AI Giants Will Pivot to Non-Invasive Brain Writing, Huberman Says

Huberman表示：AI巨头将转向非侵入式大脑写入技术 人工智能开发者将演变为生物技术公司，直接向神经回路写入信息，以非侵入式的清醒状态调控取代粗放的药物疗法与开颅植入物。 Andrew Huberman博士概述了人类表现的生理与技术极限，首先探讨了个人生物能量与下调机制为何是高绩效人群的终极瓶颈。他审视了神经技术的演变，认为该领域正在从被动记录转向通过聚焦超声、靶向病毒载体和直接自主神经监测等手段向特定神经回路进行非侵入式写入。Huberman批评了当前公共健康文化中的耸人听闻倾向，解释了主流AI领跑者涉足生物湿件接口的原因，并指出清醒大脑状态的分类是神经科学中最关键的未解难题。对话最后探讨了如何在迎合受众与保持内在科学好奇心之间取得平衡。 神经科学正从被动观察转向主动神经调控，其核心技术前沿在于利用聚焦超声和基因靶向病毒载体等工具，对神经回路进行非侵入式写入。 (22:07) 包括OpenAI、Meta和Anthropic在内的主要人工智能领跑者正处于向生物技术公司演变的节点，因为它们正试图将硅基计算与人类神经组织的模拟、渐进动态特性连接起来。 (23:30) 与划分为明确阶段（N1、N2、N3和REM）的睡眠结构不同，清醒状态下的大脑完全缺乏标准化的科学分类体系或精确的诊断测量框架。 (57:34) 能够区分生理性不良压力与有益的高唤醒良性压力的新型可穿戴袖带传感器，正在推动自主神经系统直接测量技术的发展。 (40:00)

Neuroscientist Andrew Huberman argues that human energy and nervous system regulation have become the central bottlenecks for high performers, projecting that neurotechnology will inevitably advance from passive tracking to non-invasive, closed-loop neural modulation. Biological Baselines and Downshifting Mechanics (0:00) Huberman emphasizes that elite founders who operate sustainably must identify their personal baseline sleep requirements rather than copying genetic outliers like Elon Musk. While early-stage sprints often require brute survival—an intense 'don't die' phase prioritizing the avoidance of chronic disease over intricate optimization—long-term output demands the ability to sever cognitive momentum. Because the brain functions as an anticipatory prediction machine, downshifting into rest requires satisfying three physiological prerequisites: disrupting planning thoughts through direct sensory focus, reducing heart rate via extended exhalations that engage parasympathetic vagal braking, and extinguishing proprioceptive spatial awareness. The Retrenchment of Public Health Discourse (18:24) Characterizing current public longevity commentary as overrun by theatrical showmanship and unqualified influencers, Huberman anticipates a consumer backlash against convoluted daily 'biohacking' stacks. As the public tires of unsustainable optimization protocols, he forecasts a market downturn that will weed out social media theatrics. In their place, Huberman expects a wave of credentialed physicians, pharmacologists, and neuroscientists to elevate communication with rigorous, evidence-based education. The Convergence of AI and Neurotechnology (23:30) Huberman predicts that leading artificial intelligence companies—including Meta, OpenAI, and Anthropic—will ultimately evolve into biotechnology firms by pursuing bidirectional neural interfaces. While invasive craniotomies will remain confined to clinical conditions like speech restoration in locked-in syndrome or tumor resections, commercial scale requires non-invasive write technologies such as focused ultrasound. Paired with direct autonomic sensors—such as wrist cuffs distinguishing distress from eustress—adaptive machine learning models will dynamically titrate graded neural stimulation to modulate alertness, motivation, and rest without relying on blunt systemic pharmacology. Decoding the Architecture of Waking States (57:34) While sleep stages and circadian mechanics are well mapped, Huberman identifies waking brain states as the primary uncracked scientific frontier. Science currently lacks a coherent physiological taxonomy for daytime modes such as executive focus, passive mind-wandering, or frustration. Recent hippocampal memory experiments by Susumu Tonegawa and Mark Mayford demonstrate that learned behaviors can fire even when tagged neuronal ensembles are stimulated simultaneously or in reverse, exposing how little neuroscience understands about higher-order thought generation. Developing targeted write technologies will require bridging this gap while navigating the risk of destabilizing interconnected neural networks. A core unresolved question is whether non-invasive writing tools can achieve sufficient spatial and temporal precision to guide waking cognitive states without triggering off-target effects across distributed, multifunctional neural circuits.

## Background 本期节目由商业与投资播客网络 Colossus 旗下的《Invest Like the Best》（第 494 期）于 2026 年 10 月 6 日发布。节目邀请了斯坦福大学神经生物学教授、《Huberman Lab》播客主理人及《Protocols》作者 Andrew Huberman，围绕神经科学从传统的“信号读取”走向高精度“状态写入”的技术前沿展开深入探讨，并剖析了顶尖 AI 公司向生物科技演进的终局趋势。 ## TL;DR 对谈首先从现代高绩效人群的核心瓶颈——生理精力与自主神经降挡机制切入，详解了如何通过长呼气调节心率、切断本体感觉以及新型非侵入式设备实现秒速入眠与专注切换。随后，Huberman 剖析了神经科学的范式转移：从过去数十年粗放的脑活动读取，跨越到未来利用聚焦超声、无创光电刺激及 AI 闭环自适应直接向大脑“写入”特定生理状态。他还前瞻性地指出 Meta、OpenAI 及 Neuralink 等科技巨头终将成为生物科技公司，并探讨了破解清醒脑状态、胚胎基因编辑光谱以及遵从内在真实好奇心对突破学科边界的核心意义。 ## Atmosphere & style 整场对话兼具硬核生物学机制的精深推导与硅谷前沿科技的宏大视野，语调冷静、透彻且极富穿透力。Huberman 既有严谨学者的格物致知，又不乏打破学院派条框的叛逆与坦率，将复杂的脑电传导、神经受体与科技领袖的真实案例信手拈来。听感上犹如一场高密度的智力冲浪，在理性解构人体生理极限的同时，悄然注入对人类认知前沿与生命本质的敬畏与激情。 ## Bulletin - `22:07` 顶尖AI公司（如 Meta、OpenAI、Anthropic）以及马斯克的终局是演进为生物科技公司，AI 军备竞赛的深层目标正在转向通过算法闭环对人类自身神经活动进行高精度的动态读写调控。 - `22:07` 非侵入性技术（特别是聚焦超声）被视为神经科技的圣杯，有望替代传统开颅手术，在毫秒级与细胞级精度上实现对深部脑区的定向写入与状态调控。 - `35:00` 海马体学习实验颠覆了传统“神经元放电必须依序进行”的假说，反向甚至无序全量激活神经元仍可精准触发对应行为，这极大简化了状态写入的工程难度，但也放大了回路复用带来的连锁风险。 - `01:03:01` 相比睡眠脑机制的相对清晰，解码与重构“清醒大脑状态”（如极致专注、警觉与沉浸）成为当前神经科学界与科技巨头争夺的下一个核心未解之谜。 ## Key takeaways - `00:00` 顶尖高绩效人群的终极瓶颈已经从外部团队的执行速度转移到了个人生理层面的精力带宽与恢复效率。 - `06:20` 成功入眠需要满足三大神经生理前提：切断主动思维与未来规划、通过长呼气降低静息心率、通过眼球运动切断本体感觉对肢体方位的追踪。 - `10:00` 长呼气通过“呼吸性窦性心律不齐”机制促使心脏容积变小、血流加快，进而触发迷走神经向心脏释放抑制信号以迅速拉低心率。 - `12:35` 麻省理工学院研发的双向眼罩等新一代神经科技设备，不仅能直接量化 REM 睡眠，还可通过微电流引导眼外肌运动主动诱导入睡与调节白昼神经状态。 - `17:57` 当前公共健康与长寿传播陷入了极低门槛与过度优化的闹剧，行业在经历公众疲劳的下行泡沫后，将回归由严谨 MD/PhD 学者主导的硬核科普。 - `26:18` 现有的药物与精神活性分子写入手段属于全局性的粗放调节，缺乏空间与时间特异性，而聚焦超声与皮下植入设备正成为精准写入的新范式。 - `30:00` 大脑对外周感觉和深层核团（如控制食欲的下丘脑靶点）的调控机制已十分清晰，但在记忆、思维本质等高阶认知层面仍处于理论迷雾中。 - `39:00` 大脑具备惊人的自组织意图解码能力，即使运动通路受损，直接捕获言语规划区的粗糙意图信号也能精准驱动语音合成，绕开受损的外周运动神经。 - `40:00` 下一代无创神经监测工具能直接采集自主神经电活动与瞳孔变化，帮助个体准确区分“良性压力”与“病理性痛苦”。 - `42:07` 无创神经写入未来可能借助 AAV 病毒载体植入工程化受体，配合外戴帽子发射声光刺激，并需设置口服小分子药物等“与门”安全机制防范滥用。 - `52:43` 贺建奎 CRISPR 婴儿事件暴露了科学界内部微妙的观望心态，同时引发了学界关于基因编辑敲除 CCR5 是否旨在进行认知增强的长期隐秘猜想。 - `55:00` 从常规羊水穿刺到商业胚胎筛查乃至基因治疗，人类对基因干预的接受度呈现连续光谱，技术的终局在于对生理与认知状态的深度掌控。 - `01:00:00` 科技领袖（如 Altman、Musk、Zuckerberg）正在成为神经科学前沿的真正推动者，跨学科视角的注入打破了传统学术界狭窄保守的研究边界。 - `01:06:12` 真正的学术与人生突破取决于将自身纯粹的好奇心与所从事的工作紧密结合，违背内心真实渴望去迎合世俗成功无异于长期的精神慢性死亡。 ## Detailed walkthrough ### `00:00` 精英表现者的精力瓶颈与“埃隆·马斯克效应” 在现代商业和科技的前沿，最顶尖人才面临着一个关键逆转：以往往往是世界其他人的速度跟不上顶尖人才，导致这些天才因周围人效率低下而倍感挫败；然而如今，由于技术工具极大地放大了个人潜能，真正的瓶颈已变成了这些顶级人才自身的“精力带宽”（Energy Bandwidth）。许多杰出创业者受到埃隆·马斯克（Elon Musk）同时掌管多家巨头公司的影响，产生了一种“真正的顶级人物必须同时驾驭多家企业”的认知，但也正因如此，如何生成能量、维持能量、避免倦怠、选择“清洁燃料”而非“劣质燃料”以保持长期高水平产出，成了决定人类进步速度的核心命题。 ```mermaid flowchart TD accTitle: 精力瓶颈演进与精力管理框架 accDescr: 展示顶尖人才从外部带宽限制转向内在生理能量瓶颈的过程，以及应对该挑战的评估与分期策略。 subgraph Problem["瓶颈转换"] A["传统限制：外部团队执行力滞后"] --> B["现代瓶颈：个人生理精力与认知带宽"] end subgraph Assessment["自我认知（Know Thyself）"] B --> C["摒弃盲目模仿：识别马斯克式的极端基因特例"] C --> D["诚实评估生理睡眠基线（如5/6/7小时的临界点）"] end subgraph Lifecycle["职业周期的精力策略"] D --> E["早期冲刺期（0-4年）<br/>核心目标：活下去、产出最大化、防伤病"] E --> F["成熟优化期<br/>核心目标：日/周/月尺度插入结构性休息"] end ``` 谈及马斯克这一现象级案例时，Andrew Huberman 指出，马斯克代表了多任务并行能力的绝对巅峰，在此维度上几乎无人能及。就连湾区一位极具影响力的百亿级投资人（曾参与早期 PayPal 团队）也曾私下向 Huberman 询问：“从生理学机制上讲，埃隆究竟是如何做到这一点的？”Huberman 直言在纯生理层面上无法解释。历史上确实存在极少数拥有天然基因突变、几乎不需要太多睡眠且精力无限的特例，但绝大多数人绝不能自欺欺人地认为自己也能靠每天四五个小时的睡眠维持顶级运转。 对绝大多数高绩效人群而言，最底层的逻辑始终是“认识你自己”（Know Thyself）。每个人必须对自己的真实睡眠需求做一次诚实的自我评估： - 以 Huberman 自身为例：睡满 7 小时感觉极佳；睡到 8 小时早晨反而会有些昏沉（groggy）；睡 6 小时能够正常应对；若遭遇 1 天睡眠仅 5 小时甚至彻夜未眠，仍能咬牙撑住；但如果连续两晚睡眠低于 4 小时甚至为零，生理状态就会彻底崩塌（cratered）。 - 提升能量的健康干预手段如今已广为人知：运动、充足的睡眠、合理利用咖啡因等。至于尼古丁（Nicotine），尽管它会上瘾并升高血压，但确实能让人进入“专注但放松”的状态。Huberman 强调，若想真正利用尼古丁的认知增强和专注提升效应，绝不能成为长期慢性使用者，而只能偶尔进行低剂量冲击，避免受体脱敏。 --- ### `05:00` 能量储备的差异、职业阶段与“超速运转”的代价 除了基础睡眠，人与人之间存在着显著的天然能量差异。有些人天生拥有极高的能量底色：他们能跑完长距离后立刻投入高强度工作，或者凌晨 4 点起床、锻炼、处理商业事务、陪伴家人，并在复杂的任务切换中毫无阻滞。然而，人们往往忽略了这类高能量人群的“内在生理状态”（Internal State）： - 许多表现得不知疲倦的人，实际上一直处于“超速运转”（redlining）状态；他们要么重度依赖兴奋剂，要么单纯依靠交感神经极度亢奋硬撑。这类人最致命的问题在于夜晚无法自主切断神经兴奋，导致系统无法重置。 - 在年轻时或特定高压阶段（例如攻读博士学位或初创企业的前四年、医学院实习期），80 到 100 小时的工作周是常见且可行的。Huberman 回忆自己读博时也是如此拼命，并最终获得了斯坦福大学的终身教职。在这个阶段，所谓的“优化”（optimization）往往是伪命题，真正的第一法则是“别死掉、别重伤或患上慢性病，在身体承受范围内榨取最大产出”。 - 然而，当度过最初阶段后，就必须转向长期的可持续性：学会在每日、每周、每月插入系统性的休息单元，以此释放最高维度的专注与创造力。对于在硅谷从事 AI 等尖端领域的创业者，强行要求他们做呼吸练习或保证 8 小时睡眠是不现实的，关键在于看清他们是否拥有像音乐教父吉米·爱奥文（Jimmy Iovine）或 Meta 顶尖 AI 研究员那样天生的“额外档位”（Extra Gear）。 --- ### `06:20` 神经降挡：入眠的三大生理前提与知觉切换机制 高能量人群最常见、最痛苦的瓶颈并非缺乏动力，而是无法“降挡”（downshifting）——大脑在经历了一整天高强度、令人兴奋的工作后，思绪在睡前疯狂奔逸，往往需要耗费三到四个小时才能让大脑平静下来。 要实现神经系统的成功下调并进入睡眠，生理学上有三个不可或缺的核心前提： 1. **切断主动思维与预测规划**； 2. **主动降低心率**； 3. **遗忘肢体在空间中的具体本体感受位置**。 针对第一点“关闭思维”，Huberman 指出这属于一项可以被训练的神经学技能。大脑的核心功能之一就是对未来进行预测（Prediction），因此大脑会自然而然地不断思考、预测、回忆和梳理接下来的事务。要打断这种前额叶驱动的循环，唯一的解法是进入“纯粹的知觉感受”（Pure Sensation）： - 人类大脑具备从思维/预测模式切换至纯感受模式的能力。操作方法是将意识集中在身体表面的触觉、房间内的背景声响、以及内在知觉（Interoception，例如不借助外力感受自身心跳的能力）上。 - 与心率减慢（前提二）相结合的关键技术是“长呼气”（Long Exhales），通过延长呼气时间触发副交感神经，从而在生理层面拉低心率。 - **与传统冥想的关键区别**：传统冥想通常要求将注意力高度聚焦于单一锚点（如呼吸、前额后方的某一点或双手）；而入眠所需的感知切换练习恰恰相反，它要求注意力在身体各个部位之间“流动迁移”——从脚部、腿部，移动到呼吸节律，再到外部环境的声音，随后再循环流动。这种将意识锚定在当下纯粹感知的过程，能够有效关闭负责计划的大脑神经网络，构成跨越入睡门槛的关键第一步。 ### `10:00` 呼吸机制与心率：通过长呼气关闭思绪 要让大脑真正停止胡思乱想，心率必须下降到特定阈值以下。这种心率的下降主要是通过长呼气触发的，其背后的生理机制被称为**呼吸性窦性心律不齐**（respiratory sinus arrhythmia）。虽然“心律不齐”听上去像是一种危险的病症，但在生理学上它其实是非常健康且理想的反应。 ```mermaid flowchart LR accTitle: 呼吸对心率调节的生理机制 accDescr: 展示吸气与呼气时膈肌、胸腔空间、心脏容积变化及迷走神经如何反向调节心率。 subgraph Inhale["吸气阶段 (Inhale)"] direction TB I1["膈肌下移 / 肺部充气"] --> I2["胸腔空间增大 / 心脏容积变大"] I2 --> I3["血流速度相对放缓"] I3 --> I4["脑部发出信号：心率加快"] end subgraph Exhale["呼气阶段 (Exhale)"] direction TB E1["膈肌上移 / 肺部排气"] --> E2["胸腔空间缩减 / 心脏容积变小"] E2 --> E3["单位体积血流加快"] E3 --> E4["迷走神经向心脏传递信号：心率减慢"] end ``` 人在呼吸时，随着膈肌的运动和肺部的充盈，心脏在胸腔内实际拥有的物理空间会发生改变。吸气时，心脏容积变大，血液流速相对变缓，大脑便会发出信号让心率略微加快；而当呼气时，心脏容积受挤压变小，单位容积内的血液流动反而变快，此时脑部会通过**迷走神经**（vagus nerve）向心脏发送抑制信号，从而迅速减慢心率。 因此，无论是在清醒状态下想要快速减压、降低神经系统的激活水平，还是在睡前想要顺利入眠，最简单有效的方法就是做几次深度长呼气，直到肺部气体彻底排空，随后仅做轻微的被动吸气。通常在高度紧张或兴奋时，人们倾向于大量主动吸气而仅做被动呼气；将这种呼吸模式反转为“呼气为主”，是自主神经调节的底层逻辑。 --- ### `11:14` 摇晃、眼球代偿运动与本体感觉切断 促进入睡的另一个关键维度是关闭身体对四肢方位的感知，即切断**本体感觉**（proprioception）。有大量研究探讨过在不借助药物的情况下，是什么促使婴儿以及成年人入睡。父母凭本能学会的以特定节奏摇晃婴儿的做法，在成人身上同样有效——实验室中制造的特定频率摇篮床能让成年人极快入睡。 这种机制的核心在于前庭系统与视觉系统的联动： - 当身体左右摇晃时，即使在闭眼状态下，眼球也会在眼皮底下产生代偿性眼球运动。 - 这些眼球运动通过脑干向小脑传递信号，以此控制身体平衡和本体感觉。 - 摇动带来的神经信号输入，会阻断大脑对四肢所处空间方位的精确追踪，促进入睡。 由于成年人很难拥有一张摇篮床，更不可能有人抱着摇晃入睡，因此可以通过主动模拟这种闭眼下的眼球运动来达到同样效果。对于神经科学专业人士而言，这种做法的本质是扰乱了视网膜方向选择性细胞（direction selective cells）、脑干的副视核（accessory optic nuclei）与小脑之间的神经回路，使大脑暂时“遗忘”身体的姿势与位置。在操作上，只需在闭上双眼后，缓慢或快速地将眼球从一侧移向另一侧、向上再向下，或者顺时针、逆时针转动眼球，配合一次长呼气，就能快速打破本体感觉回路。 --- ### `12:35` 神经科技前沿：MIT眼罩与脑状态精准写入 眼睛作为直接延伸出颅骨腔外的脑组织部分，是通往整个大脑的绝佳物理通路。来自麻省理工学院（MIT）的一个团队研发了一款具备双向交互能力的眼罩设备，不仅能够精确监测睡眠，更开辟了直接向大脑“写入”特定状态的可能。 当前的常见可穿戴设备（如 Whoop、Oura 智能戒指、Eight Sleep 智能床垫）虽然能估算睡眠分期，但无法直接测量快速眼动（REM）睡眠。该眼罩通过直接捕捉眼球运动来精准量化 REM 阶段；更为关键的是，它利用耳后贴片向控制眼外肌（extraocular muscles）的神经回路施加微弱电刺激，引导眼球在闭眼状态下产生缓慢的往复运动，从而主动诱导入眠。一旦进入睡眠，设备还能进一步增强 REM 睡眠阶段的深度与时长。该技术预计在7到12个月内推向市场。 不仅是睡眠场景，这种“神经读写”技术的应用正在向清醒状态拓展： - 在不远的将来（或许距离睡眠眼罩问世仅需约12个月），可能会出现具备类似功能的智能眼镜。 - 这类眼镜能精准刺激自主神经系统的特定分支，在工作时将大脑推入极致专注的高效模式，在下班回家途中则平稳切换至放松蓄能模式。 - 当前人们依靠冷水澡、咖啡因、锻炼或特定呼吸法（以吸气为主来提神，或以呼气为主来镇静）来人为调控神经状态，未来这类非侵入式刺激设备将实现对学习、工作与放松状态的程序化诱导。 --- ### `15:28` 自主神经系统的全面调控与能量管理的长期分水岭 想要实现高效的神经关机与深度恢复，需要将三大核心要素结合起来（其先后顺序并不严苛）： 1. **停止思维奔涌**（关闭心智层面的思绪） 2. **切断本体感觉**（通过眼球运动遗忘身体方位） 3. **拉低静息心率**（通过长呼气激活迷走神经） 对于面临巨大生活压力的人群——无论是创业者、背负育儿与高强度工作的职场人，还是处于心理创伤与悲痛复原期的人——如果每晚只有4小时睡眠时间，却需要6小时的生理恢复量，掌握一沾枕头就瞬间关机入睡的能力将是巨大的生存优势。 高负荷人群的生理状态会随着时间推移出现明显分化。在年轻或处于职业起步期时（例如读研究生或创立早期公司期间），许多人仅仅是“尽一切可能不被累垮”；随后他们会逐渐转向平衡与优化；而最终在各自领域达到顶尖水平的人，其生理系统与日程强度的匹配度会直观地刻在他们的外貌与健康上： - 部分人虽然事业有成，但外表显得极其憔悴、患有慢性病，生理年龄看起来比实际年龄老上10岁。 - 另一些人则表现出非凡的“额外档位”（extra gear），例如埃隆·马斯克（Elon Musk）那般展现出不可思议的连续高产精力。 - 像推特联合创始人杰克·多西（Jack Dorsey）这类人，始终维持着极度平静与深思熟虑的状态，其核心秘密在于摸索出了一套生理机制：白天全天保持警觉而平稳，夜晚则享有高质量睡眠，以此支撑起超长周期的职业生涯。 --- ### `17:57` 公共健康舆论的乱象与极低门槛的闹剧 当谈及整个神经科学与健康传播领域面临的最大挑战时，公共领域中围绕健康与长寿（health and longevity）的舆论环境被指责为一场充斥着哗众取宠的“灾难”。 在传统的神经科学学术界，即便拥有博士学位（PhD），对于“是否真正为人类理解大脑做出了实质贡献”而言也仅仅是一个必要条件，远非充分条件。然而在当前的社交媒体语境下，健康传播的准入门槛几乎为零： - 许多人只需展示清晰的腹肌，承认自己使用了类固醇，就能轻而易举收获海量拥趸。 - 还有一些人展示腹肌或臀部，宣称自己“纯自然”（all natural），便能迅速建立起巨大的信任与追随度。 - 越来越多人没有任何严谨的科学或医疗背景，仅仅凭借“我过去很胖，现在瘦了，我尝试了这些手段”的个人叙事，就自封为身心优化大师。 早期如蒂姆·费里斯（Tim Ferriss）开创的身体黑客（biohacking）与自我优化探索是极为坦诚的，他坦率分享过哪些实验值得做、哪些尝试后来证明并不明智，并且始终保持真实的个人探索边界。但随后整个领域却演变成了越来越极端的表演与闹剧，严肃严谨的科学解释在浮夸的市场营销与博人眼球的表演面前遭受严重挤压。 ### `20:00` 公共健康传播的困境与专业回归 健康科普领域正经历两极分化。顶级科普工作者与过往二线创作者之间的鸿沟日益拉大，大众传播者的声量越来越膨胀，但浮夸的表演风格正在损害公共健康讨论的严肃性。Andrew Huberman 坦言："获取关注极其容易，但维持尊重极其困难（It's very easy to get attention. It's very hard to keep respect）。"无论是 Brian Johnson、Gary Brecka 还是 Dave Asprey，受众在审视这一群体时，需要辨别究竟是谁的人格特质在真正引发共鸣，而谁仅仅是在提供娱乐价值。 整个行业已经从最初的求知热潮演变成极端追求各项指标的“过度优化”。公众正对层出不穷的繁琐养生指令感到厌倦和疲惫，甚至觉得整个生态逐渐走向荒诞和疯狂。Huberman 预判健康优化与延寿科普将迎来一轮行业下行退潮；但在泡沫破裂后，会有一批真正拥有科学、医学或药理学博士（PhD）与医学博士（MD）严谨学术背景的新同侪涌现出来。他们不会退化为迎合短视频算法的扭动热舞或自我吹捧模式，而是带来真正高水平的严肃科普。 ```mermaid flowchart LR accTitle: 健康科普演进路径 accDescr: 展示公众健康传播从过度优化走向理性专业化的三个阶段 subgraph Phase1["当前阶段"] A["狂热追求优化<br/>表演化/博眼球"] end subgraph Phase2["阵痛调整"] B["受众疲劳与反弹<br/>行业热度下行退潮"] end subgraph Phase3["未来常态"] C["专业 MD/PhD 进驻<br/>严肃硬核科学传播"] end A --> B B --> C ``` --- ### `22:07` 神经科学的圣杯：AI 巨头的生物科技终局 在神经科学研究层面，最重大、最前沿的挑战是如何以无创（non-invasive）方式在睡眠与清醒状态下直接对神经系统进行“写入（write）”。脑外科手术在治疗脑部肿瘤或严重癫痫病灶时必须打开颅骨——例如 Huberman 的儿时好友、加利福尼亚大学旧金山分校（UCSF）神经外科主任兼生物工程师 Eddie Chang，在为保持清醒的患者切除脑瘤的同时，会进行高精度的语言与言语脑区定位刺激。虽然神经外科医生常戏称钛合金修补板比原装骨头更坚固耐用，但正常人绝不会无缘无故接受开颅手术。 未来的核心分水岭在于非侵入性技术（极大概率是聚焦超声），它能在极高时间与空间精度下精准激活或抑制特定脑区。诸如帮助闭锁综合征患者开口发声、让脊髓损伤患者重新站立行走等临床突破属于英雄式的壮举；但更深层的产业变革在于底层科技公司的进化方向。Huberman 明确断言：诸如 Meta、OpenAI、Anthropic 等前沿人工智能企业，以及布局 Neuralink 的马斯克，其最终归宿都是**生物科技公司（Biotech companies）**。 这些顶级 AI 团队的幕后人员即便没有科班神经科学学位，也对大脑运作机制有着深刻理解。AI 军备竞赛的真正终局并非仅仅是构建大语言模型（LLM），而是实现对人类自身神经活动的高精度动态调制。 * **分级调控而非简单开关**：神经系统的信号传递高度依赖门控（gated）机制与分级电位（graded potentials），需要像调节旋钮一样平滑改变兴奋阈值。 * **方波式能级输出（Square Wave Function）**：通过无创设备精准刺激伏隔核（Nucleus Accumbens）及关联动机通路，人在进行深度学习或硬核攻坚时能瞬间拉满专注与动力；而在任务结束时立即切断刺激以保护精力，使全天的能量输出如同规整的方波。 * **AI 闭环自适应滴定**：AI 的真正价值在于实时监测并读懂个体的动态脑电特征，识别出某种过度兴奋状态可能导致后半天精力枯竭，进而自动微调（titrate）神经干预强度。这与 Eight Sleep 智能床垫在夜间通过调节水温被动干预 REM 睡眠和深睡类似，但直接作用于中枢神经的调控将比被动的体温传导强大得多。 --- ### `26:18` 神经读取与写入的技术路径：从开颅电极到皮下干预 回顾过去三四十年，神经科学的核心工作主要停留在“读取（read）”大脑在感知、思考和运动时的天然信号；而当下的转折点在于如何高保真度地向其“写入（write）”。神经元之间通过动作电位（action potentials）进行通信，教科书上将其简化为整齐划一的“全或无（all-or-none）”电信号，但现实中动作电位存在诸多细微形变与阈前分级电位变化。当前技术在空间与时间精度、侵入深度之间存在着权衡博弈： | 读取 / 写入技术 | 侵入程度 | 空间与时间精度 | 局限性与代表场景 | |:--- |:--- |:--- |:--- | | **胞内记录（Intracellular）** | 极高（穿透颅骨，电极刺入神经元） | 单细胞级别极高精度 | 必须穿刺细胞膜，破坏性大，仅限实验研究 | | **胞外记录 / 钉床阵列（Bed of Nails）** | 高（穿透颅骨，放置在细胞周围） | 细胞集群级别高精度 | 需开颅植入电极线（Neuralink 类技术路径） | | **fMRI / 脑电图（EEG）** | 完全无创 | 空间分辨率或深层穿透力极差 | 仅能捕捉皮层浅表热图，无法有效触及深层脑核团 | | **经颅磁刺激（TMS）** | 完全无创（穿透颅骨刺激） | 空间聚焦精度较低 | 临床用于抗抑郁，难以进行极精准靶向写入 | | **聚焦超声（Ultrasound）** | 完全无创 | 高时空精度潜力巨大 | 有望成为跨颅骨无创读写深部脑区的未来核心途径 | 在药理学层面，所有的化学“写入”手段本质上都是粗放的间接调节： * **下调阈值（提高转速 RPM）**：阿德拉（Adderall）、咖啡因、尼古丁、苯丙胺（Speed）和莫达非尼（Modafinil）通过增加神经调质释放来降低动作电位触发阈值，使全脑和躯体处于高转速紧绷状态，容易引发肢体震颤、坐立难安及思维停不下来的亢奋感。 * **抬高阈值（抑制放电）**：酒精、巴比妥类和镇静药物通过增强抑制性神经递质释放，全面抬高神经元激活门槛，间接强迫大脑安静下来。 要打破药物的全局副作用并实现精准写入，核心在于使用者愿意接受何种程度的“皮下侵入”。干预手段并不一定要凿开坚硬的颅骨，皮下植入同样具备非凡威力。例如迷走神经刺激术（VNS）：迷走神经通常被大众误认为是纯粹的放松神经，但其大量传导通路实为兴奋性通道。目前临床上常在患者颈部皮下植入一枚比一分硬币略小的刺激装置，通过电刺激逆转抑郁症伴随的嗜睡与行动迟缓。 斯坦福大学生物工程学家 Karl Deisseroth 曾展示过令人惊叹的实时调控案例：面对一名有严重自杀倾向的重度抑郁症患者，当他在交谈过程中逐步调高患者的迷走神经刺激强度时，患者的心理状态在现实对话中发生了立竿见影的逆转，直接从极度绝望切换为主动构思要出门去应聘找工作。这种客观硬件对主观世界感知的实时重塑令人震撼。 对于皮下植入技术，Huberman 直言自己持完全开放接受的态度。曾在他实验室工作过、现任 Neuralink 首席神经外科医生的 Matt McDougall 便是一位激进的实践者：他在自己以及妻子的虎口（拇指与食指间的皮下）各植入了一枚微型射频接收芯片，出门回家无需实体钥匙，只需挥手即可解锁门禁。这种典型的硅谷湾区生物黑客（Biohacking）实践，展现了人机直接交互的雏形。 ### `30:00` 植入式神经技术与大脑模型的现状 谈及神经技术的未来落地，Andrew Huberman 预测植入式设备很快会迎来爆发。很多人目前还在狂热讨论多肽药物（Peptides），但在他看来，植入人体的微型设备才是真正的下一个前沿。在大约 12 个月内，可能就有 20% 的创业圈人士会觉得：“只要侵入性不比打耳洞严重，那我何不装一个迷走神经刺激器？”尽管目前的商业化迷走神经刺激器效果尚不能让执着于“切开头皮、钻开头骨”的神经外科医生满意，但这类技术的演进速度不可小觑——正如十年前如果有人说每个人都会戴着手环或戒指（如 Oura Ring 或 Whoop）来精确监测睡眠和心率变异性（HRV），当时的人也会觉得匪夷所思。 面对神经技术的读出（Reading）、写入（Writing，如超声、电刺激、磁刺激）以及“大脑模型（Model）”这三大支柱，主持人发问：我们对大脑本身的理论模型究竟发展到了哪一步？Huberman 解释，这取决于你考察的是哪个层级： - **外周感觉与神经肌肉控制**：视网膜、耳蜗、舌头、嗅觉系统及皮肤，科学界对这些外周器官如何运作理解得非常透彻。例如，神经系统如何把不同波长的光转化为视觉图像，如何解析运动，人类都已掌握了高精度的细节。 - **深层大脑结构（下丘脑与边缘系统）**：控制饥饿、愤怒、性冲动、焦虑以及激素分泌的深层回路，本质上就像一组明确的“开关”。GLP-1 类受体激动剂之所以起效如此迅速且广泛，正是因为我们清楚知道它们在脑内的结合靶点——直接激活下丘脑中抑制食欲的神经元。但这也是把双刃剑：一旦剂量越过阈值，人对其他事物的驱动力也会一并被抑制（例如 Sam Altman 曾公开谈过他因丧失动力而减少用量或停药的体验）。 - **前额叶皮层（PFC）**：通过人类脑损伤病理研究和脑区激活实验，科学界清楚前额叶负责情境学习、制定环境策略以及抑制冲动（即“自上而下的抑制” Top-down suppression）。 然而，一旦深入到记忆和“思想（Thoughts）”的本质，神经科学便步入了一片迷雾。 ```mermaid flowchart TD accTitle: 大脑不同层级理解度对比 accDescr: 展示从外周感觉到高级认知的科学理解深度阶梯 subgraph High["高度清晰 (开关明确)"] A["外周感觉系统<br/>视网膜 / 耳蜗 / 皮肤"] B["深层脑结构<br/>下丘脑: 饥饿 / 愤怒 / 激素 (GLP-1 靶点)"] C["前额叶皮层 (PFC)<br/>策略制定 / 冲动抑制"] end subgraph Murky["充满谜团 (理论危机)"] D["记忆回路 (海马体)<br/>时序假设被推翻"] E["思想生成<br/>无教科书机制解释"] end High --> Murky ``` --- ### `35:00` 海马体记忆实验与被颠覆的“时序”假说 在涉及记忆的机制时，神经科学界遭遇了一个让人不安的“暮光之城”式危机。教科书历来坚信：大脑回路必须按照严格的时间顺序依序点火——就像发网球一样，必须由神经元 A 传导到 B，再传导到 C，形成一串精准的电信号链条。 Scripps 研究所的 Mark Mayford 实验室，以及因免疫球蛋白研究获得诺贝尔奖、后转型为顶尖神经科学家的 Susumu Tonegawa（利根川进）实验室，在小鼠和灵长类身上进行了一系列海马体经典学习实验。他们让动物学习特定任务（例如在特定区域受到电击或获得奖励、走迷宫、按压特定拉杆组合），并用化学标签“标记”下任务期间被激活的神经元。 随后进行的行为复现与对照实验得出了令整个领域震惊的结果： 1. **依序激活**：按照动物学习时的原有顺序重新激活这组标记神经元，动物如期展现出了学习到的行为。 2. **反序激活**：实验人员将激活顺序完全倒过来，从尾到头刺激这组神经元，结果动物**依然准确完成了该行为**。 3. **全量齐发（砸琴键）**：实验人员不再保留任何时序，就像双拳同时砸在钢琴键盘上一样，瞬间同时激活该模式中的所有神经元。令人难以置信的是，动物表现出的行为与原本完全一致。 - 记忆与行为的表达并不严格依赖神经元放电的时间先后顺序（Temporal ordering），这彻底颠覆了长期以来的认知。 - 对于神经技术的“写入”而言，这既是捷径也是噩梦：捷径在于或许无需极高的时间精度，只要把相关神经元一股脑点亮，大脑自己就能补全下游逻辑；噩梦在于，这些被标记的神经元同时兼任了其他无数的感知、记忆与思考功能。 - 大脑的高度复用性（Repurposing）意味着：并不存在一个只负责“打网球”而与你的“童年回忆”毫无关系的纯粹回路；刺激其中一部分，极有可能引发无法预测的连锁干扰。 --- ### `39:00` 意图、语音合成与跨越“半机械人”界线 虽然时序谜团令人困惑，且科学界至今无法在教科书里写明“一个想法是如何诞生的”，但大脑展现出了惊人的自组织能力：只要系统拥有一个明确的目标（例如“喝面前这杯水”），即便输入的刺激相对粗糙，大脑往往也能自行摸索并完成控制。 一个极具代表性的临床案例来自 Huberman 的朋友 Eddie Chang（张复伦）医生。他在治疗闭锁综合征（Locked-in syndrome）患者时，成功实现了让患者通过电脑“说话”： - 语言的生理本质仅仅是咽喉对呼出气流的精细肌肉控制。 - 面对无法将运动信号传导至咽喉的瘫痪患者，研究人员并没有去测量发音执行区域，而是花了数十年时间深入解码**言语规划区域（Speech planning areas）**输出的高阶意图信号。 - 算法捕捉到这些规划意图后，直接将其解码并转换为电信号驱动计算机发声，绕过了受损的外周运动神经。 这种非特异性既带来了工程落地的巨大优势，也引出了更深层次的哲学恐惧。主持人感叹，如果未来通过高精度的颅内技术就能像“弹钢琴”一样随意拨弄人类的神经回路，这在某种程度上剥离了人类的自主性与身份认同。谈及未来 5 到 10 年内人类是否会不可避免地跨越“半机械人（Cyborg）”的技术界线，Huberman 笑着回应，自己作为一个在帕洛阿尔托（Palo Alto）土生土长的人，对这种科技交织的演进有着非常切身的体会。 ### `40:00` 自主神经系统的直接读取与日常压力监测 公众对新兴技术的接受过程往往遵循类似的心理轨迹。正如人们最初对个人电脑充满恐慌、怀疑手机时刻在窃听谈话，或者初次在旧金山搭乘 Waymo 无人驾驶出租车时心存顾虑但很快习以为常，对神经系统的直接读取也正跨越这一心理门槛。目前主流的可穿戴设备主要依赖心率与心率变异性（HRV）来间接推测压力、睡眠和恢复状态，但这并非来自自主神经系统的直接读取。 前斯坦福大学博士后 Melissa Almest（曾就读于哈佛大学并在加州理工学院有短期研究）正在开发一种革命性的腕带袖套工具，能够直接测量自主神经系统的活动。这一工具不仅能记录不同言语与呼吸模式下的神经反馈，更打破了传统的健康认知：过去的观念认为心率上升、HRV 下降即代表负面状态，但她的研究发现，当人们在全身心投入热爱的事物而感到压力时，这种“良性压力”（eustress）反而能促进后续的高质量睡眠与高度专注。目前该技术已在警察部门和急救人员中试点，未来将推向大众。 除了腕带，未来的神经读取还将借助智能眼镜和眼罩等日常形态展开： - **瞳孔大小变化**：在动物实验中已被证实是测量专注度和警觉水平的核心指标，未来完全能够集成在普通眼镜或睡眠眼罩中进行无感采集。 - **压力尾部效应（Stress Tail）**：直观反馈不仅揭示日常生活中隐蔽的压力源，还能精确量化一次压力事件后自主神经系统的恢复时长。 - **高唤醒与负面压力的辨析**：帮助个体分清自身究竟是处于病理性痛苦（distress），还是处于高认知唤醒的积极推进状态。 ```mermaid flowchart LR accTitle: 自主神经监测与多模态读取路径 accDescr: 展示从间接指标向直接神经监测演进，以及未来外戴设备的融合方案。 subgraph SG1["传统间接推断"] A["心率 / HRV 数据"] --> B["推测整体压力与睡眠"] end subgraph SG2["下一代直接神经读取"] C["腕部袖套传感器"] -->|"自主神经电活动"| D["辨析良性压力 vs 痛苦"] E["智能眼镜 / 眼罩"] -->|"瞳孔动态监测"| F["量化即时专注与警觉"] end ``` --- ### `42:07` 无创神经写入：病毒载体、光遗传学与安全门禁 相较于读取，更具颠覆性的是对神经系统的写入与调控。对于健康人群而言，开颅植入电极是不现实的，理想状态是在不破坏颅骨的前提下实现全脑或特定脑区的精准调控。传统电刺激过于粗糙，而现代神经生物学普遍采用基因工程手段：通过注射无害的腺相关病毒（AAV）作为载体，将来自藻类等微生物的离子通道基因递送至大脑特定类型的神经元中。 在未来，这种机制可以演变为一种完全无创的大脑接口。使用者只需接受一次安全的病毒载体注射，让特定神经元表达出工程化受体。此后，只需佩戴一顶普通的棒球帽或羊毛帽，内置的发射源便能穿透颅骨发射长波长光或超声波： - **非侵入式光/声刺激**：帽子发出的波束不需要极度精细聚焦，只有预先被基因标记的特定神经元才会对该刺激产生响应。 - **连续动态调节**：借助 AI 算法，系统能够根据需求控制光/声的强度与闪烁频率，使神经元在静息、活跃到超活跃之间平滑过渡，而非简单的开/关二元状态。 - **安全门禁机制（AND Gate）**：为了防止情绪失控或恶意攻击，这类神经调控必须内置防护机制。例如设计“与门”（AND gate）逻辑——不仅需要超声或光刺激，还必须配合口服特定的化学小分子药丸才能激活通路，同时使用者也可随时断开物理设备。 针对这种调控可能引发的成瘾或滥用担忧，咖啡因与甲基苯丙胺的对比提供了现实参考。全球 90% 以上的成年人每天都在摄入咖啡因，通过阻断腺苷受体间接增强脑内兴奋性；未来对神经回路的直接调节本质上是同一逻辑的高精度版本，人们将能根据睡眠与精力状况精确设定当天的专注度。 --- ### `45:00` 医疗数据的平民化争议与精准调控演进 公众获取自身身体数据的浪潮往往伴随着医疗既得体系的阵痛。例如，当全身核磁共振成像（Whole-Body MRI）的成本大幅下降、民众得以自主筛查早期肿瘤时，神经外科等传统医疗圈层却表现出强烈的抵触情绪，《华尔街日报》甚至刊文对其泼冷水。这与上世纪 80 年代全基因组测序刚刚兴起时的社会恐慌如出一辙——当时人们担忧查出亨廷顿舞蹈症基因会带来灾难，而如今基因测序与常规验血已成为个人健康管理的日常工具。随着检测成本骤降，全身影像与直接神经读取在未来一年半到两年内走向普及已是大势所趋。 回顾神经生物学的演进历程，人类对大脑的理解正从极度粗放的破坏性研究转向超高精度的分子级调控： - **粗放解剖时代**：通过著名铁道工 Phineas Gage 的脑部贯通伤推断眶额皮层与决策策略的关系，或通过病人 H.M. 的海马切除手术理解记忆编码机制。 - **现代光化遗传学时代**：利用化学遗传学标签、光与超声波精确微调（tickle）某一段特定记忆或行为所涉及的具体神经元回路。 - **环境信号与精密回路的契合**：晨间接受阳光照射并非粗暴的“生物黑客（biohacking）”手段，而是因为视网膜到下丘脑之间本就演化出了专用于校准昼夜节律和皮质醇峰值的硬编码回路。 - **药理学与神经可塑性**：选择性 5-羟色胺再摄取抑制剂（SSRI）最初有效缓解了临床级强迫症（OCD），但在抑郁症领域被过度滥用；而当前备受关注的裸头草碱（Psilocybin）同样通过强力提升 5-羟色胺水平，在解离大脑横向连接的同时诱发强烈的神经可塑性，促成全新神经联想的建立。 ### `50:00` 神经调控工具的钝感局限与从“读取”到“感知”的解码之路 Andrew Huberman 深入剖析了当下精神医学与神经调控手段的底层逻辑。无论是赛洛西宾（psilocybin，裸盖菇素）还是 MDMA，它们之所以能在创伤、抑郁或慢性疼痛治疗中展现出惊人疗效，核心在于它们在极短时间内大幅推高了神经调质（neuromodulators）水平。在心理治疗的框架下，赛洛西宾让血清素飙升，MDMA 则同时暴增血清素与多巴胺，为大脑重构创造了极高的神经可塑性（plasticity），其效果远非日复一日单纯依靠谈话疗法（talk therapy）所能企及。Huberman 预言，未来的临床趋势必将是经颅磁刺激（TMS）与药理学的深度结合，用于精准攻克抑郁症或创伤后应激障碍（PTSD）等情绪与精神疾患。 然而，现有工具依然非常粗糙。以治疗注意缺陷与多动障碍（ADHD）的兴奋剂为例，药物通过粗暴提升前额叶皮层（PFC）活性来增强非特异性专注力；Huberman 风趣地指出，其实只要有一晚高质量睡眠加上适量咖啡因，在睡眠剥夺时再补充一点肌酸，效果就能逼近 Adderall，只是吃 Adderall 的人往往睡不好觉。更关键的瓶颈在于，所有受体在全脑范围内广泛表达，缺乏空间与时间上的特异性，因此抗抑郁药等药物不可避免地伴随性功能障碍、体重骤增骤减或性欲亢进等副作用；电刺激技术同样面临精度不足的困境。 回溯神经科学的发展，对神经系统的“读取与理解”正是沿着提升精度的路径演进的。Huberman 提到了 Dario Amodei 早期鲜为人知的科研经历：在 Mike Berry 的实验室中，Dario 曾通过记录蝾螈等动物视网膜上的光感受器与神经节细胞（ganglion cells），试图破译从特定视觉输入信号到全群神经元响应的对应关系，以此构建从图像输入到大脑深层感知的解码链条。虽然像 Markus Meister 这样在该领域奠基的学术前辈能够通过电极刺激让动物瞬间暴怒或进入交配状态，但现实生活中驱动动物从“探索互动”转向“实际攻击或交配”的内在开关极其精细。神经科学目前仅在触及这些基础行为动机的边缘，至于大众幻想的赛博朋克改造——例如随心所欲提升创造力或智商——科学界连相关的大脑回路都尚未厘清。 ```mermaid flowchart LR accTitle: 从视觉输入到大脑感知的解码路径 accDescr: 展示从视网膜接收光信号经由神经节细胞输出并最终在大脑深层形成感知的完整信号流。 subgraph Retina["视网膜层级 (输入与初步编码)"] A["视觉信号输入 (已知图像)"] --> B["光感受器响应 (Photoreceptors)"] B --> C["神经节细胞输出 (Ganglion Cells)"] end subgraph Brain["大脑深层 (感知构建)"] C --> D["全群神经元响应矩阵"] D --> E["高阶知觉形成 (Perception)"] end ``` ### `52:43` 基因编辑风暴：贺建奎 CRISPR 婴儿事件背后的科学震荡与疑云 话题转向生物技术的极端边界，Huberman 讲述了一段“令人毛骨悚然但确凿发生”的科学历史：数年前，曾在斯坦福大学从事博士后研究、随后回国建立实验室的贺建奎（He Jiankui）在国际学术会议上公然宣布，他利用 CRISPR 基因编辑技术敲除了双胞胎女婴的 HIV 受体基因（CCR5）。官方流传的动机是女婴的父亲为 HIV 阳性，他意图让婴儿免疫艾滋病毒感染。这一举动在整个医学与生命科学界掀起了滔天巨浪，所有人一时间茫然不知所措——究竟是该授予他诺贝尔奖，还是将他投入监狱？ 这一事件暴露了科学界内部极其诡异的社会学群态。当时全球顶尖实验室都掌握了相关技术，但无人敢越雷池一步，而贺建奎直接完成了这一“终极实验”。在随后的调查中，同行们开始翻检过往与他的往来邮件。Huberman 指出，许多科学家当时的私下沟通呈现出极度微妙的投机心态：一边叮嘱“小心点”，一边暗自打探“实验成功了吗？”；大家都在观望事态风向，若他安然无恙，相关学者便能宣称自己参与了这项划时代发现，一旦东窗事发，则立刻划清界限自保。随后全球伦理学界口诛笔伐，中国政府官方也出面定性其违背伦理，撤销其实验室并施加惩处。 更引人深思的是该事件鲜为人知的神经科学隐喻与后续流言： - **记忆与智力的潜在关联**：在神经科学领域，CCR5 受体已知与某些调控学习和记忆形成的神经环路密切相关。当时学界私下热议却避而不谈的核心疑问在于：贺建奎此举是否不仅是为了抵御艾滋病毒，更是为了进行认知增强（Cognitive Enhancement），制造智力超常的婴儿？至今这一问题仍无确切定论。 - **现况与行业灰度**：有传言称他目前在德克萨斯州奥斯汀设有实验室继续开展工作。Huberman 坦承无法核实其真伪，但断言他绝非唯一涉足此类研究的人；贺建奎的受罚仅仅证明了公开宣扬这些研究是危险的，迫使全球潜在的同类实验转入地下。 ### `55:00` 人类基因工程的连续光谱与失控的技术狂飙 Huberman 退一步审视了整个人类基因改造技术的演进光谱。他强调，社会对基因干预的接受度实际上呈现出一条连续而模糊的渐进线，界限从未被清晰划定： | 干预层级 / 实践类型 | 代表技术或机构 | 主要筛选 / 干预指标 | 伦理与社会接受度现状 | |:--- |:--- |:--- |:--- | | **常规产前诊断** | 羊水穿刺（Amniocentesis） | 染色体异常、严重遗传病缺陷 | 临床普遍实践，产妇可自愿选择 | | **体外受精（IVF）胚胎筛查** | Herocyte、Orchid 等商业公司 | 遗传疾病风险、自闭症关联表达、智商（IQ）、身高特征 | 商业化扩展迅速，逐步向非疾病性状渗透 | | **自然择偶与精子库筛选** | 伴侣选择 / 捐精机构匹配 | 学历背景、外貌表征、身高等显性表型 | 社会完全默许，属于自发层面的“优生偏好” | | **基因治疗与体细胞编辑** | 肌肉生成抑制素（Follistatin）基因突变疗法 | 肌肉过度生长、力量爆发力 | 动物实验成熟，已有极客在人体尝试增肌基因疗法 | | **生殖系基因敲除** | CRISPR 胚胎改造（如 CCR5 敲除） | HIV 免疫、神经可塑性与认知增强潜力 | 违反现有国际伦理共识，极度敏感但技术门槛已破 | Huberman 感叹，如果十五年前有人谈论人类通过基因疗法修改 Follistatin 来突破肌肉生长极限，大家会觉得这人疯了；但如今这些技术均已落地成真。然而他更倾向于保持审慎的乐观，因为在自媒体与互联网高度透明的时代，数字足迹无处遁形，一旦多于两个人知晓，技术暗箱便无法长期隐瞒。 正如威廉·吉布森（William Gibson）的经典格言所言：“未来已经到来，只是分布不均。”整个人类技术演进的弧线本质上就是“控制”——从基因层面的操控，到通过神经科技精准调控大脑状态。 ### `01:03:01` 神经科技的新圣杯：解码与书写白昼清醒大脑状态 当被问及若被迫创办一家公司、必须押注单一方向时最兴奋的领域是什么，Huberman 给出了明确的答案：**理解并精准调控清醒大脑状态（Waking Brain States）**。 在媒体与大众传播层面，Huberman 对当今以 X（前 Twitter）和 Instagram 为核心的全球学术“同城校园”感到振奋。跨学科人才在社交网络上直接碰撞，极具魅力且专业扎实的科普领袖正在以惊人速度重塑基因治疗与脑科学的认知边界。与他年轻时在实验室挑灯夜战、只要“不弄死自己、不烧掉实验室”就敢放手尝试的拓荒精神相比，如今的科研环境虽因合规而多了一丝谨小慎微，但新一代技术极客正在向神经系统的“读写能力”发起全面冲击。 在具体的科研探索上，睡眠领域的拼图已经相对完整： - 科学界对慢波睡眠（Slow-wave sleep）、深度睡眠以及 N1、N2、N3 各阶段的脑电架构已了如指掌。 - 温度对睡眠周期的调控机制被充分验证（例如晨间升高床垫或环境温度可有效诱发并延长快速眼动期 REM 睡眠）。 - 针对睡眠状态的测量与干预工具已经高度成熟。 然而，清醒状态的大脑依然是一片巨大的未解之谜。Huberman 坚信，神经科技的终极未来不在于睡眠的被动修复，而在于如何在人体处于清醒、工作、创造和情绪波动的白昼状态下，真正实现对大脑神经回路毫秒级、细胞级的实时读取与靶向改写。 ### `01:00:00` 解析清醒脑状态与科技巨头的神经科学跨界 相比于调节室温这种粗糙的手段，未来借助眼罩类工具对睡眠进行毫米级精确调控是完全可能的。但目前最紧迫的未解之谜在于“清醒脑状态”（Waking Brain States）。人们用快速眼动（REM）描述睡眠，这只是生理眼球运动，并不能表达主观体验；而清醒时的状态——例如当下高度警觉、专注且享受对话的体验——缺乏精准的神经生物学定义。只有彻底理解不同的清醒状态究竟最契合哪些具体活动，科技界才能真正着手以极高的时间与空间精度，向神经系统“写入”指令来精准触发这些状态。 传统的学术神经科学界以及生物技术领域存在一个通病：研究人员受训后，往往只敢在与原本课题相差“四度以内”的微小范畴里打转。而如今，最前沿的突破力量反而来自科技巨头的高管与工程师。Dario Amodei、Sam Altman、Mark Zuckerberg、Marc Andreessen 以及 Elon Musk（通过 Neuralink）正成为神经科学探索的领军人物。他们不再仅仅是编写代码的工程师，而是深入理解脑科学的前沿实践者。虽然科技从业者刚接触大脑时往往会碰壁并意识到“大脑绝非计算机那么简单”，但跨学科的视角正带来巨变。 ```mermaid flowchart LR accTitle: 神经技术驱动清醒脑状态调控的流程图 accDescr: 展示从定义清醒脑状态，到跨界科技巨头介入，最终实现向神经系统高精度写入状态的闭环流程。 