What Founders Can Learn About Excellence From MIT President Sally Kornbluth

创始人能从麻省理工学院校长 Sally Kornbluth 身上学到的卓越之道 在本期节目中，麻省理工学院（MIT）校长 Sally Kornbluth 分享了关于保持精英标准、应对重大危机以及管理复杂组织的领导力经验。她探讨了捍卫择优机制、绝不在追求卓越上妥协的重要性，以及在公众争议期间支持性董事会所发挥的关键作用。此外，Kornbluth 还探讨了人工智能时代高等教育的未来，强调写作即思考，且人类的创造力依然不可替代。主持人最后总结了关于危机管理、招聘具有突出特质的“尖刀型”（spiky）人才以及保持处变不惊的外在形象等核心领导力要点。 MIT 校长 Sally Kornbluth 引用“尝一口平庸的棒棒糖，就会永远平庸下去”的警言，强调卓越标准一旦退让便会导致组织不可逆的滑坡，宁缺毋滥是维持精英体系的核心。 (0:00) 麻省理工学院严格捍卫录取公平，杜绝一切关系户后门特权并坚持“需求无关”（needs-blind）政策，同时在疫情后率先恢复 SAT/ACT 标准化考试要求。 (2:43) 针对团队突破 150 人后的标准下滑风险，领导者应培养可信赖的骨干梯队下沉标准，直面绩效缺陷并践行 5:1 的真诚表扬比例，避免微观管理陷阱。 (6:29) 在国会听证会引发的舆论风暴中，MIT 董事会迅速且公开的坚定支持，成为 Kornbluth 抵御外部争议并稳住领导职位的关键决定因素。 (19:51) 谈及 AI 对高等教育的冲击，Kornbluth 坚信“写作即思考”，主张学生仍须掌握扎实的学科底座与基础编程能力，方能精准识别 AI 幻觉并进行高层次创造性思考。 (31:22)

MIT President Sally Kornbluth argues that institutions preserve enduring excellence by enforcing unyielding standards of merit, maintaining high-touch communication and direct feedback, and securing aligned governance to navigate public crises and technological disruption. Guarding Meritocracy Against Organizational Dilution (2:43) Kornbluth insists that maintaining institutional excellence requires an uncompromising commitment across every entrance point, asserting that organizations are better off leaving a role vacant than hiring someone mediocre. At MIT, this principle manifests in strictly needs-blind admissions, the early post-pandemic reinstatement of standardized testing (SAT/ACT), and the refusal of legacy or development-based admissions. When scaling past thresholds like 150 employees—where leaders can no longer personally vet every candidate—she notes it is far easier to halt an initial drop in standards than to recover from cultural decay. This commitment to objective excellence led MIT to become the first institution to reject a federal 'compact for academic excellence in higher ed.' Backed unanimously by the MIT Corporation's Executive Committee, Kornbluth refused the agreement because it sought to tie research funding to specific behaviors rather than pure scientific merit. Leadership Communication and Feedback Discipline (6:29) Transmitting organizational values through a growing hierarchy requires trusted human networks rather than broadcast announcements. Kornbluth cites an institutional reopening update during the COVID-19 pandemic that achieved only a 30% open rate, demonstrating that digital memos fail to engage audiences. Instead, she advocates developing empowered lieutenants who can fan out and deliver expectations personally without descending into micromanagement. In managing personnel, Kornbluth emphasizes upfront vetting followed by immediate, frank intervention when performance issues emerge. Delaying difficult conversations only increases personal friction as professional relationships deepen. To keep teams motivated, she endorses a genuine 5:1 ratio of praise to correction, warning that dishonest or unearned praise misleads employees into believing substandard work meets organizational expectations. Navigating High-Stakes Crises and Board Support (19:51) Reflecting on her congressional testimony regarding campus antisemitism, Kornbluth acknowledges she entered the hearing naively, expecting an academic debate rather than what proved to be political theater. Derailing at the end due to a factual, measured response rather than an immediate expression of moral outrage, she faced intense public fallout and personal attacks. She weathered the storm by staying calm externally, avoiding television circuits, and relying on biological 'receptor downregulation' to stay focused despite relentless incoming crises. Her institutional survival depended on the immediate, vocal defense mounted by MIT Board Chair Mark Orberg and the 75-member Corporation. Rather than panicking, board members reviewed the complete testimony transcript and actively distributed context to critics. Kornbluth stresses that an effective board governs without micromanaging, providing the structural safety necessary for leadership to function in high-pressure environments. Reimagining Education and Careers Alongside AI (31:22) As artificial intelligence automates routine information retrieval and data processing, Kornbluth argues higher education must focus on cultivating human judgment and creative thought. Because AI models can deliver contradictory or hallucinatory outputs, she maintains that students must retain foundational competencies—including basic coding and writing, which she equates to disciplined thinking—to evaluate machine results critically. Looking ahead, Kornbluth predicts traditional didactic lectures will be replaced by Oxbridge-style tutorial models where students work alongside AI tutors and gather in person for faculty-led debates and critique. She advises students to pursue genuine intellectual passions in discovery fields rather than acting as replaceable technical 'hands,' while emphasizing the necessity of keeping research tied to local industry clusters like the Massachusetts biotechnology corridor. While Kornbluth provides an operational framework for sustaining high standards, elite universities face continuing tensions in disciplining autonomous tenured faculty who resist centralization, alongside unresolved questions about which technical skills can be safely offloaded to AI without weakening independent critical reasoning.

MIT校长Sally Kornbluth认为，不论是掌舵顶尖学府还是初创企业，维系卓越的核心在于坚守绝对的唯才任用、拒绝妥协平庸，并在危机中依托清晰的董事会治理与沉着应对来化解风波。 捍卫精英选拔与坚守卓越标准 (0:00) Kornbluth指出，组织扩张本质上是卓越标准的延伸。引用前杜克大学同事的警示——“一旦尝了平庸的棒棒糖，就会永远平庸”，她主张宁可留出空缺，也绝不招募平庸之辈充数。在MIT，选拔不受外界人情或校友裙带影响，即使校内高管出面也无法干涉招生。该校坚持“需求盲审”（needs-blind）政策，也是疫情后首个恢复SAT/ACT标准化考试的顶尖学府。此外，当联邦政府要求11所精英机构签署将学术资助与特定行为挂钩的“学术卓越协约”时，MIT率先予以拒绝，以捍卫基于纯粹科学能力的资助原则。 在组织规模突破约150人的关口时，创始人往往无法继续亲自面试每一位员工，导致用人标准滑坡。Kornbluth强调，遏制下滑远比事后修复容易，关键在于尽早建立值得信赖的核心团队。在内部沟通上，针对全校邮件开信率一度仅有30%的现实，她认为依靠层层传递的一线关键人员进行个性化沟通，远比自上而下的群发邮件更为有效。 国会听证风波与危机治理经验 (15:59) 在就任校长首年，Kornbluth就校园反犹太主义问题出席了长达四至五小时的国会听证会。她事后坦言自己此前过于天真，未料到整场听证会是一场“政治剧场”，而最终因末尾一道预设陷阱的问题陷入被动。她反思当时的回答过于事实化且克制，类似Michael Dukakis在辩论中著名的冷淡表现，未能在当下展现出应有的情感共鸣。随之而来的舆论风暴让她承受了巨大个人心理压力，尤其作为一名犹太女性，面对被指责为反犹太主义的言论倍感痛苦。 她能渡过这次导致哈佛大学与宾夕法尼亚大学校长双双辞职的风波，核心在于MIT董事会的坚定支持。由董事会主席Mark Orberg带领的执行委员会与董事会成员公开力挺她，并主动研读完整听证记录分发给批评者，澄清被曲解的语境。Kornbluth强调，高素质董事会应专注治理而非微观管理，让领导者能够专注于工作本身，避免在危机中产生随时可能被解职的恐惧。 实验室思维与日常团队管理 (8:23) Kornbluth将自己的管理哲学归功于培养研究生的生物学实验室背景。面对科研受挫带来的低谷期，领导者无法替下属做实验，必须深入了解每个个体的内生动力。在工作反馈上，她提倡5比1的正向称赞与纠错比例，但坚决反对为了粉饰太平而给出虚假夸奖。一旦出现信任破坏或严重低效，必须立即坦率沟通，否则时间一长，伴随对下属家庭与私生活的熟悉，艰难对话的开展将变得极其痛苦。 面对顶尖机构中难以推行裁员且高度自治的明星学者，Kornbluth认为这些高绩效者通常身怀世界级科研成果，不应盲目集权。只要管理层以清晰、严谨的逻辑直接沟通，大部分顶级人才都能接受被拒绝的结果。在应对持续不断的危机邮件时，她借助生物学上的“受体下调”概念，让自己在高度刺激下适应压力，保持幽默感并依靠核心圈子的倾诉来化解负面情绪。 AI时代的教育重塑与人才培养 (31:22) 针对AI普及和“放弃大学”的思潮，Kornbluth认为大学的核心价值不在于单纯的技能传授，而在于校园社区的全面熏陶。MIT秉承“知行合一”（mens et manus）的理念，强调人类的创造性思考不可被AI取代。虽然学生可以借助AI作为增强工具，但底层逻辑依然不可缺失：例如学生必须掌握基础编程框架以识别AI的“幻觉”和错误，而坚持自主写作则是梳理思考路径的关键过程。 面向未来的教学模式，传统宣讲式大课或将逐步被淘汰，取而代之的可能是由AI担任助教的小组研讨制。对于年轻人的职业规划，Kornbluth建议选择神经科学或免疫学等需要深度创造力的前沿领域，避免充当容易被自动化取代的“纯执行手脚”。同时，她主张学校应为学生奠定广泛而坚实的知识底座，并致力于强化波士顿与剑桥当地的科技生态，促使科研成果在本土加速转化。 无论是应对政治危机还是AI对传统认知模式的冲击，Kornbluth的领导实践表明，精英机构与创新企业在规模化过程中，既需要严苛捍卫基于实力的底线标准，也依赖董事会的战术掩护与内部赋权；然而，如何在非营利大学高度分权且终身教职的环境与商业公司追求敏捷的治理逻辑之间取得更广泛的平衡，仍是有待各界持续探讨的课题。

[0:00] Sally Kornbluth's former Duke colleague: If you take a lick of the lollipop of mediocrity, you will suck forever. [2:43] MIT Admissions Staff: We do not do that at MIT. We just don't. [6:29] Sally Kornbluth: They cared, but they still didn't want to read leadership emails. [8:23] Sally Kornbluth: It doesn't cost you anything [10:40] Sally Kornbluth: What can you give them that they need so that you get what you need from it? [15:59] Sally Kornbluth: Oh my god, nobody would say that on our campus [19:51] MIT Board: We just hired her. It was fine. Go away. [19:51] MIT Board members: Did you listen to the whole thing? [31:22] Sally Kornbluth: writing is thinking [34:06] Sally Kornbluth: If you were just going to be the person who were the hands, don't do a PhD [36:32] Sally Kornbluth: You can't cut corners just because someone's going to do something that you don't think requires the full suite of educational opportunities [39:37] Host's old colleague: when you explain, you lose

