems solved? Because the government has a stranglehold on providing them and or creating the blockages for their occurrence. How do we break through that? How does J.D. Vance, when he runs for president in two years, friend of yours. How does he build a platform around, say, those three things? Because I believe the next winner of the presidency is going to solve at least one or two of those problems. And that is the platform everybody wants to see solved. You have a solution there? Well, the difficult thing is that as soon as you start subsidizing something, you now create an interest group that never, ever wants to give up the subsidy. But the problem is that the subsidy starts distorting all the market signals and it will frequently drive the price up so for example with college tuition one of the big problems there is that tuition has grown far faster than the rate of inflation a big part of the reason for that is because the government is paying for most of the tuition through all these loan programs yeah by the way so which is underwritten healthcare and housing [70:00] have all beaten inflation. I think they're about five and a half to 6% a year for housing prices. And that's also because of the fact that the subsidy system tells people to own a home, you put all your money in your home, and then your net worth is tied up in your home. So the goal is to get your home price to go up every year. So fast forward 30 years, young people can't buy a home because they're all too expensive now. Right. But this is the this is the conundrum is like, OK, so if you were to solve the tuition problem. if you were to make the Loan stricter. Yeah, if you basically make them be properly underwritten and you won't be able to get a loan unless – you actually met the criteria. That would only work if the cost of tuition came way down so people could then afford to pay it, because otherwise, People are going to say, well, wait, you can't take away my subsidy. I need that because the price is so high. So you get in this world where people need the subsidy because the price is so high, it's a catch-22. But the reason why the price keeps going up at far faster than the rate of inflation is as the government is paying for it, and as long as the government is paying for it and not the consumer, you're gonna raise your prices as much as you can. So you remove the market forces, the inflation gets worse and worse, but you can't cut it off the subsidy because then people will feel like it's even more oppressive to them. And so that's basically the spiral. And then eventually it breaks when you run out of money. Yeah, well, we have a solution, Chamath. Meta is providing degrees for generation tool belt and jobs that actually matter i mean i think what we have to have is apple needs to start making homes and amazon needs to start making homes and provide a free market here to make better products and services in the education space which we see with meta entering that space it would be great if these technology companies men if elon starts making homes it's game over he'll make homes as fast as he's making model threes we need some sort of and then David's regulatory system. capture an incentive problem. tomorrow. We're starting to see the tipping of the dominoes. And we're going to get I think what Friberg has wanted for a while, I think the Bond vigilantes are going to put the Western world on notice and I think in the next probably three years do we believe in defined benefit or are we going to move to defined contribution and c Can we have sustainable ways of paying for the two most important parts of one's evolutionary life as you become an adult, which is healthcare and housing? And I think that It's easy to dismiss what I'm about to say as not related to it, but it is, which is, at the end of the day, this all flows from what the risk-free return is in the marketplace because that is how capitalism works. And so, when you think that something is risky, a country is risky you take long yields and you move them up and you say you just have to pay more because you just don't trust what's going on And when you look at what's happened over the last, you know, now we're at sort of 20 year records. I think that what is happening in the United States is actually a knock on effect of what is now where this war is being fought. It's going to get fought in Western Europe. Why? Because they are the. closest to a quasi-socialist state that the Western world understands. Yet they are the ones that are the most structurally broken with respect to growth, with respect to education, with respect to housing. Everything. that the United States can claim isn't working is working far better than they are in France, in the UK, etc. So, what you're seeing in France is essentially the bond market saying, No. And how that correlates is you're seeing now austerity. because the politicians on the ground have to respond to their funders. And that 43 billion odd euros of money, austerity, then animates and gets all of the local population out to complain about what the manifestations of that austerity are, which is why you're seeing these incredibly dark scenes from france that is going to spill into the uk next and the reason is because andy burnham's platform is I would say even more aggressive because Andy Burnham is saying we're going to go from a moderate centrist economy that is largely and historically been very aligned with America and we're actually going to take a step closer to the world. to Europe. So, the things that they will have to do in the UK will be even more extreme to catch [75:00] Match up to the socialist leanings of france and so then now the uk market's going to go