subgraph Phase1["科学突破需求"] A["定义清醒脑状态<br/>(专注、警觉、沉浸)"] --> B["解析状态与认知活动的匹配"] end subgraph Phase2["跨界技术融合"] C["科技巨头与工程师入局<br/>(Musk, Altman, Zuckerberg等)"] --> D["顶尖神经医学团队协作<br/>(如 Eddie Chang, Neuralink)"] end subgraph Phase3["终极落地"] B & D --> E["高时空精度<br/>写入神经系统"] end ``` - **脑科学四巨头与核心探索者**：Sam Altman、Mark Zuckerberg、Dario Amodei 与 Elon Musk 被誉为当前脑科学前沿的顶级选手；加州大学旧金山分校（UCSF）的神经外科专家 Eddie Chang 也是该前沿领域的核心领军人物。 - **健康信息周期的预测**：任何领域在取得巨大飞跃前都会经历周期的震荡——先是狂热涌入、公众科普泛滥、信噪比严重恶化，随后热潮退去，最终在 2027 至 2028 年左右沉淀出低噪音、高可用性的普及技术与实用知识。 - **告别旧课题**：新书《Protocols》收录了已完全确证的知识。关于“清晨晒太阳”这类基础科普已经讲得足够透彻，接下来的焦点将全面转向向清醒神经系统读取与写入信号的技术前沿。 --- ### `01:03:01` 科技走向人体内化的终极前沿 对于脑机接口与神经技术的狂飙突进，主流媒体常常试图用恐慌情绪来减速，强求从业者公开发表“我们不会毁灭世界”的声明。这类作秀既无法真正安抚公众，也毫无建设性。建立安全防护体系与伦理机制固然不可或缺，但在该领域倾注心血的都是全球最聪明、最具洞察力的人。从当前人工智能的技术演进逻辑推导，技术与人体之间的物理与神经亲密度必然会越来越高，神经技术正是人类科技探索的“终极前沿”。 --- ### `01:05:00` 追随真正的好奇心：导师 Barbara Chapman 的善意 在被问及“他人为你做过的最善良的一件事”时，一段往事揭示了学术生涯的真正转折点。读研究生期间，实验室轮转后，原本加入了一个研究分子遗传学技术观测神经元的热门尖端实验室，但内心却始终魂牵梦绕于第二个轮转实验室那项相对冷门但让自己彻底着迷的课题。于是常常在深夜偷偷潜入那间实验室独自做实验。 后来担任博士生导师的 Barbara Chapman 教授发现了这件事。她的实验室当时已非常成功、屡屡在《Science》等顶级期刊发表论文。她主动找来，毫无被拒绝的芥蒂，坦诚而大胆地说道： > “你究竟在哪个实验室并不重要。真正决定你能否成功的唯一要素，就是你有多么渴望知道你所研究课题的答案。” 她鼓励断绝与那个热门实验室的关联，回到真正热爱的地方。最终师徒二人联手发表了 8 篇第一作者论文，学术生涯由此起飞。这种善意不仅在于对学生的信任，更在于她看穿了年轻学者缺乏反叛大众共识的勇气，并推了一把，指引其走向内心真正的好奇。 ```mermaid flowchart TD accTitle: 顺应直觉与好奇心的决策对比 accDescr: 对比为了大众所谓成功而妥协的选择，与追随内心真实好奇心带来的长期产出路径。 subgraph PathA["世俗功利路径"] A1["追随热门领域与大众期待"] --> A2["缺乏自驱力与深层渴望"] A2 --> A3["实验/创业遭遇阻力时极易枯竭"] end subgraph PathB["真实好奇心路径"] B1["深挖内心热爱的冷门/本质问题"] --> B2["极度渴望探寻底层答案"] B2 --> B3["构建独特技能与认知壁垒"] B3 --> B4["时代周期轮转后占据前沿领先地位"] end ``` - **顺应周期的长期价值**：遵从好奇心进入彼时“无人问津”的领域，随着学科周期的轮转，离开博士实验室时所沉淀的技能恰好契合了学科下一阶段的核心需求。 - **创办播客的初心**：放下全职实验室转向公众媒体同样源于这种内在驱动力。尽管毫无媒体经验，但内心深知这套知识体系能真正帮到大众，便义无反顾全力以赴。 --- ### `01:06:12` 拒绝世俗规训：做自己才能避免精神的慢性死亡 导师那句话的深层精髓在于：**“成功的决定因素，在于你将自己真正的好奇心与你所从事的工作结合得有多精准。”** 无论是投身科研、创办企业，还是做出一档真正优质的播客，过程都极其艰难险阻。若做的事情并非出自真心，根本不可能熬过漫长严苛的磨练。 Elon Musk 能够赢得广泛的敬佩（或部分人的嫉妒），核心不在于财富，而在于他完全无视世人指手画脚的能力。曾有做客播客的嘉宾主张顶级科技领袖应该“务实地帮扶眼前人”，而不该去搞什么太空与火星探索。这种论调极其荒谬——“这就像要求 Metallica 乐队去演奏 Grateful Dead 的乡村民谣一样”。Metallica 之所以成为 Metallica，是因为他们深爱着重金属摇滚；Grateful Dead 亦然。 世间最大的悲剧，莫过于人们为了追求所谓的世俗“成功”，而压抑并背离了自己内心真正的好奇心。不遵从好奇心去解答真正关切的难题，在精神层面上无异于一场漫长而痛苦的慢性死亡。 ## Notable quotes > `20:00` It's very easy to get attention. It's very hard to keep respect. — Andrew Huberman > `01:05:00` The single most important variable in whether or not you're going to be successful is how much you want to know the answer to what you're working on. — Barbara Chapman > `55:00` The future is already here, it's just not evenly distributed. — William Gibson > `01:06:12` You can't ask Metallica to play like the Grateful Dead. — Andrew Huberman > `22:07` For decades the field has focused on reading from the brain, and the next era is about writing to it with precise control over specific brain states. — Andrew Huberman > `22:07` The leading AI companies will become biotech companies. — Andrew Huberman > `05:00` Don't die, don't get sick, don't get badly injured, and extract the maximum amount of output you can within the boundaries of what your body can tolerate. — Andrew Huberman > `01:06:12` The biggest tragedy is doing what you think you should do instead of what you are obsessed with doing. — Andrew Huberman ## Mentioned Andrew Huberman, Elon Musk, Jimmy Iovine, Jack Dorsey, Eddie Chang, Karl Deisseroth, Matt McDougall, Sam Altman, Mark Mayford, Susumu Tonegawa, Melissa Almest, Markus Meister, Dario Amodei, Mike Berry, He Jiankui, Barbara Chapman, Mark Zuckerberg, Marc Andreessen, Tim Ferriss, Brian Johnson, Gary Brecka, Dave Asprey, Phineas Gage, Patient H.M., William Gibson, Metallica, Grateful Dead, Huberman Lab, Protocols, Neuralink, OpenAI, Anthropic, Meta, MIT, Stanford University, UCSF, Scripps Research, Whoop, Oura Ring, Eight Sleep, Waymo, Herocyte, Orchid, CRISPR, CCR5, Follistatin, GLP-1, TMS, fMRI, EEG ## Video description Andrew Huberman is a neuroscientist and professor at Stanford, the host of the Huberman Lab podcast, and the author of Protocols. Most of our conversation is about where neuroscience goes next. For decades the field has focused on reading from the brain, and Andrew believes the next era is about writing to it with precise control over specific brain states. He explains why he thinks the leading AI companies are really building toward this, and why understanding our waking brain states is the problem he most wants to see solved. Please enjoy my conversation with Andrew Huberman. For the full show notes, transcript, and links to mentioned content, check out the episode page ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠here⁠⁠⁠⁠⁠.&nbsp; ----- Become a Colossus member to get our quarterly print magazine and private audio experience, including exclusive profiles and early access to select episodes. Subscribe at ⁠colossus.com/subscribe⁠. ----- ⁠Ramp’s⁠ mission is to help companies manage their spend in a way that reduces expenses and frees up time for teams to work on more valuable projects. Go to⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ ⁠ramp.com/invest⁠⁠ to sign up for free and get a $250 welcome bonus. ----- Trusted by thousands of businesses, ⁠Vanta⁠ continuously monitors your security posture and streamlines audits so you can win enterprise deals and build customer trust without the traditional overhead. Invest Like the Best listeners get a special offer of $1,000 off Vanta when you go to ⁠vanta.com/invest⁠.&nbsp; ----- WorkOS⁠ is the infrastructure B2B and AI-native companies use to sell to enterprise. It covers everything enterprise security requires: SSO, SCIM, RBAC, Audit Logs, AI governance, and more. Trusted by 2,000+ fast-growing companies, including OpenAI, Anthropic, Cursor, and Vercel. ----- Rogo is the AI platform for finance. They're building agents for Wall Street that are trained to understand how bankers and investors actually do work: from diligence and modeling, to turning analysis into deliverables. To learn more, visit rogo.ai/invest. ----- ⁠Ridgeline⁠ has built a complete, real-time, modern operating system for investment managers. It handles trading, portfolio management, compliance, customer reporting, and much more through an all-in-one real-time cloud platform. Visit⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ridgeline.ai⁠. ----- Editing and post-production work for this episode was provided by The Podcast Consultant. Timestamps: (00:00:00) Energy as the Bottleneck for Top Performers (00:06:20) How to Downshift and Fall Asleep (00:14:06) Neurotech for Sleep and Focus (00:17:57) The State of Public Health Discourse (00:22:07) The Biggest Problem in Neuroscience (00:23:30) Why AI Companies Will Become Biotech Companies (00:26:18) How We Read Brain Activity (00:42:07) Crossing the Cyborg Line (00:43:30) Measuring the Nervous System Directly (00:57:34) The CRISPR Babies Story (01:00:24) Embryo Screening and Genetic Engineering (01:03:01) Understanding Waking Brain States (01:06:12) Technologists Becoming Neuroscientists (01:09:35) The Kindest Thing [Source](https://traffic.megaphone.fm/CLS1239827470.mp3)

[20:00] Andrew Huberman: It's very easy to get attention. It's very hard to keep respect. [65:00] Barbara Chapman: The single most important variable in whether or not you're going to be successful is how much you want to know the answer to what you're working on. [55:00] William Gibson: The future is already here, it's just not evenly distributed. [66:12] Andrew Huberman: You can't ask Metallica to play like the Grateful Dead. [22:07] Andrew Huberman: For decades the field has focused on reading from the brain, and the next era is about writing to it with precise control over specific brain states. [22:07] Andrew Huberman: The leading AI companies will become biotech companies. [5:00] Andrew Huberman: Don't die, don't get sick, don't get badly injured, and extract the maximum amount of output you can within the boundaries of what your body can tolerate. [66:12] Andrew Huberman: The biggest tragedy is doing what you think you should do instead of what you are obsessed with doing.