[0:00] The way you maintain meritocracy and [0:02] excellence is to make sure that each [0:04] person you bring in, and for us this [0:05] means all of our faculty, all of our [0:07] staff, all of our students, we have to [0:09] consistently focus on excellence. Um, [0:12] there was a colleague of mine at Duke [0:13] who had a sign in his office that said, [0:15] "If you have a take a lick of the uh [0:18] lollipop of mediocrity, you will suck [0:20] forever." [0:21] >> Uh, that's pretty good. [0:22] >> And I I always liked that. [0:34] >> [music] [0:41] >> Hey everybody, we have Sally Cornbluth [0:44] on the pod today. She's president of MIT [0:46] and a real gem. She's had a really [0:48] interesting journey. Within a year of [0:50] being hired as president of MIT, uh she [0:54] was summoned to Congress and spoke [0:56] alongside the president of Harvard and [0:58] UPEN about all manner of things. And you [1:01] probably all remember Elise Stefanic [1:03] really asking her some tough questions [1:05] and all three presidents not really [1:08] nailing that. So we get into that and [1:10] her response to it and I think her [1:12] response was excellent. uh neither the [1:15] Harvard uh president nor the uh UPEN [1:19] president made it through uh that [1:21] crisis. She not only survived but she [1:23] thrived in it. She also had is having [1:27] more recent run-ins with the compact for [1:29] excellence in higher education that the [1:31] federal government gave her which was [1:33] pretty interesting and her reaction to [1:35] that I thought was really interesting [1:37] and showed terrific leadership. It was [1:39] only 11 or 12 institutions that got it [1:41] and I thought MIT shined on that. [1:43] She is good in a crisis and talks a [1:46] bunch about that. Sally and I talked a [1:48] lot about what I refer to as sustaining [1:51] meritocracy. She's got a giant [1:54] organization and over 150 years, you [1:57] know, they are synonymous with [1:59] excellence and synonymous with [2:00] meritocracy. And the natural state of [2:02] things is you kind of regress to the [2:04] mean and most companies do that. So we [2:06] talk about how they keep the standards [2:08] high and how they sustain it over long [2:10] periods of time. I think it's super [2:11] applicical to scaling CEOs. She's got a [2:14] great quote in there. Um, the suck in my [2:17] head. If you take a lick of the lollipop [2:20] of mediocrity, you suck forever. I think [2:24] she hits the nail on the head in that [2:25] one. And then she gets into just some [2:26] blocking and tackling of how to be a CEO [2:30] of a large organization. Some things [2:32] that I needed a reminder on and [2:34] hopefully they're useful to you. And [2:36] lastly, we talked a bunch about how to [2:38] manage how to hire a board and how to [2:39] manage a board and how to get the most [2:41] out of a board. I hope you like it. [2:43] She's a gym. My nephew Jack is applying [2:46] to business schools. He's a wonderful [2:48] kid and I really want him to go to [2:49] Sloan. I went to Sloan. I teach at [2:51] Sloan. And I've worked hard to influence [2:54] the process. I went through the front [2:55] door. I went through the back door. I [2:57] went through the basement in the attic. [2:59] Every door I go to, they're like, "We [3:02] don't care who you are. You're not going [3:03] to influence the process." 100%. [3:06] >> I love that. I love that we were the [3:08] first place to bring back SAT and and [3:11] ACTs after COVID. I love that we're [3:14] needs blind. Can you talk about that and [3:17] it's like it's easy for companies to be [3:20] meritocracies like we are. It's hard to [3:22] scale it like how do we how do you scale [3:24] that? Well, you know, first of all, I [3:26] should say I had the same experience [3:27] when I got here because [laughter] [3:29] I was someone said, "Oh, can you, you [3:31] know, speak to someone about getting my [3:33] kid into MIT and I was like, well, you [3:36] know, I could just tell them to take a [3:37] closer look." No, no, [3:39] >> I cannot. I was new here. They were [3:41] like, "We do not do that at MIT. We just [3:45] don't." So, it is it is pure merit. [3:48] >> Um, you know, it's interesting. And I [3:50] think the way to scale meritocracy, if [3:54] you want to extrapolate, is how do you [3:55] scale excellence? [3:57] >> And the thing is that when you start a [4:00] company or when you start a new team, a [4:02] new job, um you're probably the best [4:04] person out there at it, right? And [4:06] there's a temptation to just hire [4:08] people. And honestly, bringing in [4:12] >> just people, it's better to have nobody. [4:15] The way you maintain meritocracy and [4:17] excellence is to make sure that each [4:19] person you bring in, and for us this [4:21] means all of our faculty, all of our [4:22] staff, all of our students, we have to [4:25] consistently focus on excellence. Um, [4:27] there was a colleague of mine at Duke [4:29] who had a sign in his office that said, [4:31] "If you have a take a lick of the uh [4:34] lollipop of mediocrity, you will suck [4:36] forever." [4:36] >> Uh, that's pretty good. [4:38] >> And I I always liked that. And you know, [4:41] so we really pay attention that each [4:45] hire is excellent so that when people [4:48] collaborate across MIT, they know it's [4:50] not just them that are that are going to [4:52] do do good work. They know that the [4:54] people next door are also going to do [4:56] good work or the people down the hall. [4:58] So I think that is a very big part of it [5:00] is just having a mindset that [5:03] consistency [5:05] is really important. And to your point [5:07] about going from one door to another [5:09] door to another door, it has to be a [5:11] consistent message across the whole [5:13] institution because if one of those [5:15] doors had been open, [5:16] >> then you could bet people would have [5:18] flooded through that door. [5:19] >> Would have gotten through if there was a [5:20] door open. [laughter] [5:20] >> Exactly. Exactly. [5:22] >> He did get in. Obviously, I feel bad for [5:25] him because everyone assumes I got him [5:27] in. I had nothing to do with it. [5:29] >> Well, you know, the other thing is that [5:31] um merit can be defined in a number of [5:33] different ways. is I mean I think people [5:35] think that sure you got to get high [5:37] grades to get into MIT you got to get [5:39] have great test scores to get into MIT [5:42] but there's so many dimensions and I [5:44] would define it in this way we want to [5:46] bring in people who can benefit from the [5:48] fantastic environment we have and who [5:50] will contribute to that fantastic [5:52] environment and there can be a lot of [5:54] dimensions on which this is true they're [5:56] not all completely quantifiable a lot of [5:59] them comes come through in the [6:02] >> tangible is cultural or is it written [6:05] down? [6:06] >> It's not exactly written down because [6:08] it's a little bit like the if you know [6:10] it you see it and that that obviously [6:12] could be uh subject to a lot of [6:14] objectivity. So I don't mean that it's [6:17] not written down in the sense that we [6:18] don't have certain criteria in mind. I [6:21] just mean there's some flexibility to it [6:23] to take into account a wide variety of [6:26] what you or I might call excellent [6:28] fantastic people. [6:29] >> Got it. Okay. I I coach all these CEOs. [6:33] They're terrific. And like something [6:36] happens around 150 employees where the [6:38] the founder no longer interviews [6:40] everyone, [6:41] >> right? [6:41] >> And the bar drops a little bit, right? [6:45] >> And they wake up at 500 employees and [6:47] you're like, "What the hell happened to [6:48] my company?" You're that CEO. You drop [6:50] the bar and