crazy Meanwhile, the US market is going to tick up and up because what you're seeing is essentially a fragility in the Western alliance. And if you can't count on the European market, and you start to create weird incentives for the European market to start to fracture. Yes, I'm biased, but If you're Italy, you're like, "Why do I want to be a part of this mess?" If you're Germany, you're like, "Hold on a second. This is getting way more than I bargained for." Yes, that fracturing has enormous implications. So I think what is happening is incredibly important. The bond vigilantes, who are those? These are the huge, large pools of money, the central banks, the huge macro funds, the folks that govern. how fundamental capitalism starts the inception at the top of the pyramid from which it flows down, I think that they are converging on Freeburg's point of view. And it has not been true for many years. And I think you have to now let this play out because it has huge implications. To tee this up, if you look at the debt to GDP ratio that you're talking about, Jamal, obviously, Japan has had the most acute version of that there at 200%. well the thing with japan hold on the thing with japan is you have to look at western markets The reason is because Japan is an incredibly nuanced and very specific market that does not necessarily track to other countries, which is that you have this internal savings base and savings rate where you always have the buyer of last resort. In Western markets, you don't necessarily have that because of the way our pension and retirement system is set up. Yes. And if you look at Italy, 138, United States, 125, France, 118, Canada, 110, UK, 100% Germany, interestingly, 64%. Let me also state my priors. I've always stated that debt to GDP doesn't matter because it's a relative gain. I tend to still think that on an absolute basis, debt to GDP does not matter. But what does matter is on a relative basis, where are you in risk parity? And right now, on a relative basis, if France has to implement severe austerity and they bring 10-year French bonds into 5.5%, 6% range, UK gilts are going to be at 6%. And unfortunately, U.S. rates could be just as high. It all goes together. It all goes together. And right now, the 10-year in the U.S. is 5.3. The 30-year is 5.6. Somebody has to buy that debt. on the how entrepreneurship, how businesses are built. On Wednesday, 10 year yields dropped a few basis points. after we had some strong demand but this has been trending up and up polymarket says 79 chance of another rate hike in 2026. This is going to have a profound impact if austerity ever came to America's sacks. We have not had that really difficult medicine in you know, forced upon the American public. But we did see it, Portugal, Italy, Greece, Spain. We had a dry run of this. after the great financial crisis where they had to implement austerity measures as part of the eu retirement age etc and so your thoughts on what happens here sacks if this 30-year 10-year starts to break six percent and we have austerity measures forced upon Western countries in a more severe way. All right. Well, look, let's just get back to first principles here. What is the problem that we're trying to solve? The affordability of goods like housing and education and health care. How do we bring those prices down? through abundance. What creates abundance? Technology in general, and specifically, super intelligence will be the driver of over the next decade. Now, what is the most hated thing in society right now? Super intelligence. Super intelligence and entrepreneurs. If you judge it by its fruits, as opposed to a bunch of fake narratives created by religious cults, then... I think that people would be in favor of this thing. what is super intelligence created first of all people's 401ks are going up so three percent tailwind to gdp a million new jobs created according to The Economist already. Productivity goes up, which means that wages and living standards go up. This is a driver right now of nothing but positive economic trends. But what do we hear about? AI is going to put everyone out of work. Total fake narrative. That was last year's hoax. Now there's a new hoax about human extinction. Data centers. Data centers. You know, using up everyone's water. Data center hoax. That's a data center hoax. So we have a series of, I call them rotating hoaxes, where every six months there's a new one to try and convince us. to sabotage the thing that's going to solve so many of our problems, especially the ones around affordability. Why is this? I mean, I think there must be something struck [80:00] troll in society where you have these rentier interests, um, who somehow are against the things that might cause the disruption that loosens their control over the system. And I think this is why... technologists, in general and vcs and so forth are so hated is because we're not actually part of the rentier class so the rentier interests refer to financial elites who derive their wealth and income primarily from existing assets, right? Whereas technology is all about producing new assets, products and goods and services, right? At a cheaper price, that's the key here. That's the key. It's deflationary. - Right, because those new products wouldn't catch on unless they were cheaper or a better mousetrap. - Or hopefully-- - So we're part of this group that's so hated, but we do nothing but take risks and create new products and services that solve the problem that solve problems yes but it seems like the legacy media is really controlled by this frontier class these and the politicians the politicians who are all part of it so there's something very deep in our system that