[0:00] Andrew Huberman: I can't tell you. [3:15] Andrew Huberman: ridiculous. [3:15] Andrew Huberman: two nights like that of four to zero hours of sleep and I'm cratered. [5:00] Andrew Huberman: don't die. [15:48] Andrew Huberman: look like shit. [18:24] Andrew Huberman: I feel like the public discourse around health and longevity is a fucking disaster. [20:00] Andrew Huberman: it's very easy to get attention. It's very hard to keep respect. [66:12] Barbara Chapman: Listen, I'm totally okay with the fact that you didn't join my lab... but you just seem to really love this stuff. And the only thing that's really going to determine your success is how badly do you want to know the answers to the questions that you're working on? [68:30] Andrew Huberman: That's like telling Metallica to play the Grateful Dead. Metallica are Metallica because that's what they love... and the Grateful Dead are the Grateful Dead because that's the music they love. [68:30] Andrew Huberman: a slow death.

[0:00] Ramp is the only platform built to make your finance team leaner, faster, and better, saving businesses 5% annually on average so you can stay focused on growth. Ramp just opened for business in the United Kingdom, so if you're running a business in the UK, you can now use Ramp's AI-powered finance platform to manage cards, expenses, bills, approvals, and accounting all in one place. I run my business on Ramp, and so should you. Learn more at ramp.com slash invest. Thank you. Open AI, Cursor, Anthropic, Perplexity, and Vercel all have something in common. They all use WorkOS. And here's why. To achieve enterprise adoption at scale, you have to deliver on core capabilities like SSO, Skim, RBAC, and Audit Logs. That's where WorkOS comes in. Instead of spending months building these mission-critical capabilities yourself, you can just use WorkOS APIs to gain all of them on day zero. run on WorkOS. WorkOS is the fastest way to become enterprise ready and stay focused on what matters most, your product. Visit WorkOS.com to get started. Rogo is the AI platform purpose-built for financial institutions, serving tens of thousands of bankers and investors at hundreds of leading firms worldwide. Rogo's AI agents autonomously execute large chunk of your firm's workflows. They can screen deals, draft sims, run buyer outreach, and do diligence on data rooms, all with full security and compliance handled. stronger than the last. To learn more, visit rogo.com slash invest. there's a very interesting thing happening that relates to a piece of your work which i haven't seen you talk about as much and so maybe it's a fun place to begin which is that it feels like we're now bandwidth limited on the most talented people's ability to do work Like it used to be that the rest of the world couldn't keep up with the most talented people. And so they would get frustrated that no one was moving as fast or as efficiently as they were. There seems to be something flipping, which is actually some of these people would earn a really high return if they could generate more energy, maintain more energy, not burn out. I heard this from several great entrepreneurs this summer that I think maybe Elon has just gotten into everyone's head that the real greats are going to run multiple companies. technology unlocks so much potential for the most talented people. And so this seems like an actual bottleneck is like these people's energy. And I'd be curious in all your research and study what you've learned about the component parts of this, like generating energy, maintaining it, clean versus dirty fuel, long-term performance. Like all of these things are fascinating and I think actually important to the degree of progress. And so I would love you to just riff on all that for a while. with the caveat that Elon represents the apex of being able to do multiple huge things. simultaneously. And I don't think he has any peers in that regard. In fact, one of the best multi-billionaire investors from the Bay Area who may have gone to Stanford and may have been part of the PayPal early team, has asked me, how is it that Elon does this? I'm like, I can't tell you. Like physiologically. Physiologically. I mean, there are examples throughout history of people who seem to require less sleep. They just have more energy. And most people can get their energy up There are a lot of healthy ways to do that now. And I think now we've undergone this huge shift. whereby everyone understands you have to sleep eventually and you have to know your own sleep requirements. So to dilute oneself into thinking you only need four or five hours because Elon only needs four or five hours, that's ridiculous. an honest self-assessment about how much sleep one actually needs to perform well and perhaps best and what you can get by on. If you can't get that, that's an important thing to understand. But most people know, like for me, it's seven hours. I feel great. Eight probably puts me into the day a little bit groggy. Six, I can manage just fine. Five for one day or a sleepless night. I'm good. two nights like that of four to zero hours of sleep and I'm cratered. And I think most people are similar. What's really changed is that we have effective, healthy ways to build our energy up. We know to sleep. We can talk about exercise, caffeine. Some people are using nicotine. Yes, it's addictive. It raises blood pressure, but it does put people into that focused but relaxed mode. I always say if you really want the benefits of nicotine, don't use it very often. If you want the actual cognitive enhancing effects or the focus enhancing effects, then you have to not be a chronic user and then spike your system with nicotine at low amounts every once in a while. Setting Elon aside with the understanding that he represents that extreme in terms of energy and sleep requirements being very low, I think what's really happening now is people are hopefully doing this self-assessment. Like I would hope people would have an honest conversation about themselves and say, at this point in my life, with or without kids, [5:00] at this stage of my company, at this stage of my career, whatever it is, at this age, how much sleep do I need in order to feel awesome, good enough, or it's not enough? And I think most people know that. Okay, so that's step number one. The next one is something most people don't ever talk about, which is some people actually just have more energy. They can focus for long, long periods of time. They have a ton of energy to run. They have a ton of energy to then go work. But what we never ask is, What is that person's internal state? So I know people that can do all of that, but they're kind of redlining and they're using a lot of stimulants to do it. or they're not using stimulants to do it, but they're redlining. And then at night, they're having a hard time switching it off. And so they're not resetting their system. Some people, however, are these like wake up at 4 a.m., work out, handle business, deal with family, go off to work, back to business, come home, and they pivot and task switch really fast. I don't think these things are necessarily stable features throughout our entire life. Seems to be a lot of young guys come up to me. They're like, what should I do? The Oracle said it best, like know thyself. You really have to be honest with yourself about what you need in order to do something for a long period of time. Now, that said, I do recall working 80, even 100 hour weeks in graduate school, and that served me well. I mean, eventually led me to tenure at Stanford, etc. I do think that if you can pull it off, I think from zero to maybe first four years of starting a company, or if you're in medical school, the goal is don't die. Get as much done as you can, and don't impair yourself with an injury or chronic disease. We don't really define this thing optimization. And then you start to pivot into how can I insert components of rest along the way, maybe during the day or across the week or the month that really allow me to access my best levels of focus and creativity. But I'd be naive and I'd be lying if I said somebody starting a company in the AI realm in the Bay Area or elsewhere these days, you're going to sit them down and go, listen, you really need eight hours of sleep. I do your resistance training, should do some breath work. Like, no, none of that. What they should determine is how much energy they actually have. And you've met these people, right? Occasionally you'll run into like a Jimmy Iovine or some of the folks doing AI at Meta, and they just seem to have an extra gear. Some people have this extra gear. Now, there's one other component we can deep dive into any of these that nobody talks about. And that's the ability to switch off your thoughts, your planning and your concerns at the end of your day and drop into sleep. Yeah. Can we go deeper in that one? Because to be very selfish about it, if I was to self-diagnose, the biggest problem I have is downshifting. Sometimes we'll take literally four hours to go from a crazy, exciting, awesome day to literally even be able to fall asleep or just like stop the mind. If I could bestow myself with any gift, it would be what you're describing. So yeah. What is the research there of how to do that? So in terms of sleep, there are a couple of prerequisites for sleep. There are a bunch of don'ts. So obviously, you don't want to drink caffeine too late. Being under too much bright light late in the day can make it difficult, but a lot of us can't control that. Alcohol, cannabis, these things impair the quality of your sleep, help you get to sleep. We all know that by now. But in terms of being able to turn your thoughts off, This is a very, very difficult thing for people for good reasons. The brain wants to think, plan, remember, and organize itself around what's coming next because such a huge aspect of brain function is a prediction of what's happening next. So first requirement, Turn your thoughts off. Second requirement is slow your heart rate down. Okay, so that's purely of body. And then third is... you need to forget about the position of your limbs in space. So let's work through each one systematically. For the first one, turning your thoughts off is a skill. We can think, we can predict, we can remember, but we also can get into pure sensation. And that's the gateway to turning off your thoughts. And this doesn't come from Eastern tradition per se, although this has been discussed for a long time in those traditions. But there's a neuroscience correlate to all of this, of course, and we can talk about those. So how do you do that? You have to focus on pure sensation. So you start focusing on the surface of your body. If you can detect your heart rate, Some people are very good at what's called interoception. They can actually get pretty accurate measurements of their heart rate without pressing on a vein or artery. And they're just listening to their own breathing. They're listening to the sounds in the room. They're bringing their conscious awareness just to the body. And it generally helps. And this relates to the second thing to slow the heart rate down by just doing long exhales. Now, this might sound very woo, but I want to be very clear. This is not meditation. Because meditation has you do something very different. Meditation has you focus all of your attention on your breathing or all of your attention on some region behind your forehead or your hands or whatever. In this sort of practice, what you do is you actually migrate your attention to your feet, to your legs, to your breathing, to sounds in the room and back again. Now, this might sound really squishy. But you're moving away from thinking and planning to pure sensation and perception. And you're literally bringing your perception into the present. Remember, this is one door in front of turning off [10:00] thoughts. To actually turn off your thoughts, your heart rate actually has to drop a certain amount. Those long exhales drop your heart rate through something called respiratory sinus arrhythmia. Arrhythmia makes it sound dangerous or bad, but it's actually very good. As you're breathing, the size of your heart is literally changing based on the amount of space it has because the movement of the diaphragm and the filling of the lungs. And when we inhale, our heart rate actually speeds up a bit. When we exhale, The heart actually gets a little bit smaller volume-wise, and the blood, therefore, that's in it, moves a little bit quicker per unit volume. And the brain sends a signal to slow the heart down. So when you exhale, you're slowing the heart rate down through this thing that we call the vagus nerve. So if you want to slow your heart rate down anytime in waking to de-stress or just bring your level of activation down, or if you want to fall asleep, It really helps to do some long exhales until lungs are empty. and then just passive inhales. Normally, when we're really activated, we're doing a lot of inhaling or passively exhaling. And then the last piece is to be able to turn off your perception of your limbs relative to you. It's something that we call proprioception. The three ways to do this, one is old school, everyone accepts it. One is pretty wacky sounding, and then there's a new technology that's coming out that I wish I had developed that I think is gonna make people more comfortable with the second one. So here goes. Believe it or not, there've been studies of what puts babies to sleep and what puts adults to sleep without pharmacology. And it turns out that the rocking pacing that we do with babies at a certain rate. Parents figure this out on their own, puts them to sleep. Turns out that you can build, and they've done this for studies, a bed that rocks at a certain frequency. and adults will fall asleep very quickly. What it has to do with is the fact that when you're moving side to side, your eyes are generating under your closed eyelids compensatory eye movements. Our eye movements all the time are communicating through our brain stem to our cerebellum. to control balance and proprioception. So when people are rocked this way, somehow they fall asleep more quickly. That's interesting, actually. And the frequency at which that works best and fastest is known. Now, I don't have a rocking bed. I don't know anyone large enough to rock me and hold me. And if they did, I'm not sure I'd feel comfortable with that. Can't imagine you would either. So there are two ways to approach this. One is something that looks kind of wacky but works, which is to mimic the eye movements under closed eyelids that help you forget about your limb position. I've been talking about this a lot in public education recently, and I confess it makes me feel really weird as a neuroscientist. Somebody just prefaced this by saying, please don't just put the part out into the world allowing me to first say, that this confuses the circuit between, for the aficionados, these direction selective cells in the retina, the accessory optic nuclei of the brainstem and the cerebellum. and you forget about body position. And basically what you do is under closed eyelids, you move your eyes to one side, then the other, slowly or fast, doesn't really matter, then up, then down. You can roll them counterclockwise and clockwise. You can do a long exhale, but really what you're doing is confusing this system. The reason I got so interested in this and the reason I went to the literature on rocking of babies and adults to fall asleep is because a group out of MIT approached me and they said, we'd like you to try this eye mask. The eyes are two pieces of brain, they're just not in the cranial vault, which makes them unusual in terms of parts of brain. So the eyes are an amazing gateway to the rest of the brain. Everyone knows about eye masks to get the darkness out. These guys have developed eye masks that measure rapid eye movement, because right now your whoop, your aura, your eighth sleep, they're not measuring REM directly. This can measure REM directly. But they designed this mask specifically to make people fall asleep faster. Like, I want that thing. So I got to try it and it works very well in the daytime or at night. And what does it do? It generates these slow back and forth, up and down eye movements through a small amount of stimulation behind your ears, because it turns out you can access the extraocular muscles well enough from that location. So just an eye mask with some straps. You sometimes feel a little bit of the stimulation and you're out. And then once REM sleep starts, It actually helps amplify the REM stage. You can get much more REM. So I wish I had a stake in this company. I don't. These technologies are about seven to 12 months from release. I'm super excited about them. I'm also super excited about the fact that people can just do this on their own. And obviously, babies get rocked. You could do the non-mask tool approach, or you could use the mask. I will say, and I don't want to take us too far off course, but for those interested in emerging technologies and AI, in particular, neural technologies, I do think that in sleep, reading from and writing to the brain is going to be a huge area. But of course, I'm also interested in reading from and writing to the nervous system in waking states. Like you can imagine that within 12 months or a year of the release of the technology that I just described, you can imagine somebody developing eyeglasses like this. that are stimulating some component of the autonomic nervous system. [15:00] so that I'm exceptionally focused at certain times of day. And then when I'm walking home, I'm in a slightly more relaxed state to capture some more of that energy. We're going to start shifting brain states into focus. into modes for learning, and we're going to start doing it with technologies. Right now, we're doing cold showers, caffeine, exercise we're doing inhale emphasized breathing versus exhale emphasized breathing to get more alert or calm if you think about it all breathing exercises can be summarized as inhale emphasize or exhale emphasize So, Yes, we can do that stuff deliberately, but you can imagine that pretty soon, there will be non-invasive stimulation devices that will allow people to do this, but it will start in sleep because we're all kind of comfortable with the idea now of measuring our physiology and sleep. a little bit in waking states. But this is where this is going. So to return to this in sequence, you need to learn to turn your thoughts off. you need to forget about your body position and slow your heart rate down. And the order in which you do that doesn't really matter. You ideally want to combine all three. But I think that's going to be a significant achievement for humanity. Because if you look at recovery from illness, healing, learning, stress control, burnout, right? If you're going through grief or psychological stress or trauma healing, if you're trying to start a company and you got kids, or even if you're just trying to work the maximum number of hours and you have four hours to sleep and you're somebody who needs six, you want to be able to hit the pillow and shut it off. And we don't have good ways to do that except the tools I described. So I think we are starting to go about things in a more healthy way. But what I'm talking about is people doing a thorough self-assessment of sleep need, ability to turn off thoughts, and then training those things. In fact, a lot of hard driving people will come to me and say like, what workout should I do? I'm like, look, How much time do you have? In graduate school, I worked out twice a week. I don't talk about this, but I lifted weights and I ran really far because I just wanted to maintain while I worked my ass off publishing as many great papers as I could, And then later I'm doing, you know, three workouts a week with the weights. I do cardio three times a week, long, medium, and