then you're trying to pick [6:53] it up at 500. What would you do? [6:55] >> What advice would you give that CEO? [6:57] >> Well, I would say this. First of all, [6:59] again coming to the fact that in your [7:01] head or in reality, you may be the best [7:03] person. You've held up that standard. It [7:06] is a lot easier to stop that slide than [7:08] to recover from it as you say. So the [7:10] most important thing when someone's [7:12] starting is to build a strong team to [7:14] whom they can delegate and who they can [7:17] trust. In other words, there's a little [7:19] bit sometimes of trust but verify at the [7:21] beginning. you want to see if someone's [7:23] really good, but after that [7:24] micromanagement is really problematic [7:27] and uh it's easy to criticize, but you [7:30] really have to let people run a little [7:32] bit. And I think to your point, you have [7:34] to build a strong structure of [7:36] lieutenants along the way when you're [7:38] getting to 150 to 500, etc. And if you [7:41] can't convey that message down through [7:44] the troops, [7:46] >> yeah, [7:46] >> you're never going to be able to get [7:48] control of it. I have to say even at a [7:50] place like MIT we do have trouble though [7:52] sometimes communicating those messages [7:54] even if we have a hierarchy. My best [7:55] example of this was actually from my [7:58] previous institution. Okay. Everyone was [7:59] bugging us when are we going to reopen? [8:01] This was after co when are we going to [8:02] reopen? When are we going to reopen and [8:05] you know we sent out an email to [8:06] everybody about when we were going to [8:07] reopen. It had a 30% opening rate. Okay. [8:10] >> So no one really cared THAT MUCH. [8:12] >> THEY CARED BUT THEY STILL DIDN'T WANT TO [8:14] read leadership emails. I really think [8:16] that hight touch series of people who [8:19] can fan out and carry your message [8:21] personally becomes really important. [8:23] >> Sally, I have trust issues and I found [8:26] almost all the CEOs I work with have [8:28] trust issues and it's a very very small [8:31] amount of employees. They deeply trust [8:33] to hire and do things crossfunctionally. [8:37] Do you have trust issues? [8:39] >> Um I would say no. Okay. in the sense [8:42] that I try to I try to sort of nip that [8:44] in the butt up front, which is again, [8:47] you know, people criticized me a little [8:49] bit at first when I came in. Well, why [8:50] aren't you turning over these positions? [8:52] Why aren't I think I needed time to [8:54] assess the capabilities of every person [8:56] >> and so when I make a decision, I really [9:00] feel good about it and have a have a gut [9:02] feeling. Now that said, if someone does [9:05] violate the trust or if someone is [9:07] really doing a bad job, I think it's it [9:10] behooves you to tell them in a very [9:12] straightforward, clear way right away [9:15] >> because once don't wait because once [9:17] time has gone on, it becomes harder and [9:19] harder and harder. You know their kids, [9:21] you know their family, you know their [9:22] dogs, you know, sorry, you're doing a [9:24] really bad job. See you. You can't do [9:25] that. Yeah. And so, um, I find that [9:29] really just being very frank when [9:31] someone's doing less than I want to be [9:34] very clear about what it is, but also [9:36] the flip side is to tell people when [9:38] they're doing well. Okay? [9:39] >> And I think that is sometimes hard to [9:41] remember. You know, it's very easy to [9:43] kind of think, "Oh, they're doing a [9:44] great job and I'm doing such a great [9:46] job. Of course, I'm leading such a great [9:47] team." I think people really, really [9:49] like to hear it when they're doing well. [9:51] And if you can do that, then you can [9:53] tell people when things aren't going [9:55] well. There's an old CEO trope that you [9:57] should give praise five times. You [10:00] should give praise to everyone sort of a [10:02] correction. Do you buy into that ratio [10:04] or is it a bunch of BS? [10:06] >> Oh, I totally buy into that ratio. I [10:08] think that's absolutely right. And you [10:09] know what? [10:10] >> I struggle with that. [10:11] >> It doesn't cost you anything. [10:12] >> No, it doesn't. [10:13] >> You know, if you're showing gratitudes [10:15] toward gratitude at the people who are [10:16] doing things, it doesn't cost you [10:17] anything. Now, if you tell people [10:20] they're good when they're not really [10:22] that good, um, just to kind of [10:24] temporize, that's where you really run [10:26] into trouble because you can't blame [10:28] them for thinking they're doing a great [10:29] job when you're just kind of saying that [10:31] to keep them happy and keep things [10:33] moving along. So, I think that ratio is [10:35] good if you really believe it, [10:37] >> but if you don't believe it, don't [10:38] propagate that message. [10:40] >> Okay. I spoke with a bunch of people who [10:42] work for you to prepare. Um, did my home [10:45] >> none of it's true. [laughter] [10:46] I they said wonderful things. Okay, [10:48] maybe [10:48] >> somebody said something that was echoed [10:50] a bunch that you're bold but not [10:52] overbearing. And when I was a CEO, I was [10:55] bold and overbearing. [laughter] [10:59] What advice would you give to someone [11:00] like me? I'm a I'm like most CEOs like [11:03] how [11:05] >> well you know I I have to say that the [11:07] training of a scientist who was training [11:10] graduate students taught me a lot. [11:12] >> Okay. [11:13] >> So being a graduate student can be [11:14] fabulous. you're doing wonderful things. [11:16] It's so exciting. It's so interesting. [11:18] >> It can also be one of my students called [11:19] it the dark tea time of the soul, you [11:22] know, you're slogging away. It's like [11:24] horrible. And but you have to keep [11:26] pushing. Uh that's mostly when your [11:28] experiments aren't working. When things [11:30] are working, it's like super exciting [11:32] and super fun. [11:33] >> And so, [11:34] >> by the way, same for CEOs and founders. [11:35] Exactly the same. So, but what I learned [11:38] was that you want things from the people [11:41] who are working, particularly when [11:42] you're starting a lab, their experiments [11:44] have to work because you're not doing [11:45] the experiments every day. And so, you [11:47] have to think about ways to motivate [11:49] them. And sometimes it feels a little [11:51] bit mchavelian. Y [11:52] >> you know, but what it really is about is [11:54] thinking what did they need? [11:56] >> What can you give them that they need so [11:58] that you get what you need from it? And [12:01] I learned a lot about individually [12:04] catering to people's motivations, trying [12:07] to figure out what makes them tick and [12:08] makes them really productive and what [12:10] makes them happy. Now, that doesn't [12:12] scale infinitely, but for your your [12:14] immediate team, [12:16] >> there really is a huge amount of [12:19] individuality in terms of what what [12:20] makes people do good work. [12:22] >> And I think really getting to know the [12:24] people individually becomes extremely [12:26] important in terms of motivating them. [12:29] And sometimes you're like, why do they [12:31] think that's fun or interesting? But you [12:33] have to indulge that, [12:34] >> especially here. [laughter] [12:37] >> Okay, that's very helpful. There's a [12:39] columnist named George Will. [12:41] >> Yes. [12:42] >> Um he described it, I think accurately, [12:44] as inexplicably [12:46] like [12:46] >> like why? [12:47] >> Yes. Ebullant. [12:49] >> And I just again I'd ask from my own [12:51] perspective like my day is like having [12:53] meetings and reading my inbox. [12:56] It's mostly bad news, you know. [12:59] >> I get really, really, really annoyed [13:01] >> and you get down. I would get down and [13:03] all CEOs get down. [13:05] >> Why are you so bulliant in the face