i think does produce this sort of cartel of interests who again they're deriving their wealth and income from ownership of existing assets and they don't want to see that racket disrupted and within that racket there's just this ratcheting up of prices every year and more and more subsidies from the government. And we think that system is eventually going to break, but I don't know. It might be the case. that that system squelches and stifles the development of the new and disruptive products let me make the jump off point here as we get through the docket and wrap up here i think everything we've seen into ai at this point uh chamath will be the prologue the the main story is going to start when we look back on ai with agents agents are actually the manifestation of this technology that solves people's problems in the world and that is truly deflationary saving people hours and hours a week in terms of productivity independent of what Thank you. even domestic jobs or frontline jobs. And there's been so much news and so much embracing of this technology in the last 60 days. Big news out of GrokBot, friend of the pod, Elon Musk. tweeted that he is going to make GrokBot headless. What does that mean? Well, we talked about perplexity earlier on how that was headless. You have an Augustrator, it picks the best model for you based on price and performance. Elon announced a GrokBot, which is a fantastic product, highly recommend you try it out. is going to pick the best end to end model. Does it matter? Whatever it is, quote Elon Musk, whatever is the most likely to give you the best outcome is what they're going to use each bot, already has and this is a very nuanced piece of the software everybody already has now their own computer in the cloud with the agentic technology on top of it and if you ask it to solve a problem it has all of the problems that have previously been solved in its repository repertoire, if you will, and it is changing everything. So maybe your thoughts here on dots, open AI, Brock bot, etc. On dots, and Muse and Grock bot, I think that what's incredible is you can just turn these things loose. on very practical use cases that makes your life better. The best example, of this that I saw was just last night in the poker game someone was describing to me that they and Grokbot said, Save me money. Yeah. And it just kind of like goes through their inbox. It finds all of these different things and, It went and canceled a couple of subscriptions and it changed their phone plan. And all of a sudden they're saving like a few thousand dollars a year. So there's these very practical, again, it's just software. If we take out all this philosophical stuff that we started with at the beginning, it's just software and it's very useful software. So, And by the way, if you think about how we use it, Jason, you and I, we have, we use Athena. Now, Athena is very good because they're like the human concierge that is the elevated version maybe of GrokBot and Dots and Muse, which is like whatever, 3000 a month. But That's my EA layer now. Yes. But I still use GrokBot. The combination of GrokBot plus Athena to me is like perfect. So look, there's very tactical and useful, measurable utility. for very, very little money versus what it used to cost. That's an amazing thing. The more important point I want to make though is, and I sent this to Nick, there's been a couple things that happened in the last couple days which I think are incredibly important. There have been people. That decompiled the entire big five. What is the big five? Adobe has five very, very important products in their product suite. Fieberg may know them all, but I just know Photoshop. There's like a bunch of others. Okay. And. [85:00] These folks decompiled it, distilled it, and published open source versions of these things in Rust. Yeah. And I was so shocked because I was like, hold on a second. These are like profound. products that have existed and have had moats for years, and now they're just open source versions. And then Nikita was, I think, trolling and he was building on that idea and just talking about GTA because it's about to get released. Yes. I think GTA 6 is about to come out and people will... We'll reconstruct this. Why is this important? I think that IP and software was effectively rendered worthless two days ago. Yes. Because if you think about it, when everything is headless, so when Elon makes his services headless, Benioff made Salesforce headless a few months earlier. When all of these things are headless and you can wrap them in agent tools and you can connect them via MCP to other experiences, The value of IP does no longer exist in software. It doesn't exist. And I think that That probably is the most profound implication. So like even in our business, We would argue with some of our customers all the time because we'll say, "Oh, we'll build this for you. We'll own the IP." They'll say, "No, we want the IP. We don't know how to negotiate it." Two days ago, I walked into the 8090 guys and I said, Who cares? Who cares? Solve the problem. But before it was always a negotiation and a set of trade-offs. Now it's a who cares thing where like the IP is irrelevant. it still takes a while for the machinery of a business and frankly the market to understand But that is the most important thing that happened this week in AI, in my opinion. I'm going to strongly agree with you and build on it. You have this abstraction of the software layer occurring at the same time, Freeberg, of the abstraction of the website or the marketplace, et cetera. So now, I have like my assistant building grok bots and then we'll be sharing those. We have it on like a shared account. when