short. take a day off, maybe do some sauna and cold. My life's a little bit different now or a lot different. But I do think those first years, you're trying to just not die while you create this thing. Then you start moving into some balance and optimization. And then you get the people who have really achieved some level of mastery or virtuosity in their craft. And you really get an assessment of how accurately their schedule matched who they really were at a physiological level by looking at them and going like, they look like shit. They have a chronic illness. They look 10 years older than they were supposed to. Or there are certain people who are like phenoms, like in Elon, or you look at, Some people can just go, "Holy crap, they just got an extra gear." And I'm speculating here, but I'm guessing that Jack Dorsey was always a pretty calm, thoughtful guy. That's what people tell me of him. And that's how I experience him too. But I'm guessing he's also put some work into it. He's figured out that staying alert and calm all day and sleeping well at night is the key to having a really long, successful arc. Since you've been covering it for so long, I'm curious what you think the biggest challenge or problem or issue is facing your whole field. Like if I think about this as the quest to learn and teach, to understand everything. our health and our wellbeing and ways of making it better and What is the biggest challenge? Facing the field today, do you think? I mean, I feel like the public discourse around health and longevity is a fucking disaster. He is a bunch of clowns. And people are doubling and tripling down on the theatrics. I'm not saying I get everything right or that I have all the answers. I'm definitely not saying that. I got into this at the right time, but... The downside is that it was so devoid of anyone else. There are people talking about health only in their area that they worked on in their lab, but not broadly. that, you know, the bar for entry in public discourse around health is basically zero. You know, it's like you show your abs and you admit you're on steroids. You have a following. Some dude shows his abs or some woman like shows her glutes and claims to be all natural. And we believe them then. And suddenly they have a following. So, I mean, if that's the bar for entry, I come from a field where having a PhD isn't even nearly sufficient. It's necessary, but not even nearly sufficient to say that you've contributed something to the understanding of the brain. Like we could take somebody like Tim Ferriss was talking about a lot of this stuff early on, the biohacking self-optimization. And what I love about Tim is he's remained Tim. There was this period where he experimented. He had some things that he felt probably weren't a good idea. You know, he talks about those like things that he wishes he hadn't done and things that he's happy he did. And then he's also in investing and all these other things. I think Tim's fantastic because he's being Tim. So I'm not referring to that. And then you started getting these more and more extreme versions where people were coming in with basically no science or health training and saying, hey, like, here's how you optimize the mind and body. And I know this because I'm doing all these things and I used to be fat and I'm not fat anymore. And I think that's fine in principle. But then it just keeps getting crazier and crazier. And so what you [20:00] seeing now is there's a Big gap between, if I'm just honest, between the A-tier public health educators and what used to be the B-tier educators. Public educators are just getting bigger and bigger and bigger. Because one has to be true to self, but you also have to take the endeavor seriously. It's very easy to get attention. It's very hard to keep respect. So I do think that the public health discourse is harmed. when it becomes about the theatrics. And I'm not trying to not name names here. I mean, I think everyone can look out on the landscape of everyone from Brian to me to Gary Brecka to Asprey to anyone else that are a lot in the space. And just go like whose persona resonates, not just what captures the entertainment value. So what I would expect will happen next now that we've gone from high amounts of interest in this to this optimization, optimization. Now it's turned into are we over optimizing? I think we're going to see a downturn in the number of things that people are willing to do each day. They're like, it's just too much. It's gotten kind of kooky. It's kind of crazy. And the characters, I need more serious peers. I need serious peers with training. in either science or medicine, maybe pharmacology, and they're out there. But I think those people are going to rise to the surface. We haven't seen them yet. So I look forward to having colleagues like that. And then I think what's going to happen is we're going to get many, many more people trained as PhDs, trained as MDs who are going to be doing... serious science health discourse. Not regressing to the TikTok dance model, the focus on me model, and really educating people in a way that's unique that I – I'm certain I haven't even tapped into. Someone's going to come along that's better than me. So I think this field is going to go into a downturn now. The public education around health and optimization, longevity, et cetera, is going to go into a bit of a downturn. And then I think we're going to get an influx of a lot of great people. And then it will be where I would like to see it. How about on the research side, biggest problem facing the field? Well, I think the coolest problem that's also the biggest problem is how we can non-invasively write to the nervous system in sleep and waking states. So, My best friend from childhood, Eddie Chang, is chair of neurosurgery at UCSF. He's a bioengineer. He's doing things similar-ish to what Neuralink is doing, but removing brain tumors and this kind of thing. I've been with him to surgeries, like, right up there next to the tumors. Patient is awake. He's probing areas of the brain. He has permission to map areas related to speech and language, etc., So breaching the skull is a big deal. The neurosurgeons say, well, you know, a titanium plate would be better. It's stronger. It lasts longer. But most people don't want to have to do that. They do it because they need something taken out like a tumor or an epileptic foci. I think the big one now is who can develop, it will probably be ultrasound, but a non-invasive way of selectively activating or quieting brain areas. with high amounts of temporal precision. Like getting people who are in locked in syndrome to speak as Eddie has done. or getting people who have a spinal injury to walk, which Neuralink and Eddie and others are trying to develop technologies to do, that is heroic, right? But ultimately, and I'll go on record saying this, anyone who thinks that, Meta, OpenAI, Anthropic, or Elon's got Neuralink as a stated neural interest. Anyone that thinks that those are LLM AI companies is perhaps going to be surprised at this prediction. Every single one of those is going to be a biotech company. They're all interested in the brain. You talk to any of those guys. They may not have formal training in it, but they understand a lot of neuroscience and they will all tell you. Here's the goal, right? The goal is to be able to read and write from your brain non-invasively. And there are straightforward ways to do that. So the big thing is, how can you stimulate a brain area with temporal and spatial precision? How can you quiet that brain area with spatial and temporal precision? How can you dial up or down the level of quieting? Because it's one thing to say silent versus activated, but the nervous system has a lot of graded signals in it. I mean, we don't think about those signals, but the communication between areas is gated. It's graded. thresholds for activation change. even just for things like drive and motivation. A non-invasive device where you could literally crank up your level of motivation by stimulating the nucleus accumbens and related pathways while engaging in a particular learning task or in hard work and then turning it off so that you can capture more energy. You have more of a square wave function on your energy output. That's coming. And those companies are all about that. That's what the real arms race of AI is about. So I sort of chuckle when I think like, okay, like the LLMs are great or terrible, depending on who you are. And I believe they're great. I'm a big fan of AI, but the dynamic modulation of your own neural activity, using non-invasive technologies, not pharmacology, but then using AI to be able to understand, okay, this level of activation in this brain area for you, Patrick, is related to a hyper-motivated brain state that makes it very hard for you to do anything else the rest of the day. And it will just learn to titrate. But you say, well, that sounds kind of science fiction. How far off is that? I mean, my eight sleep changes the temperature throughout the night to give me more REM sleep or deep sleep. It's already doing that. It's just we're OK with these things in sleep. And it's doing that through [25:00] a powerful but indirect method, temperature change. But it's no small thing to dynamically communicate change and measure the degree of change. in the nervous system in order to access a given state sleep or otherwise. better. And... That's what these companies want to do. So my field of neuroscience has to embrace this. And they are, that the people who are trying to figure out What are the natural signals occurring in the brain when people think, when people see, when people move their limbs in a particular way? That's been the last 30, 40 years. That's all reading from the nervous system. Now it's about writing to the nervous system non-invasive. I have a stupid basic question. What is the method by which we can read and how precise? is our ability to read activity? And what does activity even mean in different parts of the brain? Great questions. So it depends on how invasive you want to get. So most people understand, or if you have a no big deal, that neurons communicate through action potentials, electrical signals. And we were taught, and it's still in the textbooks, that they're always the same size and shape for a given neuron. Turns out that's not even true. So there's a variability there, but we sort of have this all or none. Like a neuron either fires or it doesn't, but there's graded potentials. You get to a threshold, the neuron fires, communicates with the next neuron, depending on the milieu there. how many other inputs, what they're doing, the next neuron fires. You sort of set off a chain or not. So you can record from neurons by... putting an electro close to them, extracellular recordings. You can put a bed of nails nearby, record lots of those extracellular signals. You can impale a neuron, record intracellularly, All of that requires breaching the skull, sticking wires basically down into the brain. The methods that are really exciting are things like ultrasound, transcranial magnetic stimulation, which is used for the treatment of depression and other things. The spatial control there of that's for stimulation is not great. The recording from the nervous system, You can go from what I just described, intracellular recording at the most extreme, then extracellular recording, and then you go all the way to like fMRI. or EEG, when people see these like heat maps of the brain, where you really can only measure signals in the surface, like at the cortex, which is interesting, but you're not going to get to the deeper layers. A big problem now is how important is it to understand the natural patterns of activity in detail? Before we start poking around, and stimulating with writing. And obviously there are two ways to write, at least two. You can use pharmacology. So what does most pharmacology do, whether it's alcohol, barbiturates, sedatives of some kind. What are you doing? You're increasing the threshold for activation. You're quieting the brain by just increasing the threshold for activation. You're doing it indirectly. by increasing inhibitory release, so there's less excitation overall. You think about Adderall, caffeine, nicotine, speed, modafinil. And you're talking about lowering the threshold because those generally work by increasing neuromodulators that increase the likelihood that neurons will fire. So you're just kind of ramping up that baseline. You're sort of creating higher RPM across the brain and body. That's why they lead to some shaking, some agitation. Like everything's like primed to move, primed to fire, primed to think. It's hard to shut off your mind when you're under the influence of stimulants. easier when you're under the influence of these other drugs that increase inhibition. So the real question for trying to get to better reading and writing to the nervous system. is about spatial and temporal precision and how willing you are to put something underneath the skin. I say skin because you don't necessarily have to go under the skull. Like let's take vagal stimulation, for example. Vagal stimulation, most people think about as a way to calm down, but most of the vagal pathways are excited toward it. And so common treatments for depression now, if somebody's willing, is to put a small, a little smaller than a penny sized stimulator, under the skin of the neck and stimulate the vagus in order to overcome this sort of sedation that accompanies depression. So vagal stimulation increases alertness. Neuroscientists have known this for a long time, and a lot of people are getting... Pretty good. results in their treatment of their depression by increasing vagal stimulation. And it's dynamic. You can say, how do you feel? I feel okay. You can actually see this in real time. There's an amazing thing that my colleague Carl Diceroth, one of the greatest bioengineers alive, Let's talk about where he's talking to a patient and she's suicidally depressed. and then he starts increasing the stimulation on her vagus and she starts moving from that to a discussion about how she feels willing enough to go out and apply for a job. in real time. It just tells you that these excitatory mechanisms and these depressive mechanisms or inhibitory mechanisms are profound in terms of our perception of the outside world, our perception of ourself, So this would be very nice to have. The question is, are you willing to go under the skin? I am. A guy that came up, not mainly through my lab, but he spent some time in my lab. He's now the head neurosurgeon at Neuralink, Matt McDougall. Amazing guy. has a little receiver implanted under the skin of his hand between his thumb and his index finger. And his wife does too. And it opens their door to their home, to keys. Like, I was like, what are you doing? He's like, it's a sort of biohacking Bay Area-ish kind of stuff. He's like, yeah, like, I'm just playing with the idea of, [30:00] What is it like to have a small device under your skin that can do stuff in the world for you? That's very neural link-ish. And obviously they have much bigger aspirations than that. But I think within 12 months, 20%. of the entrepreneurial world will be... asking, huh, if it's only as invasive as getting a piercing, Yeah, maybe I want a little vagal stimulator. And that's coming next. People are thinking like peptides, what are we going to do? I'm not worried about peptides. I'm thinking implantable devices is where it's going to go next. The commercially available vagal stimulators. Some have shown some promise. They're not super impressive to my neurosurgeon colleagues who like to get under the skin and under the skull. But I will say there's some interesting results there. but we're not quite there. But think about 10 years ago, If I told you, listen, you're going to be wearing a ring or a band on your wrist, I'm that measures your sleep and HRV. That's pretty out there. I'm so fascinated by this. So there's reading the underlying activity of the brain at different levels of fidelity. There's writing. You talked about ultrasound stimulation. I think there's electrical. I think there's magnetic stimulation potentially of these areas of the brain. One central part that I don't really know much about is the model of the brain itself. How far along in our understanding of that we even are? I know there's some things that we can say activation of this area produces this thing, but like it feels like we need those three things, like good input modalities, good reading methods, and a model. Yeah. I'm glad you're asking. I mean, these are my favorite topics. It's just I mean, it depends on what your pain points are. So. at the periphery, the retina. the cochlea, the tongue, the olfactory system, the skin, the skin, neuromuscular control. which isn't at the periphery, but I think we know pretty darn well how the nervous system works. And we know how different wavelengths of light are transformed into visual images at a, not even a first approximation, but we understand that in significant detail, how motion is processed. We understand a fair amount about memories, although I want to put that on the shelf for a moment because there's an important finding in the memory field that will make you feel like it's in kind of twilight zone, perhaps in neuroscience. It worries me all the time. The hypothalamus. So circuits that are deep in the brain that control hunger, anger, sexual appetite, anxiety, lack thereof, hormone secretion, those are switches. So do we understand those? Yes. Do we understand everything? No. but we understand them. I mean, one of the reasons the GLPs are so effective is, And one of the reasons they got so much velocity is because we understand where they bind in the brain, and they bind to these hypothalamic and other areas, of course, that activate neurons that are involved in the suppression of appetite. And so it's a very simple equation, really. GLP levels up thousandfold. People aren't as hungry, but it turns out they're not as driven for other things, too, if you get beyond a certain level. I think Sam Altman talked about his experience with that at one point and why he decided to either back off the dosage or stop taking them. So as you move in from the periphery of the sensory organs, right, and as you move up from the base of the brain, the deeper limbic structures and hypothalamic structures to the prefrontal cortex, we can say, yeah, like your prefrontal cortex, we know based on lesion studies in humans, activation studies. is really important for contextual learning, for strategy setting in different environments, for suppression of impulse, what we call top-down suppression. So do we understand what the prefrontal cortex is doing? Yeah. The place where it starts to get murky is, is at the level of memory, we'll talk about that in a moment, and at the level of memory, Thoughts. You won't find a chapter in a neuroscience textbook that says this is how thoughts are made. And we can talk about what's starting to emerge there, these dynamic attractor states. We'll get to those in a moment. But here's the eerie results from the study of the hippocampus in memory. that freak me out and a lot of neuroscientists don't want to talk about this because it throws all the read-write stuff that we were talking about earlier, all the technologies and pharmacology into a complicated, place because it doesn't allow for clean matching of conceptual understanding of the brain from experiments. to levers that actually allow us some control. The experiments are like this. Mark Mayford's lab has done this down at Scripps. Susumu Tanagawa, who won a Nobel for his work on immunoglobulins, but then became a very successful neuroscientist. They've done these experiments. where they have an animal do some sort of learning task. Could either learn that they get shocked over here and not