of [13:07] I'm sure there's this [ __ ] show going on [13:09] [laughter] behind you. [13:11] >> So, yeah. Well, you know, in part, I [13:13] think I'm explicably a brilliant, which [13:15] is, you know, there's some unbelievable [13:17] things going on here. Every day I hear a [13:18] discovery that makes my jaw drop. So for [13:21] me to be able to enable the discoveries [13:23] of the people around me, I'm, you know, [13:25] essentially helping science move forward [13:27] that's way better than the science I did [13:29] myself. Yeah. [13:29] >> So that's pretty fun. [13:31] >> Yeah. [13:31] >> The other thing is you two things about [13:33] it. One is you got to find humor in some [13:36] of this. Like you really can't make some [13:37] of this stuff up. Yeah. So that kind of [13:39] keeps me, you know, a little bit up. The [13:41] other thing is this um phenomenon in [13:43] biology called receptor downregulation. [13:46] What that means is if you have a [13:48] receptor for a drug or for any molecule, [13:51] if you keep stimulating it over and over [13:53] again, eventually it kind of like goes [13:55] away and downregulates. So the last few [13:57] years have been just like a constant [14:00] incoming. And so eventually you're like, [14:03] okay, well, I could live miserable all [14:05] the time, okay? Or, [14:07] >> you know, I can rise above it and find, [14:09] you know, pleasure in all the cool [14:11] things we're doing. It doesn't mean I [14:12] don't get pissed off every day about [14:13] something because I do. But you can do [14:16] that without like becoming depressed [14:18] about it if you ignore the fact that a [14:21] lot of it is a lot of it is pretty [14:23] annoying. [14:24] >> Okay. You've been through some stuff. [14:25] You get down. [14:26] >> I get down when it's personal. Okay. [14:29] >> So, one thing about the last few years, [14:31] some of the attacks, so for instance, [14:32] after I testified in Congress on [14:34] anti-semitism, I got a lot of incoming [14:36] that was very personal. And I'm Jewish. [14:39] To be called an anti-semite over and [14:40] over again, I did not find [14:42] >> um appropriate. And so when it's [14:45] personal, when people say negative [14:47] things about me, some of the emails [14:49] people people sent to me, I was sort of [14:51] like, does wash your mouth out, folks. [14:53] >> But would you get down? [14:56] >> Transiently. [14:56] >> Okay. [14:57] >> My my I would get a little bit. Now, I [15:00] don't get down about all the other junk [15:02] that's coming into everybody in our [15:03] sector. But yes, I would get transiently [15:05] down. But the way I deal with it is to [15:08] um just like, you know, talk to people [15:10] incessantly. Like I'm a little bit of a [15:12] you know, wear it on my sleeve, get it [15:14] out of my system. It may wear down the [15:16] other people a little bit, but [15:18] >> that's definitely how I operate. [15:20] >> Okay. You brought that up. Um you just [15:23] started. You barely know your way around [15:25] the infinite quarter and you got [15:27] summoned to Congress. [15:28] >> Yes. [15:29] >> Uh what was your You did not see that [15:32] coming, I assume. I did not see that [15:34] coming. [15:34] >> The summons, what was going on inside of [15:37] your head? Who did you call? [15:39] >> Well, first of all, it wasn't a choice. [15:42] People ask me, why did you go? [15:46] >> And I went because it said if you don't [15:48] go, it's a summon and you will be [15:50] subpoenaed if you don't come. Okay, [15:52] that's the first thing. Um, [15:54] >> look, you know, the first people I [15:56] obviously talked to are the team, [15:57] including people like Alfred Iron [15:59] Ironside, you know, our VP for [16:01] communications. We talked to I talked to [16:03] a lot of other uh friends in [16:06] universities. I did not I was naive. I [16:10] think the biggest disappointment to me [16:12] of that whole thing was not that it was [16:14] hard to deal with personally. It was [16:16] that it was theater. [16:18] >> It was [16:19] >> and you know I was a political science [16:20] major in college focusing on American [16:22] political institutions and I was so [16:25] disappointed in the quality of the [16:28] dialogue and the questioning. I I know [16:30] that sounds really naive, but I thought [16:33] we were really having an interchange, [16:35] but it wasn't. It was political theater. [16:37] Yes. And I was ill well, not that I was [16:39] ill equipped. I actually think I was [16:41] prepared pretty well. And if you listen [16:43] to to the full [16:45] >> testimony, the full four hours or five [16:47] hours or whatever it is or read the [16:49] transcript, [16:50] >> it was basically fine until sort of the [16:52] gotcha moment at the end. But it [16:53] reminded me a lot. You know, I should [16:55] have done I it reminded me of the [16:57] remember the Michael Dukakus moment uh [16:59] when Yeah. Well, no. Uh when they were [17:02] asking about his wife being attacked and [17:03] it was very factual. [17:05] >> I wish. And so I sort of did that. I [17:08] maintained a very factual. What I should [17:10] have said was, "Oh my god, nobody would [17:12] say that on our campus." But you know, [17:15] hindsight is 2020. [17:16] >> Okay. So you wish you had a doover. [17:18] >> I not really because I don't want to do [17:20] it again. But but I would have handled [17:22] it differently. [17:23] >> Okay. So I assume you were prepped by [17:26] your lawyers and everybody for that and [17:27] you're sitting there listening to the [17:29] president of Harvard and the president [17:30] of UPEN. You're watching the whole thing [17:32] and then you answer the question kind of [17:34] similarly to them a little bit better. [17:36] >> No, I was first on that question. That [17:38] was the problem. [17:38] >> Oh, interesting. They'd been impressed. [17:40] Harvard was first. [17:41] >> Yeah. They'd been going Harvard, Penn [17:44] for every question until that last [17:46] question. [17:47] >> Did you know you had messed up that [17:49] second? [17:50] >> No, I didn't. When did you find out that [17:53] your 15 minutes [laughter] [17:55] was right now? [17:56] >> Well, well before I saw it on Saturday [17:57] Night Live, but I I uh you know, I was [18:01] >> Did you look at your phone or you have [18:02] to go? [18:03] >> No, because most people who talk to me [18:05] were like, "Yeah, you did pretty well." [18:07] They didn't realize that that moment [18:08] would become viral. You know, as soon as [18:10] I read the newspaper, I saw it. And you [18:13] know, then I started whenever I start [18:15] getting a lot of texts from friends from [18:18] my old days at Duke. How you doing? I [18:20] know there's something really bad out [18:22] there. When I get like 10 texts in a [18:24] row, it's like, "Oh my god." [18:25] >> But, you know, [18:26] >> so many of these I talked to and I've [18:28] interviewed go through major crisis. [18:30] This is a whopper. [18:31] >> Um, what was it like? [18:34] >> So, first of all, I think [18:37] >> you people want their leaders to be [18:39] calm. Yes. [18:40] >> And measured. And so, first of all, I [18:43] think no matter what for me, no matter [18:44] what I was thinking inside and a lot of [18:46] it was like, you know, [18:48] >> fine. I'm happy to hear that cuz you [18:49] were calm and measured, [18:50] >> but I you have to be calm and measured [18:52] on the outside. I mean, you have a human [18:54] reaction. [18:55] >> Um, but I just I felt it was so unfair. [19:00] >> And it's interesting because one of my [19:01] biggest it's sort of almost a joke in my [19:04] family. One of my biggest sort of [19:05] Achilles heel heels is the notion that I [19:08] don't want to be accused of something I [19:10] didn't do. And you know, I even watch [19:12] lots of crime shows etc. And the ones [19:15] that really like I can barely watch are [19:17] the people who like are falsely accused [19:19] of crimes. So I felt during that whole [19:22] incident that I was being accused of [19:26] having certain views or thinking certain [19:28] things and people extrapolated from that [19:30] and that to me was the crux of it. I [19:33] felt it was unfair. But you know we have [19:35] a lot to do around here. So I had to [19:38] move past it. I thought you handled the [19:40] aftermath what and I think this is a [19:42] lesson for every CEO through crisis like [19:45] you clarified your remarks and you got [19:47] to focus back on [19:48] >> you just got to focus back now I was [19:50] >> that was one good thing happened the [19:51] second good thing happened is the board [19:53] publicly came out to you and the other [19:55] two [19:56] >> y [19:56] >> stepped down shortly after those remarks [19:58] that's right [19:59] >> so talk about both those things let's [20:01] start with the remarks the clarification [20:03] who did you call how did you think about [20:05] that it was very quick like that was [20:08] excellent Excellent. That was excellent [20:09] leadership. [20:10] >> You know, I tried I tried to not create [20:14] another news cycle [20:16] >> and I think one thing we did for quite a [20:18] long time was [20:20] >> to be very quiet about it. I mean, you [20:23] know, I wasn't about to start like [20:25] showing up on TV shows and I was just [20:27] very clear with our community. I think [20:29] that's the first thing. [20:30] >> Okay. But the other thing in terms of [20:32] our um our board or corporation being [20:35] supportive that that is actually the [20:39] pivotal moment in terms of me surviving [20:41] at MIT. Yep. [20:43] >> Because they were like we just hired [20:44] her. It was fine. Go away. [20:46] >> Yeah. [20:47] >> And that really created an environment [20:50] that I felt I had a lot of support. I [20:53] could keep moving on with what we were [20:54] doing at MIT. And you know, to this day [20:56] when I do something, I don't look over [20:59] my shoulder and think, "Oh my god, are [21:00] they going to fire me over this?" [21:01] Because, you know, we have a great [21:03] relationship. We talk. Now, that doesn't [21:05] mean they won't fire me for something [21:06] someday, but at the moment, I will say I [21:08] I feel like [21:11] transparency, great interchange, and I [21:13] knew right away they were supportive. [21:15] The other thing is many many people on [21:18] our board did listen to the whole thing, [21:20] did read the whole transcript and when [21:22] people would come to them and say, "Oh [21:24] my god, how could you not fire her?" [21:25] They would say, "Did you listen to the [21:28] whole thing?" And then they would [21:30] distribute it or talk to people about [21:32] it. And so it wasn't only that we're [21:34] being supportive in in voice, they [21:37] actually were propagating the correct [21:39] message. So I think that made a huge [21:42] difference. My takeaways were you were [21:44] calm, you were in a crisis, you were [21:46] calm, [21:48] you had a good board that backed you, [21:49] which is invaluable, and you clarified [21:52] your statements in a thoughtful way. And [21:55] I think that's I think that's a best [21:57] practice right there that all CEOs can [21:59] kind of learn from. Okay. There's been [22:01] some similar I wouldn't say crisis like [22:04] actually crisis averted. Um the federal [22:07] government sent you the compact for [22:08] academic excellence in higher ed. They [22:10] sent it to 11 institutions including us. [22:14] You were the first to respond which was [22:16] interesting way before the deadline and [22:19] you basically very politely said no. And [22:22] here's why. Why were you first? [22:25] >> Well, so a couple of reasons. First of [22:27] all, when I received the compact, I [22:30] looked at it. I said, "There's no way [22:31] I'm signing this thing." [22:32] >> Okay. Right away. [22:34] >> Right away. [22:34] >> And I called uh Mark Orberg, the chair [22:37] of the corporation. I said, [22:38] >> "Mark, I'm not signing this thing." [22:40] thing and he goes, "Who's going to make [22:41] you sign this thing?" Yeah. So, right [22:42] away I knew I had the chair's support [22:45] and we went to the executive committee. [22:46] Everyone was supportive and then [22:47] eventually the whole board. So, that was [22:49] the first thing. [22:50] >> Okay. [22:50] >> Part of the reason I think we came out [22:52] quickly was rather than collectively is [22:55] because this has been true throughout [22:57] all these crises, every school has a [22:59] little bit of a different story to tell. [23:01] Yeah. [23:01] >> And the crux of this coming back to your [23:04] first question was about merit. You [23:06] know, I didn't care what the particulars [23:08] of the letter were. They talked about [23:10] giving preferential funding based on [23:12] behaviors or commitments, not on [23:15] scientific merit. And to me, the bedrock [23:18] of scientific excellence in this country [23:20] is scientific merit. Now, MIT has a [23:23] great story to tell in terms of merit. [23:26] You know, [23:26] >> awesome story, [23:27] >> awesome story to tell. No legacy [23:29] admissions, no development admissions, [23:31] under 200k uh family income, free [23:34] tuition. We reinstated the SATs before [23:38] other schools, etc. We had such a great [23:40] story to tell that in terms of merit. I [23:43] didn't really have any hesitancy. We [23:45] didn't have to think, well, how can we [23:46] couch this? The story sort of told [23:48] itself. [23:49] >> So, I wasn't trying to, you know, you [23:53] know, make a huge statement or speak for [23:55] anyone else. I just felt like we had a [23:58] good story to tell and we're going to [23:59] tell it. The other thing is I didn't [24:01] want to poke anyone in the eye. I [24:02] thought the tone that we were striving [24:04] for was thanks but no thanks. Like we'll [24:07] compete with other people. [24:08] >> The letter was extremely well written. [24:10] Who wrote that? [24:11] >> Okay. So, we have a fantastic writer, [24:15] Martha Edison. [24:15] >> Okay. She has a PR team. Sure. All CEOs [24:18] have one of these. [24:18] >> Well, someone should have a great writer [24:20] to help them. I'll tell you that. But [24:22] what Martha did was generate a long [24:24] draft on all these points and then four [24:27] or five of us iterated for multiple days [24:30] around the clock. Okay? you know, so I [24:32] would not have written everything she [24:34] wrote by scratch, but there are uh [24:36] pieces of that that I did write myself. [24:38] There are pieces of that that various [24:40] members of the team wrote and actually [24:42] we came down to a very few words at the [24:44] end that actually took us a couple extra [24:46] days and we were sitting with the [24:48] executive committee of our board [24:49] discussing the letter and we wanted to [24:51] get some of the words exactly right. [24:54] Yep. [24:54] >> And so we were arguing about different [24:56] things and I remember our EVP Glenn [24:58] Shore who is a financial person finally [25:01] like came up with a phrase and everyone [25:02] was like okay we're done here. So there [25:05] was a lot of group think thinking about [25:08] this um and I think we came to a common [25:11] message because we hang together well as [25:13] a team and we were able to be blunt. You [25:15] know Martha writes something I'm like [25:17] that is not what we want to say. [25:18] >> Okay. [25:19] >> You know she's like okay tell me what [25:20] you want to say. brought up the board a [25:21] bunch, more than I thought you would. [25:23] Uh, and I all my CEOs have to build [25:27] boards before they go public. [25:28] >> Yes. [25:29] >> Advice on building a board in managing a [25:32] board. [25:32] >> Yeah. So, I obviously didn't build this [25:34] board, but I do have input via our [25:37] governance committee. Now, in terms of [25:39] who joins the