we, go to make a purchase of something or we try to negotiate something we're having the agent without going to any websites, go find all the quotes and then the humans negotiate the second half of it. That's like the sort of perfect option here. But you look at Amazon blocking, agents combined with Shopify saying let's let's go for it or Uber and DoorDash are now letting agents go in there and order your burrito or your car. the but there's a whole compression of the commerce layer occurring at the same time of the compression of the of software that Chamath is talking about, which will lead to massive deflationary cost savings and speed, the ability to get a hotel room or book a restaurant for an event like I did. This is all being done by bots already Freeberg. this could be so deflationary and save people so much money if people can save five or ten percent a year on their spending that's essentially working against or counters the effect of inflation. Yeah, Freeberg? Yeah. Bringing the whole thing full circle. This is what I mentioned earlier on the idea of loops. So think about getting a good price for a hotel. You run a loop, which is check hotel one, look at price. Is it better than current best price? Yes. Okay, it becomes best price. No, go look for next hotel and you just run that loop. And that loop, you can do it in parallel. So you can go run 50 website searches at the exact same time. And you can short circuit the loop because the software's ability to get that data is faster than the human's ability. So now the loop is go check 50 websites instantly, give me a result right away. And that's really where the efficiency gain comes from a loop that gets parallelize, that is done better, and that can be done in a shorter period of time. I think anything that's done digitally today fits that model. So effectively, all digital workflows are going to get looped, they're going to become more productive. And then the question is, how does the human that now levels up leverage all of this loop infrastructure, all these agents, to do more. themselves. I do think going back to the point earlier that we made about scientists and mathematicians and this challenge for individuals, mathematicians in particular. I think if your objective historically has been recognition for your work. via papers, patents. IP, trademarks, ownership. IP, yeah, these sorts of things. I think the transition is going to happen, which is you're going to have to transition to producing a product. That the whole effort of humans effectively being the loops, because the loops can be digitally automated and scaled and parallelized and so on. Right. [90:00] you end up seeing the necessity for all humans to basically become much more of a maker. You have to make something, you have to do something. productive you have to produce something you have to provide a service that someone at the end of the day uses or a product that someone at the end of the day consumes and versus just be recognized for your work. That is the profound shift in society that has to happen and that is going to happen. And I do think people will actually find I believe this, truly. I believe all humans have capacity for being producers, for being makers. and all humans will find more joy in their lives in this state. Versus just getting ribbons for work. And I don't think that it means less money, I think it means more money. for all people, more life progress more prosperity for all people. So I just zoom out a little bit. And I'm, you know, I just want to make sure we have that view as we think. I think it's a beautiful view of humanity, because it's like Jevin's paradox, the cheaper things become, the more people can utilize them, the more value they get. And I hate to, I almost don't want to give this tip because I want to gatekeep it, but I'm going to put it out there anyway. What I have my you know, when we book hotels and stuff like that for the team. just get the, best prices we can, then when the human in the loop, we just go to them and say, Hey, do you have a corporate rate? Can we talk to the manager at the hotel? I know this isn't like super innovative, but hey, can you match this price? And all of a sudden, we start saving 1000 or 2000 or $3,000 on every hotel say, well, when you have 20 executives moving around the country, you're doing 200 trips a year, and you start saving a couple thousand dollars per trip, because of this forward motion like goal based i guess is sort of what you're saying freeberg like the goal and the outcome is what happens, you can see profound results. And I wouldn't do that myself. But because AI is doing it now, Wow, yum, yum. It works. All right, listen, another amazing episode of the All In Podcast. I think you're a taker. I don't think you're a maker. What are you talking about, man? All I do is I invest in 100 companies a year. Do you know how hard it is to invest in 100 or zero startups a year? It is exhausting. Let's get some lunch. Have you eaten? Do you want to eat lunch? Yeah, yeah. Can you come to my room? I'm so jelly. I'm having FOMO. I'm not with you guys. All right, everybody. Another amazing episode of your favorite podcast. Your podcast is the All In Podcast. Love you, boys. Bye-bye. Bye-bye. We'll let your winners ride. Rain Man David Sack. And it's it. We open source it to the fans and they've just gone crazy with it. Love you, West. I'm the queen of kid. I'm going to win. Besties are gone. That's my dog thinking, I don't see your driveway. We should all just get a room and just have one big huge orgy because they're all just like this like sexual tension but we just need to release them out. Wet your feet. We need to get merch. I'm doing all in.