there, they get a reward over here and not there, or they learn a maze or they learn a lever press combination that gets them a reward. They can learn pretty sophisticated things for a mouse. Monkeys can do this too. Humans can do this too. Because ways to read from the brain, including the hippocampus, a deep-ish brain structure, they've gotten pretty good. You can say, okay, these are the patterns of signals that happened as the animal learned. And here are the patterns of signals that occur every [35:00] - Time the animal expresses this thing. We imagine there's a correlate where every time somebody serves a tennis ball the way they perfected it, that the same cluster of neurons is activated in the brain. They're communicating with each other, the same circuit, rather. And in a particular sequence, A leads to B leads to C and so on, you get the tennis serve. The data tell us that if you tag the neurons that were involved, in this behavior. and then reactivate them, you expect the behavior, right? And indeed, that's what you find. hypothetically, you serve the tennis ball. I'm able to measure, let's just say with the highest degree of fidelity, that the patterns of activation from start to finish that occur is you toss the ball in the air and you serve. Great. Now I stamp those neurons with a chemical tag. This has been done. Then I come back later and I activate those neurons in the same sequence. And you do the serve perfectly. And I say, great. but then they did the important control experiment. What if I activate those neurons in the opposite sequence. What if I activate all those neurons at once, like banging on the keys of the piano? It's not the same as playing the song. If I just say, what are all the keys involved in this motif of music? And instead, I just bang on them all. You get a very different sound coming out of the piano. Turns out you get the exact. Same behavior. This is problematic for the field of neuroscience. Because what it says is that this notion of the circuits firing in a particular sequence, at least for memory, it's probably not true. So now if I want to write to your nervous system, maybe I'm better off because all I have to do is activate these same neurons. And it doesn't matter what sequence I activate them in. That's kind of cool, right? Like you want someone to walk. Maybe you just need to activate the brain areas involved in walking and the body and the other brain circuits will figure it out. But here's the problem. First of all, it's very hard to activate these select sets of neurons with high spatial and temporal fidelity to begin with. But the other problem is these neurons, I didn't tell you, are all also... involved in lots of other perceptions and behaviors. So it's not like you have a circuit for serving a tennis racket that isn't also used for some aspect of thinking and remembering about something from your childhood. I mean, that's one of the beauties of the nervous system is it repurposes different neurons to participate in different circuits, depending on the perceptions and behaviors that you're experiencing. that you're engaging in. Our model of the brain works at the level of the periphery, at the deeper brain structures. But once you start getting into memory and thinking, It's complicated because we don't yet have a good model of whether or not the temporal ordering of firing of neurons matters. and every textbook says it does. And the flow from like eyes to the thalamus up to the cortex to create a visual perception, we know that direction of flow is critical. If you cut that off at the level of this deeper brain structure of the thalamus, there's no perception. We know this from human patients. But memory, thinking, brain plasticity, these things all rest on the assumption that the temporal order exists Matters. And apparently it doesn't. And that's one of the unspoken, like, within our field. And people are trying to understand that and what it means, but it's going to present a serious problem for the neurotechnologies. Again, it's going to alleviate some problems. because maybe you don't need as much spatial fidelity. But in my mind, this is a huge gap in understanding. And then the other piece is that we don't really understand what a thought is. And so if you're going to get in there and start stimulating neurons in even just motor cortex to generate movement. without the acknowledgement that these neurons are extensively interconnected with structures involved in thinking and planning, you could send the system into haywire. That said, If there's a goal in mind, like drink the glass of water, the brain generally figures it out. And that's what's incredible. I mean, there are people working on brain damaged patients that are finding that if you just go to the areas of the brain where the signals are generated to inform the action output ends of the brain, then you can do pretty well. That's how my friend Eddie Chen gets people with locked in syndrome to speak through a computer. He's not measuring from the speech production areas. It's carefully characterized for many years, decades. the signals coming out of the speech planning areas. And because the signals can't get down to the pharynx and larynx, because speech is just the pharynx and larynx controlling exhales. What he's doing is he's just transforming those signals into electrical signals so that the person can talk through a computer. So there's advantages and disadvantages to this nonspecificity. I think that's the big one. That's the really big one. What about all this excites you most and scares you most? Like, Some of this, it feels to sort of strip me of my humanity or my identity in some weird way. If I knew that you could just like play me like a piano or something like this, given high precision technology that's underneath the skull or something. The bigger question is, like, where do you actually think we're going like five, 10 years from now? What kinds of things do you think we'll have? But I'm also just curious about the cyborg line here that we're probably going to cross that gets philosophically weird. I have to acknowledge I was born and raised in Palo Alto. [40:00] Oh, and... I have friends in the technology medical sector, so sometimes I get a little bit of the blinders on about how other, you know, a lot of people are terrified of AI. And I'm told lately that I'm not scared enough. Yeah. But I acknowledge that people are concerned, afraid they don't understand it. But I remember the same discussions about computers. Which is not to say that we don't need to be careful, but People are okay now with reading from the nervous system. People are OK now with the fact that our phones are recording everything that we're saying. Remember why we're like, is it really listening? How could it really know? It's a little bit like the autonomous driving. In San Francisco, like I got in a Waymo. I didn't even think about it. I just like got in. I think measurement of the nervous system directly in waking states is, is the next mental hurdle, and I think it's going to be a healthy one to cross. Right now, we get heart rate, HRV, and a few other things, but mostly heart rate and HRV. And from that, we get told how stressed we are, how well recovered we are, how well we are sleeping. None of that is direct readout from the autonomic nervous system. I actually have a former postdoc of mine who did her undergraduate Stanford graduate work. at Harvard and that was at Caltech briefly. Her name is Melissa Olmes. She's a brilliant neuroscientist and now biotechnologist. She's developing a tool. It's like a little cuff that actually directly measures activity from the autonomic nervous system. and then gives you a readout. This is so cool. I got to try it. It gives you a readout at the end of the day of how different activities and different patterns of speech and breathing are in you related to levels of distress versus eustress. Right now, we just think all levels of increased heart rate and decreased levels of HRV are bad. But she's discovering that there are certain things in your day, like when you're stressed out doing things that you love, that actually sets you up for better sleep, more focus, The data have hinted in that direction for a while, but these devices are the sort that she's developed, and she's brought this to police departments, first responders, and eventually this will be a broadly available technology. direct measurements from the nervous system of how stressed are you? when you are stressed? Or are you actually doing great? You just are in a high arousal state that might change your perception of how stressed you are. I think accurate feedback about those things will have a real positive effect on our health metrics. Maybe you'll discover that there are a lot of things in your day that really stress you out that you weren't aware of. You'll also learn how long is the tail on your stress from a given stressful interaction, stress in air quotes. how well are you focusing will probably need to be picked up from measurements of pupil size changes. This is how it's done in animals. This is why I think confidently it will be done from eyeglasses. So you need a lot of access from a wristband and eyeglasses. or an eye mask in sleep, I think people are going to be really comfortable with that. The big one is how willing are people going to be to set up their entire brain for a vulnerability to, to non-invasive control. So we're not going to drop wires into people's brains. for a while, unless they have a clinical issue, maybe something under the skin, but you still don't have access to all the neurons. What I want for me is I want access to all the neurons where I can turn things up or turn them down. At least. by brain structure and I have control over it. There has to be some benevolence written into that, you know, and vaulted into the device, right? We can't have things that are just like, and you just suddenly you're in a rage. You would hope that the companies building these things would put safeguards in place. They have them for vagal stimulators. Why wouldn't they have them for brain stimulators? And so the vulnerability... or let's call it accessibility that I'm talking about, would be the following. My lab, when I was running a lab, and a lot of people tickle neurons, write to neurons, increase their activity a little or a little bit more or a lot these days, not through electrical stimulation, but by... putting some sort of gene in them that isn't normally expressed in them, Typically, these are genes that express channels that come from either algae or other things like non-human little things. Sounds scary, right? But the way this is done is you can do it locally by making one injection. Okay, problem, got to breach the skull. Or what a lot of people do is, is they will inject an animal with, an adeno-associated virus or some sort of virus that can stably express genes in all the types of cells that you want. You can direct them to neurons. You can direct them to immune cells. You do this by putting specific promoters. Sounds fancy, but straightforward in primates and mice and so on. You can do this in humans. They've done it in humans for other purposes, but it's more localized, like into the eyes. So here's where we're going. This is where we're going to end up. I know it sounds scary, but at some point in the not too distant future, If you're willing and it can be done safely, you get an injection of a virus. Everyone goes, oh, my God, it's going to make me sick. No, it's not going to make you sick. It's not the active virus. The virus is just a vector. It's just a carrier for some genetic cargo that puts into your brain a channel such that, If you have a hat on, wool cap or baseball cap, that it can direct stimulation of only those neurons because it will cast a nonspecific stimulus, like a cone of maybe light that can go through the skull, like long wavelength light, that seems the most likely, or ultrasound. In a non [45:00] specific way. So not illuminating or activating the whole brain, but it'll sort of send a beam of sound or light onto these neurons that if they express that tag, will be active. You can increase the intensity of the light. You could flicker the light. We'll work through all the different parameters that allow you to turn those neurons from inactive to active to more active to hyperactive and dial it down. And then with most likely through AI, you're not gonna be turning a knob. You, if you're somebody who is suffering from severe disease, motivational issues, or you have healthy levels of motivation, you want to ramp it up. We'll just increase the stimulation. Now, Does this open up into people cranking the circuit for themselves. Yeah. But we also have this thing called caffeine and we also have this thing called methamphetamine. Methamphetamine is a problem. 90% plus of the adult world population drinks caffeine every day, which increases excitability in the brain, basically. Yeah, it's like a blunt version of the same thing. Yeah, it blocks the thing that shuts down excitation, but the net effect is more excitation. So you know how much coffee you can drink at a given time of day, given how much sleep you've had. you'll decide how much alertness you want to have. I don't have a better word for it, the opportunity or vulnerability of a given set of neurons will be set by some genetic thing that will be introduced through a virus. Maybe it's an and gate. Maybe you also have to take a pill. So you can't just have the virus and the ultrasound, you need a third factor there. So It could be an AND gate. It could have any number of different safeguards in there. But these people with vagal stimulators can unplug them, too. I know it sounds scary, but it's absolutely going to happen for the people that want to do it. I think Not everyone's going to want to do it. And right now, the reading and measurement from the brain and body are freaking some people out. I put out a post on X a few weeks ago, like, oh, if the cost of whole body MRI comes down, I can imagine most people would want it. I know people who have told me they've saved patients' lives because they went and got this done and saw a tumor and we took it out. And the neurosurgery community did not like that. I thought it'd be like, oh, more business for them. But they did not like it. the public gaining more access to their own health data is scaring some sectors of medicine. Dermatologists, ophthalmologists love the AI and the more measurements better. But I think the Wall Street Journal just put something out saying like whole basically said it's bad. But you don't have to do it. I'm assuming it's safe if the cost is low enough or insurance coverage. Like, I just don't get the logic. But my friends in medicine tell me we all know within 20 months, I don't know why 20 and not 24, within 20 months, it's going to be commonplace because the cost is coming down. Just like blood testing. Everyone was resistant to blood testing and genetic testing. I'm old enough to remember the 80s. when it was like, will you get your genome mapped? It was like, what if you have the Huntington's gene? What if you're going to get, would you want to know? They had like episodes of 60 minutes about this. Now it's very simple. You want to know, you go take the test. You don't want to know, you don't take the test. No one's talking about whether or not getting your genetics done is worthwhile or not or scary or not. It's up to you. Same thing with your blood. It's so interesting to me that through all this, So many of the things that you've written about or I've seen you talk about are indirect ways to affect these same end states. And it sounds like what you think the arc is, is just. away from the blunt tools and towards direct, highly precise tools through technology over time. more and more specificity. That's what you want as an experimental neurobiologist. The early days of neurobiology were like, lesion this, or find a patient who had a spike through his head, and oh, that's what the orbital frontal cortex does? It's involved in morality? No, it's involved in contextual decision-making strategy. That's Phineas Gage. Certainly not the whole story, but it got us through HM. He's got lesions in his hippocampus. Can't remember new things. We now are at the level of tickling specific neurons that are involved in a particular memory or expression of a particular behavior using chemical genetic tags, using light, using sound. I mean, we only have so many different ways to manipulate neurons. But in the health realm, I love that you bring this up because it really bridges the two areas that we've been talking about public health discourse. and neuroscience, reading, writing to the nervous system, And it bridges the two because what are we talking about? Get morning sunlight. Why? To set your circadian rhythm, spike your cortisol. It seems like a blunt tool. But there's a circuit in place from your eye to your hypothalamus and on and on that is there for that purpose. So it's not biohacking, right? This is important. It's not like using a paperclip to fix my glasses. That's not what the paperclip was designed for. It's not a good solution. It's a temporary solution. The problem with a lot of pharmacology, like SSRIs, believe it or not, helped so many people with true clinical grade OCD. They were overprescribed for depression. And really what they do is increase serotonin, which allows for more neuroplasticity. Now everyone's down on SSRIs, understandably so, although they have their place. But everyone's really excited about psilocybin. What is psilocybin? It increases serotonin. and massively so. It unmasks lateral connections in the brain, allows you to form new associations in [50:00] very condensed timeframe. Sometimes it works great, sometimes not, but it's interesting. Same thing with MDMA. Create really massive increases in neuromodulators in a specific context if we're going to work on this thing, this pain, this issue, this trauma, this depression, and you get more plasticity than you do if you're just lathing away at the problem through talking about it day in, day out. You need the therapy, but what did you really do to create plasticity? You put serotonin through the roof with psilocybin, or in the case of MDMA, you increase serotonin and dopamine, and you have this heightened opportunity for plasticity. Okay, TMS, or what I think will soon happen is combine transcradial magnetic stimulation and pharmacology in order to work through a clinical issue that is of emotions or mind, like depression, like PTSD. So these two things are going to start to intersect. Like ADHD with stimulants, yep, ramps up activity in the prefrontal cortex, and you can focus better on things that you normally wouldn't be able to focus on. It's just a nonspecific focus enhancer because it increases activity there. You could also do sleep creatine, in particular when sleep deprived. Is it like Adderall? No But those two things great night sleep plus some caffeine you pretty close to Adderall level effects It just that the people taking Adderall aren getting great sleep And so we really talking about tools of different types different margins of safety but they are all very blunt Because all these receptors are expressed everywhere. So you get sexual side effects, weight gain or excessive weight loss with certain antidepressants or hypersexuality with certain drugs, etc. Because these receptors are all over the brain. The same thing with electrical stimulation. You just want more spatial and temporal specificity. It has to go that way. Neuroscience went that way in terms of reading and understanding the nervous system. You can talk about Dario's work, for instance. No one talks about his work in this realm, but he was recording in Mike Berry's lab from retinas, usually a salamander retina or some other creature, and they record as many neurons as you