board, you know, first of [25:41] all, we have a very large board, 75 [25:43] people. So, that is not like most uh, [25:46] you know, most companies. [25:47] >> But you must have a core group. [25:48] >> We do the executive committee. Yeah, [25:51] >> look, you want a range of expertise, [25:53] >> but fundamentally you want people who [25:55] are smart, with integrity, and open to [25:58] arguments, [25:58] >> okay? [25:59] >> You know, if people come in with very [26:00] strong views and they're unwilling to [26:02] think about other viewpoints, it's going [26:04] to be incredibly frustrating. Part of [26:06] the reason I brought it up, honestly, is [26:08] because it's unusual [26:10] to have a board that you sync with so [26:13] easily. I have friends all over the [26:15] country and some of them have great [26:16] boards, some of them don't and have to [26:18] always kind of be maneuvering. What I [26:21] really like is their governance but not [26:23] management [26:24] >> and so they let me sort of do my job and [26:28] so I've been, you know, I've been pretty [26:29] lucky that way. That doesn't mean we [26:31] agree on everything. In fact, some of [26:32] the things in the compact letter I had a [26:35] lot of phone calls [26:36] >> I see [26:36] >> with people and said no, I'm not going [26:38] to do that or no, I'm not going to say [26:39] that. But it's kind of an open exchange. [26:42] I don't really feel like I have to parse [26:44] my words and I don't think they do [26:45] either. That's what I think is critical. [26:47] People that you can be honest with [26:49] that'll be honest with you and it's not [26:50] existential [26:52] like if someone is real feeling like [26:54] >> that's like they were all also calm in a [26:55] crisis in boards all deal with crisis. [26:58] >> They were calm and many of them are CE [27:01] have been CEOs or are CEOs. It's not [27:04] like you want to fill your whole board [27:06] with CEOs but you want people who have [27:09] navigated crises been on other boards. [27:11] Yeah, [27:12] >> been in high stakes situations and [27:14] understand that [27:17] staying calm is a huge piece of managing [27:19] to navigate the crisis. [27:21] >> Okay, just kind of bringing it back to [27:23] MIT, [27:24] it's a pretty decentralized place, like [27:27] really decentralized. How do you decide [27:30] when it's like, no, that thing that's on [27:32] the edge, I'm going to bring that in the [27:33] center and we're going to centralize [27:35] that. Uh, and then how do you decide [27:38] like you've got some big initiatives [27:39] you're pushing out there. How do you [27:40] decide? Yeah. So the the the [27:42] centralization thing is a lot thornier [27:44] than the what initiatives you're going [27:45] to do. Everybody wants one of their own [27:48] everything. Yeah. And when you're in a [27:50] budget crisis or you really have to it's [27:52] getting rid of duplication is incredibly [27:54] difficult. Everyone views their own [27:56] research program their own you know team [27:59] as like you know their personal babies [28:01] right you and so it's hard. So we try to [28:04] do it when it's absolutely necessary [28:07] either for efficiency or financial [28:09] reasons or because the function will [28:11] just work way better and then you kind [28:13] of navigate the noise until people [28:15] realize like yeah actually this is kind [28:17] of better in terms of the initiatives at [28:19] a place like MIT a lot of it is from the [28:21] bottom up in that we look around we see [28:24] what great work is going on and we try [28:26] to capture it. [28:27] >> I see. So what I wanted to do with these [28:29] initiatives, we have a you know a suite [28:31] of presidential initiatives. Now I [28:33] wanted to encourage interdisciplinarity [28:35] and that comes back to everybody being [28:37] at a high level. So whoever you [28:38] collaborate with, you're going to get [28:40] good partners. Um [28:41] >> can I interrupt you on that? [28:42] >> Yes. [28:43] >> Every CEO I talk to asks me, "What do I [28:46] do with the primadana? The brilliant [28:48] asshole." Sorry, [28:50] >> every place has them. [28:51] >> I We have a lot of primadas at MIT. [28:55] >> How do you deal with the primadanas? So [28:57] you know most I've look I've encountered [28:59] primadanas in my career. Most of the [29:01] primadonas that I've seen at MIT and [29:04] there's not that many compared to some [29:06] places at least they have the goods. [29:09] >> Okay. [29:09] >> The worst thing is like someone who [29:11] thinks they're great when they're not. [29:13] Like if someone wants to come to me and [29:15] say my program is really important, this [29:18] is really, you know, critical and [29:20] fantastic and like you know they've just [29:22] won a Nobel Prize. I'm not going to [29:24] argue with them. I mean, the stuff here [29:26] is really amazing. I really haven't had [29:28] someone come to defend something that I [29:30] thought was really like low quality. [29:32] >> Okay. [29:33] >> But that does make it harder to take [29:35] things away. [29:36] >> How do you take things away? Like CEOs [29:38] all the time, they're like, "We're [29:39] decentralized. Decentralized." And it's [29:40] like, "No, no, no. We're going to [29:41] start." Particularly now there's a thing [29:43] called founder mode. You heard of it. [29:45] Yeah. [29:46] >> And like you look at Jack Dorsey for [29:47] example, he just laid off 40% of [29:49] employees. Lots of people bring [29:50] centralized [29:52] >> micromanagement's in style. [29:54] Uh [29:55] >> well what's your take on all that? First [29:57] of all remember that a big part of our [29:59] workforce is tenured. [30:01] >> So [laughter] [30:01] so that's a very different you know [30:03] environment to navigate in. Um you know [30:06] what I found all along these roles like [30:08] you want to say yes a bunch of times but [30:10] you have to say no a huge amount of your [30:13] time. And I found if you engage someone [30:16] and really give them a good rationale [30:18] you've really thought through it. It's [30:19] not just like I don't like you or I [30:20] don't want to do this. Most people will [30:23] ultimately accept it. Not all people. [30:24] I'm sure there are people out there that [30:26] hate me because I said no or took [30:27] something away. But in general, I think [30:30] people understand that you have to make [30:33] these decisions. [30:34] >> Yeah. Okay. Education's at a funny [30:37] point. It's an interesting time to [30:38] interview you. My world, there's a lot [30:41] of criticism of higher ed between the [30:43] Teal Fellowships and Mark Andre saying [30:46] to not go to college and Alex Karp has [30:48] the Palunteer Fellowship for people. So, [30:51] all that's going on, which is [30:52] interesting, [30:53] >> and now we have AI. [30:55] >> What the hell does education look like [30:57] in five years? [30:58] >> That's a great question. We're talking a [31:00] lot about that. First of all, the [31:02] college experience is so much more than [31:04] some specific set of skills. Yeah. So, [31:08] people talk to me, alums talk to me [31:10] about how MIT changed their lives. It's [31:12] not because of some particular class or [31:13] some particular skill they acquired. [31:15] It's the whole environment. [31:16] >> My class actually, it was my class. [31:18] >> It was your class. [laughter] they they [31:19] say that by name you know and so that's [31:22] you know the whole college experience is [31:24] in part is important but I think [31:27] >> I agree with that like I say Sloan it's [31:29] like I got some pedigree from it I [31:31] learned a bunch and I got a great [31:32] network you can learn this stuff [31:34] anywhere now [31:35] >> that's right and the other thing is [31:37] fundamentally we're still training human [31:39] beings you know no matter what AI does [31:42] we want people who graduate from MIT to [31:44] be used you to use AI as a tool that can [31:46] augment their performance And so they [31:50] have to learn to live in this [31:51] environment. So I don't know how you [31:53] think about this, but for me like I [31:55] still am not sure how much you need to [31:57] have in your head to think creatively [31:59] versus how much you have to offload. [32:01] Yeah. Um when we think about what we [32:03] need to train our students to do, if you [32:05] ask an AI agent a question three [32:06] different ways, you get three completely [32:08] different answers. And so how do we [32:10] teach our students to think about that? [32:12] Um how much do you have to know to know [32:14] if AI is hallucinating or just plain [32:17] wrong? So there's a discussion going on [32:19] among some of our folks about, you know, [32:21] do our students really need to know [32:22] basic coding and like yes they do [32:25] because they have to know if what [32:27] they're looking at is correct and they [32:29] have to sort of have a concept in their [32:31] heads to ask the right questions [32:33] regarding coding. Same with writing. [32:35] Writing is thinking. So to just ask, you [32:39] know, AI to write something for you, [32:41] it's not the same as thinking through a [32:44] draft, thinking through a problem. Yes, [32:46] you can have AI help you and iterate. [32:48] So, I think what we're going to see is [32:51] first of all at a place like MIT really [32:53] thinking about how you use these tools [32:54] most effectively. Second thing, MIT's [32:56] always been this you know uh you know [32:59] hand in hand in mind kind of place where [33:01] building things is really important and [33:03] in one in one way AI is forging ahead [33:06] but I think physical AI uh is not really [33:09] quite there yet. You know, I I've seen [33:11] like, you know, videos of, you know, a [33:14] robot trying to like, you know, uh, [33:17] bring a a can of Coke across the room [33:19] and, you know, they turn the Coke upside [33:21] down and are trailing the Coke the Coke. [33:23] There's a lot we need to do. [33:25] >> Classrooms go away. [33:26] >> It's a good question. I don't know. I [33:28] think you could imagine a kind of [33:30] Oxbridge style tutorial system coming [33:33] out where [33:34] >> you know students work in groups with an [33:36] AI tutor that then go to class and have [33:39] a discussion with faculty. I don't know [33:41] if we're going to see the traditional [33:43] dididactic lecture [33:45] >> survive over many years when there's so [33:47] many different ways to take in that [33:48] factual material. But the human [33:50] interaction, the discussions, the [33:52] critiques from faculty, from fellow [33:54] students, that really is what produces [33:57] the, you know, college experience and [33:59] the education that you come out of a [34:01] place like MIT with. So, do I think [34:04] that's going away? I don't think so. [34:06] >> Okay. If you had a 17-year-old daughter, [34:10] >> if I did, [34:11] >> uh, what would you want her to study? [34:14] >> Oh, that's interesting. Well, I have a [34:16] 30-year-old daughter who is a surgeon. [34:19] And I think that's a pretty good thing. [34:21] I think that's a pretty good thing to [34:22] have done. [34:23] >> Parts of surgery have actually already [34:24] gone away, but I don't think the [34:25] >> whole Yeah. Yeah. So, [34:26] >> as someone who's had a lot of surgery, I [34:28] don't think [34:28] >> that seems to be a pretty good thing. [34:29] Even if, you know, she's very interested [34:31] in robotics. [34:32] >> If you were 17, you're 17, you're going [34:34] to school. What do you study? [34:36] >> Yeah. That's very interesting. Um, I [34:38] think there are a lot of areas of [34:40] science that are still, you know, if I [34:42] were picking an area of biology, for [34:43] instance, now I'm I'm a biologist. Yes. [34:45] >> I would pick neuroscience. I think the [34:47] mysteries of the brain [34:48] >> or immunology which I think is just [34:51] >> going to impact all areas of our health. [34:53] >> So [34:54] >> so nent [34:55] >> yeah there are areas that I think have [34:57] many many years of discovery ahead of us [35:00] >> but I would say only pick areas where [35:03] you really have a passion and you really [35:05] want to be a thinker. So when I was in [35:07] graduate school, you could get a PhD [35:09] sequencing a gene, a single gene, and it [35:12] was technical work, but it wasn't [35:14] actually super intellectually engaging. [35:16] I think now about companies that are, [35:19] you know, for instance, science [35:20] discovery companies where the really [35:23] great student will now have this the [35:25] equivalent of lots of hands at their [35:27] their disposal, but they still got to be [35:29] the creative one and do the thinking. If [35:31] you were just going to be the person who [35:32] were the hands, don't do a PhD [35:35] because that that [35:37] >> that could change five years from now. [35:39] Like Google outsource information. These [35:41] AI systems are kind of outsourcing [35:43] cognition. It's I don't know. I I think [35:46] it's going to change a lot. I don't [35:46] know. Like I'm thinking about my course [35:48] and like what do I do next year? I'm [35:49] trying to think about how I [35:51] >> I mean even if you're thinking about [35:53] being a physician, think about how much [35:54] AI is changing image analysis. Would you [35:57] be a radiologist now? Maybe, maybe not. [36:00] you know, uh would you be a pathologist [36:02] now? Maybe, maybe not. I don't know [36:04] because we still don't know how much [36:07] human judgment we know in in terms of [36:10] like is this cancer, is this not cancer, [36:12] but that's not the only judgment folks [36:14] are making. So, I I think we don't know [36:16] enough yet. [36:17] >> Okay. [36:17] >> And I want students to be what we really [36:20] want is students to be curious and teach [36:22] them to think critically. A 17-year-old [36:24] doesn't have to pick their career, nor [36:26] do they have to pick a career that's [36:28] going to last the rest of their lives. [36:29] Yeah, [36:30] >> because people pivot constantly now. [36:32] >> Um, do you care where like what the [36:37] vocational path of graduates do? Like [36:39] computer science students like a lot of [36:41] them go to quant shops. A lot of them go [36:43] and do their PhD. Some do startups. I [36:47] wish more would do startups. Uh, does [36:49] MIT care? [36:50] >> Uh, you know, that's interesting. So, [36:53] I've this this is not a new question to [36:55] me because even when I had a biology [36:57] lab, Yeah. About half my students went [36:59] to industry and half went to academia. [37:02] And at that time even that was [37:03] controversial among a lot of faculty who [37:06] really thought we should only be [37:07] spending our time and effort training [37:09] PhDs and we're going to go into [37:10] academia. [37:12] >> I have never particularly cared where [37:14] people go. Okay. [37:16] >> Except that wherever they're going to [37:18] go, they have to be prepared to a level [37:21] of excellence. So if you tell me you're [37:23] going to industry and you tell me you're [37:25] going to academia, I don't care because [37:27] you while you're here, [37:29] >> you have to advance the same level of [37:31] excellence in your preparation. And so [37:34] >> I think that that it's not like we [37:36] tailor our education to someone's [37:40] specific future career goal. Okay. [37:43] >> We want them to have the kind of [37:44] knowledge base and ability to navigate [37:48] the world that will enable them to do [37:49] anything they want to do. But that means [37:51] you can't cut corners just because [37:54] someone's going to do something that you [37:56] don't think requires the full suite of [37:58] educational opportunities. [38:00] >> And you talk about related to this, the [38:02] Kate initiative you have, which