could. And then you knew what visual signals you were giving the retina. So you knew the input, you knew the responses at the level of photoreceptors, the output cells, what we call ganglion cells, the signals going off to the brain. And then the goal was get as much information about the total collection of neurons responding to an image so we can move from image to perception deep in the brain. Now you're still a few steps away from the actual perception which takes place in the brain but those and the guys that they were trained under Marcus Meister and others those were heroes of our field because dropping electrodes in an area and buzzing in like the animal goes into a rage or it starts mating or whatever That super cool But then you have to move to like, what's the nature of the switch in real life? And people are studying those things. What moves an animal from exploring mating to actual mating? What moves an animal from exploring an aggressive interaction to an actual aggressive interaction? But these are all still very low level behaviors and motivations. And we're just starting to understand those. So the cyborg thing, like, can I increase your creativity? We're not even close to that. We don't even know what circuit or IQ or something. Yeah. I mean, you could imagine stimulating more memory capacity, but you might give up something. You want to hear something super creepy. This might freak out your audience, but it's true. So a few years back, there was a guy, His first name is He. That's how you pronounce it. I can't pronounce his last name. He actually had done his postdoctoral training at Stanford, and then he moved on to his own lab in China. He announced to the world at a meeting that he had done CRISPR in twin babies. He had deleted the HIV receptor. And the story was, and I don't know if it's true or not, but that the father of these children was HIV positive and he wanted to save these kids from getting HIV AIDS There was this moment in the medical science community where no one knew how people would react Were they going to give them a Nobel prize or throw them in jail? He went to like the end experiment. We know how to do this now. The technologies exist, but he just did it. And there was this really weird sociology around this in the science community where people who had worked with him prior, who had exchanged emails with them. Their emails now were like people were digging into the emails. And it was clear that the communications from a lot of these people were sort of the, hey, be careful, but like, is it working? Is it working? But everyone was waiting. Like if he gets in trouble, I didn't know the guy, right? And I didn't know the guy for the record. I didn't know the guy. I knew of him because I knew the lab he worked in, but it's like, am I going to be a part of this? Like, was I important for this important discovery or should I dissociate myself from this? It turns out the world decide like, no, ethically, this is bad. The Chinese government came out and said that his lab would be taken away and he'll be punished in some way. I don't know how he was going to be punished. You know what? Disappeared. Yeah. Do you know what that guy's doing right now? He has a laboratory purportedly in Austin, Texas. Okay. Now I don't know if that's true or not. So my first reaction is like, how did that happen? My second reaction is, wait, of course he wasn't the only person doing this. Like you, [55:00] oh, this is happening in labs, all over the world. It's just no one's talking about it because he proved that it's dangerous to talk about. Now, I think it's very important to have the ethics and the committee set up for this kind of technology. But then you take a step back and say, where are we in the world of genetic engineering humans? And you say, okay, well... They do testing of amniotic fluid in pregnant women, babies. It's commonplace. People can refuse it, I believe. But in general, that's done. People are screening embryos for IVF. There are companies like Herocyte and Orchid. Some are more focused on trying to see if there are any diseases. Some are trying to link certain genetic patterns of expression to autism. but also to IQ, to height, et cetera. So you can see where this is going. And then, of course, there's this thing that we call partner selection. People choose, in most cases, not always, but they choose who to get sperm from or eggs from in the context of a relationship, right? Or if people are using a sperm donor. And they're choosing on the basis of, like, you look at the criteria, and the criteria are like height, where they go to school, what do they look like, you know? So the lines between these haven't been cleanly defined, but you can see there's a gradation there. Now, I don't know if he is actually in Austin, Texas, or if he's actually doing these experiments. I don't know that. I want to be very clear. But I find it. fascinating and important to understand that people are going to run with these technologies. How fast they run, where they get implemented isn't clear. What about memory? The HIV receptor supposedly... is also linked to some aspects of neural circuit formation that are linked to memory. So there was this other layer of those experiments that everyone knew in the neuroscience community, but wasn't talking about, which was, was he actually trying to both render those kids immune from HIV infection, but also make them super smart? And no one's been able to answer that question conclusively, but There is this idea that maybe people are genetically engineering people that have some gene expression or lack of gene expression, which renders them more cognitively or physically capable. And you go, oh, that sounds crazy. Yeah. And the folistatin gene has been mutated in animals. And there are people doing gene therapy for folistatin to try and build more muscle. If this was 15 years ago and you and I were having this conversation, He'd be like, guys out of his mind. Everything I just described, has happened. Why it happened, exactly what they were trying to do, unclear. Whether or not it's still ongoing, whether or not this guy's actually in Austin, Texas doing these experiments or he's someplace else or he's sitting in a prison someplace in China, I don't know. Okay, I think it'll be interesting. The internet will tell us. I guess I'm more excited than concerned, provided I'm more excited than concerned, provided I'm People are talking about it and people are thinking about it really carefully. One of the beautiful things about social media is nothing stays a secret for very long. if at all. Nothing stays a secret. There are no secrets anymore. You can have a vaulted conversation with somebody. That absolutely is true. But as soon as more than two people know about something, like you can't exist in this world without leaving a digital footprint. So sooner or later, We'll hear. It seems like the arc of all this is just control, like control of genetics, control of brain states, control of all these things through technology over time. And this is very much like a William Gibson features here, just not evenly distributed yet. And these are all the features here type things. What excites you most of all this stuff? If you had to pinpoint what? Like if I force you to go start a company or something and just focus on one thing. I spent a lot of time thinking about media and culture and just humans generally. So to me, the most exciting thing is that because of social media, we're all in the same campus. X is that campus. Instagram to some degree, but those are the campuses for the communication about these things. And so what I'm excited about. is the idea that we've got so many more people learning about things in their field and adjacent to their field. that there are a bunch of things that we can't predict that are going to happen. Like someone in the human gene therapy space who's an amazing educator, has the charisma and the ability to talk about these things. but also has the chops, really understands it, is going to move that space forward at tremendous velocity at some point. The issue is that here was a scientist just like doing his thing and he just kind of went rogue and did it. He didn't check with anybody. When I was coming up, you did experiments at night. And the goal was like, don't kill yourself. Don't burn down the lab. Don't break protocol. But like, do experiments. Now it's changed a little bit. Everyone's a little bit more paranoid. I'm excited that somebody, and they're starting. I think that people are thinking about neurotechnologies to read and write from the nervous system in awake states. My own interest right now is really understanding The waking states, we understand slow wave sleep, deep sleep, N1, N2, N3 sleep. I wrote a whole chapter in the book about sleep, its architecture, how temperature changes one or the other, heat up the bed in the morning or the room in the morning, you get more REM. We know all that stuff. We know so much about sleep and how to measure and now nudge and measure. [60:00] maybe even dial in sleep really precisely with these eye mask type tools. Changing the room temp is pretty crude compared to that. But it's also helpful. I want to understand waking brain states. I don't know how to describe the state that we're in. REM is not language about how you feel in REM. It's rapid eye movements. There's something associated with the state that we're in right now. Like I'm alert, I'm focused, I'm enjoying our discussion. What is this state? We need much better understanding of waking states and which waking states lend themselves better to which activities. Then we can start writing to the nervous system to generate those states. with a lot of temporal and spatial precision. That to me seems like the most important thing to resolve right now, and it's totally doable. I'm not the person to do it because I'm no longer running a lab. The problem with neuroscience as a field is a little bit of a problem with academia, but even the biotech sector related to neuroscience. is that people tend to work on things that are like four degrees off from what their training was. Actually, even though I've never met him, I think Dario... even though I disagree with a lot of things I hear from him, about AI. I think that it's interesting. Something came up through neurosciences in this quite different but related field. Sam, I've spoken to, is extremely smart about the neuroscience. Extremely smart, as is Sam. Zuck, as is Andreessen. And Elon obviously understands what he's. trying to achieve with Neuralink. And he's extremely savvy with respect to like, what can be done now, where we need to go. So in a lot of ways, the technologists, who used to just engineer stuff, like they're becoming neuroscientists. And I love that because neuroscience has always benefited from bringing people in from other fields. You can't give them the whole problem entirely because... As soon as they sink their hands into a brain, they realize this is not a computer. It's something a whole lot more complicated. But bioengineers are getting very, very good. And people like Eddie Chang, obviously Neuralink's doing this, Matt McDougall and the folks there. And Sam Zuck and Dario, like, that's a pretty killer lineup. And Elon, those four, those are the big four. Those are the athletes of brain exploration now. I think most people don't realize that. And I'm not sitting here like where the kids say, I'm not trying to like glaze. I think it's just awesome. You want, people coming in who have outside expertise and can leverage the best minds, like the best neuroscientists and engineers. So I'm all excitement about it, frankly. But I think we need better definition of waking brain states. I think that's the biggie. And the reason I mentioned those different silos that I've touched into is that anytime you see a field, whether or not it's action sports or it's science or it's medicine, take a big leap forward. There's always this thing that we talked about earlier where, People get into it. Some people go public facing, they kind of push the margins a little bit. Then it gets too crowded. People start getting wacky to get more signal to noise on them. And then the thing kind of dies down. And then there's this surge of really accessible, useful technologies and information. So I think 2027, 28 are going to be a low level noise in my estimation around the health space. I mean, that's what the book protocols was about. That's my thesis. We got it. I'm not going to keep talking about morning light. For God's sake, I'm starting to tire myself out, you know? So the reason I can move on from is because I know it for certain. What's in the book are the things I know for certain. So now that my thesis is turned in and people can do with it what they will, I think it has things of value, certainly information, but hopefully things of value to people. Now it's time to move into some other realm. And you can probably tell from today's conversation, I didn't realize this was going to come up. I am very, very interested in reading and writing from the awake nervous system. Eddie Chang and I talk three times a week for hours at a time. We take walks. Where is neuroscience? What's next? I'm just super excited. It's going to be great. I promise you there are safeguards we need in place. But like the best people are on these problems, the best people. And that's why it pisses me off when the traditional media tries to slow these people down. Like, OK, everybody, make your public statement about how you're not going to blow up the world. It's like, of course, like you want them to do that. Also, by the way, it doesn't pacify anybody. Of course, we need the ethics. We need the safeguards in place. But I'm nothing but excitement about this. Then again, I'm born and raised in Palo Alto. I did not exactly expect our discussion to focus so much on neurotech and waking brain states and things like this. But I honestly hope you and I can do this once a year to check in on the state of this stuff. It does seem like if you really think ahead from the current stuff in AI to where this is all going, of course, inevitably, the intimacy of technology into us is going to increase. And this is like the final frontier. So incredibly fun to talk to you about. When I do this, and I hope it's the first of many, I ask the same traditional closing question. What is the kindest thing that someone's done for you? The kindest thing. All right, this is a bit of a weird one, but I've carried the message that I got forward through basically everything. Once people hear it, if they thread it back through portions of today's conversation, I think they'll know how I've done that. So when I was in graduate school, I initially joined a different laboratory. You do rotations. And I joined a laboratory doing excellent work. that was very popular at the time that was [65:00] really on the cutting edge of molecular genetic techniques for looking at neurons. And it was all the rage at the time. But I had this absolute love obsession for the work that I had done in my second rotation. on stuff and in a field that was pretty crowded. And I found myself having joined this other lab and sneaking into the lab, I that interested me more. at night and doing experiments. And the woman who ran that lab, which is my graduate advisor, Barbara Chapman, she came to me one day and she said, "Listen, I'm totally okay with the fact that you didn't join my lab. Her lab was already very successful anyway, publishing papers in science, et cetera. And she said, but you just seem to really love this stuff. And the only thing that's really going to determine your success is how badly do you want to know the answers to the questions that you're working on? I was like, okay. So I broke up with the other lab. I joined her lab. We published eight first author papers. My career went wrong. But that was all... the consequence of working on things to try and answer questions that I was most curious about. That was an incredibly kind thing to do. It also required a lot of humility and boldness on her part because she was basically saying like, dump them, come back to me. And I had to trust that it wasn't because of some... sense of rejection in her. And it wasn't. She had extremely strong ego. Honestly, I think that's the kindest thing anyone's ever done for me. because it showed a belief in me, It showed that she understood that I didn't. have enough self-understanding and enough, frankly, enough guts to just say, I don't want to do the thing that everyone else is super excited about. I want to focus on the thing that I'm most excited about. which takes a strong mind and is much easier to do once you've had some significant success in a field. It's easy to do when you're like, oh, yeah, everyone said that that was done. That area was done. No one listens to that music. But then we had like two platinum albums. It's easy to just carry on. But when you're making the decision to go into this area that you know, most people are like, why would you go there? Like no one works on that anymore. And it turns out that the wheel of the field rotated too. And I found myself exiting my PhD lab with the skills that were best suited to what was most important in the field next. And I had to be very careful because now the landscape matched my internal curiosity. And now I had to make sure, yes, I'm going into this for the right reasons. And it was true. And likewise, with starting a podcast, I loved running a lab. I truly did love teaching at Stanford. I still do. But it was really... I just have this itch. I've got this knowledge in me that I know people can benefit from. I did not anticipate what came with being public facing. That was like not anticipated. I had no training. in that, in being media. And Like, fuck it. I want to do this. And I just pulled all the stops out and just went for it. Here we are. Can you say that line one more time? What was her phrasing of it? Success is determined by how... precisely you match your genuine curiosity to the work that you do. You do have to understand the field you're in and pay attention to it because all experiments are so hard. Building any company is hard. Doing any podcast, if it's done right, is hard. So you have to make sure that it's really true to you and not to puff him up too much. But I think one of the reasons why Elon is admired by so many people. Yeah, some people dislike him. But I think one of the reasons why deep down people either greatly admire him or envy him is not his money. It's his ability to ignore things. what people say about what he should be doing. I had a guy come on my podcast who I actually like and respect, and he was telling me all the things that, Elon and the top technologists should be doing. They should be helping people this way. And I'm like, wait, electric cars. And I do think open AI is heading into some very interesting things that relate to medicine and benevolence there and on and on. And he's like, yeah, that's what they should be doing. Not this other stuff, not Mars. That's like telling Metallica to play the Grateful Dead. Metallica or Metallica because that's what they love. That's the music they fucking love. And the Grateful Dead or the Grateful Dead because that's the music they love. You have to be yourself. The big tragedy is when people don't go in the direction of their genuine curiosity because they're trying to... quote-unquote succeed. So the kindest thing someone did for me was to gently... But very clearly, tell me, follow your curiosity, answer the question, solve the problems related to that, and you can't go wrong. And to not do that is kind of a slow death, frankly. A wonderful, awesome place to close. Andrew, thank you so much for your time. Thanks. Thank you. [70:00] may maintain positions in the securities discussed in this podcast. To learn more, visit PSUM.VC. Outro Music.