I love. [38:04] Why are you doing it? And [38:07] >> do you care what happens in the city [38:09] around you? Does that industry matter to [38:11] you? [38:11] >> It really does. Okay. Um, first of all, [38:15] we've all seen the impact of biotech on [38:17] the surrounding area [38:19] >> and you know as well as I do that, you [38:21] know, we've seen a lot of the AI action [38:23] go to the West Coast driving me crazy. [38:25] >> Yeah. And [laughter] so, look, [38:27] >> I think first of all, I want our [38:29] students and our faculty to be able to [38:32] bring to the world as rapidly as [38:36] possible their great discoveries. So, [38:38] that's one thing. [38:39] >> Yeah. But I also do want to see the [38:42] people who are here populate our local [38:44] ecosystem. [38:45] >> Okay. You do care about that. [38:46] >> I do. I do because first for a number of [38:49] reasons. One is just for you know base [38:50] economic reasons. I think it's important [38:52] for the country for the region rather. [38:54] But another region reason is that MIT is [38:58] part of this rich ecosystem and you know [39:01] students are going back and forth [39:03] faculty are going back and forth. I want [39:05] there to be a d a continuing dialogue [39:07] between what's going on at a place like [39:09] MIT and what's happening in the in the [39:12] real world kind of thing. And so [39:14] proximity is really important. We want [39:16] to for instance we're thinking about [39:18] whether our students should be doing [39:19] more co-ops and have more opportunities [39:22] in these companies. Things have to be [39:23] proximal for you to be able to do that. [39:25] So there's a lot of reasons to want that [39:27] to happen. And the vitality of [39:29] Massachusetts has always been about sort [39:31] of the education economy and cutting [39:33] edge tech economy. And I don't want to [39:36] see that go down the drain. [39:37] >> I don't either. Sally, uh, my first [39:40] exposure to you was a congressional [39:42] hearing. [39:42] >> Oh, [laughter] great. [39:44] >> You said I want to talk to her. [39:45] >> No. And as as an alum, I was like I was [39:48] I was my NPS was low on you. Like that [39:52] was my exposure. I was like, who is she? [39:53] I didn't know anything about her. Like [39:54] ah. And then how you handled that [39:57] crisis. How you handled the hecklers at [40:00] last year's graduation. [40:01] >> All right. I almost forgot about that. [40:03] [laughter] [40:04] >> You're you're an outstanding leader and [40:06] I'm really proud to be associate at MIT. [40:08] Thank you so much. [40:10] >> Thank you. [40:10] >> Thanks very much. [40:11] >> Hey [40:11] >> everybody, I hope you like that [40:13] interview with Sally. I think she's a [40:14] real gem. A couple of my takeaways, some [40:18] things I learned or maybe forgot from [40:21] when I was a CEO. She's handled a couple [40:23] whopper crises really well and several [40:26] of the other podcast guests have. Like [40:29] Vlad from Robin Hood's been through a [40:31] couple whoppers and some of this stuff [40:33] rhymes and almost all of you will [40:34] probably go through some major crisis. [40:37] Um some of my takeaways are even though [40:40] you're in an absolute tornado in size [40:44] you're kind of boiling up, you're losing [40:46] it somehow magically you have to show a [40:50] calm exterior. It is not easy. It's [40:52] unnatural. But in a crisis, the CEO [40:54] needs to do that. If you have made a [40:57] mistake, like she made a mistake in [40:59] front of Congress. [41:01] There's an expression old colleague of [41:02] mine used to use. When you explain, you [41:04] lose. You like you're spinning it, [41:06] you're explaining it, you're explaining [41:06] it, and people just tune out. Made a [41:08] mistake, just admit it. Clarify your [41:11] remarks if you need to clarify them. And [41:13] do it quickly. Don't sit on it. Another [41:15] hack she has that we had at HubSpot, [41:17] have an excellent writer on board [41:19] because you have to verbally communicate [41:21] stuff, but so much is done in writing. [41:23] And in her case, in HubSpot's case, many [41:25] times a very very well-written post [41:29] internally and externally is gold. And [41:32] we had a couple of couple of moments [41:34] where that really mattered. And I took a [41:35] lot of writing classes. My co-founder [41:37] was a good writer and we leaned on that [41:39] heavily. So if you're not a great [41:40] writer, it doesn't even have to be your [41:42] conserson. somebody on your team should [41:44] be an excellent writer. And the last [41:45] piece is just she has a very good board [41:48] and her board's on board with her. [41:50] That's underrated. You want your board [41:51] to be behind you. We talked about [41:53] meritocracy and sustaining excellence. [41:55] My co-founder and I and Huba graduated [41:57] from MIT and we tried to keep sort of [41:59] the culture of MIT into HubSpot. And the [42:03] way I would like to describe this is my [42:05] mother, my every time I visit my mother, [42:09] uh all her friends kids seem like [42:10] they're always looking for jobs. So, I [42:12] get a stack of three or four resumes [42:14] every time I visit her and of course [42:16] they refer them back to HubSpot. HubSpot [42:18] didn't care that I referred them. In [42:20] fact, it might have even been a black [42:21] mark that I referred them. And so, much [42:24] more often than not, they didn't get the [42:26] job. That really irritated my mother, [42:28] but I think was good for HubSpot. Um, [42:30] another thing about this I think she [42:33] didn't talk about this but I think is [42:34] like a lot of founders ask me like who [42:37] should I be be interviewing and when at [42:39] what scale should I stop interviewing? [42:41] I'll give you my two rules of thumb. You [42:43] should all obviously interview the [42:45] people who are directly reporting to you [42:47] in the circle in that next level out. [42:50] Um, I also think like as you're scaling [42:53] fast, you can interview everybody up to [42:55] 100 employees. Not the first interview [42:57] but the final interview. And a couple of [42:59] my CEOs are doing that around 100 [43:01] employees. It starts to really break [43:03] down there like they have 10minute [43:05] interviews on their calendar, but I [43:07] think you should have veto rights up [43:08] until about that point on everyone [43:10] because you're kind of setting a [43:11] foundation. Two other hacks that we had [43:14] at HubSpot that really worked on the [43:15] meritocracy side. Um, one is, you know, [43:19] she talked about this, but like they [43:20] didn't necessarily write down this idea [43:23] of of sort of excellence in meritocracy. [43:25] It was cultural. We had something [43:27] similar. We had our five values and in [43:30] the form you would fill out after you [43:31] interviewed someone all five values [43:34] would be on there and you'd scale you [43:36] rank them on a scale of 1 to four. That [43:38] kind of built the values into the system [43:41] and we go back later and compare how are [43:42] we doing on those values and then for [43:45] every person we interview would score [43:46] them. And my only advice on interviewing [43:49] particularly execs is it's very tempting [43:51] and we did this a lot to hire the person [43:54] with [music] the least amount of [43:55] weaknesses that everyone liked as [43:58] opposed to hiring the person who maybe [43:59] has some weaknesses and some people were [44:02] negative on but some people really [44:04] thought were great. You know, we when we [44:06] started hiring spiky people, in fact, [44:08] Yamin, the current CEO of HubSpot, [44:09] started doing [music] this, it really [44:11] paid off for us. Anyway, those are some [44:13] of my takeaways. I hope you like the [44:14] pod. I think she's incredible. I'm a [44:17] proud uh graduate and uh I'll see you on [44:20] the next pod.