刚刚,诺贝尔奖得主成Anthropic新员工

marsbitPublished on 2026-06-20Last updated on 2026-06-20

Abstract

诺贝尔奖得主John Jumper宣布离开工作近9年的Google DeepMind,加入AI公司Anthropic。Jumper是AlphaFold项目的核心领导者,该AI模型彻底改变了结构生物学,曾因此获得诺贝尔化学奖。他的离职被视为谷歌的重大损失。 就在两天前,Transformer论文核心作者、Gemini联合负责人Noam Shazeer也宣布离开谷歌加入OpenAI。短短72小时内,谷歌连续流失两位顶尖AI人才。分析认为,新兴的AI实验室能提供“个人改变公司轨迹”的机遇,这对顶尖研究者颇具吸引力。 Jumper在2017年博士毕业后即加入DeepMind,并迅速领导AlphaFold团队。在他的带领下,AlphaFold系列模型成功预测了数亿个蛋白质结构,将生物学研究效率提升了上千倍。此次加入Anthropic,Jumper将助力该公司在生命科学领域的布局。Anthropic此前已收购生物科技公司Coefficient Bio,并推出专注于医疗和生命科学的AI工具。 目前,Anthropic、OpenAI和Google DeepMind三大AI实验室均已将生命科学视为下一个核心战场,竞相投入巨资,旨在用AI技术颠覆药物研发、基因组学等传统领域。Jumper的转会标志着这场顶级人才与技术竞赛进入新阶段。

诺贝尔奖得主,入职Anthropic了!

今天,AlphaFold核心领导者John Jumper官宣:离开工作近9年的Google DeepMind,加入Anthropic。

用一个AI模型改写了整个结构生物学的诺奖得主,转身走了。

Hassabis很快回应:「感谢John在过去9年里非凡的伙伴关系!我们凭借AlphaFold取得的成就改变了世界。」

合作9年、共享诺奖,这大概是科技圈最体面的一次告别了。

而就在两天前,Transformer论文的传奇共一、Gemini联合负责人Noam Shazeer刚宣布离开Google,去了OpenAI。

不到72小时,谷歌连丢两张王牌。

一个花27亿美元买回来也没留住,一个9年感情也没留住。

博士毕业6个月,直接领队AlphaFold

John Jumper在生命科学界,可以说就是「用AI改写整个学科」的代名词。

1985年,Jumper出生于阿肯色州小石城,一个普通的美国南方小城。

本科在Vanderbilt拿了数学和物理双学位,之后进入芝加哥大学一路读到博士,研究方向是理论化学。具体来说,就是用计算方法模拟蛋白质的动力学行为。

数学给他建模的直觉,物理给他对复杂系统的理解,理论化学让他比任何纯AI研究者都更懂蛋白质这个问题本身。

三个方向合在一起恰好是解蛋白质折叠问题最稀缺的知识组合。

2017年拿到博士学位后,Jumper直接就加入了DeepMind。

值得注意的是,彼时他几乎没有深度学习经验,简历上最突出的不是对神经网络的掌握,而是对蛋白质物理的理解。

但Hassabis看中的恰恰是这一点。

紧接着,他就做了一个谁都没想到的决定——让这个毕业仅6个月、连深度学习都要边干边学的年轻人,直接领导AlphaFold团队。

没有过渡期,没有「先做几年研究员攒攒资历」。

Hassabis赌的是,解蛋白质折叠这道题,懂蛋白质比懂AI更重要。而Jumper接下的,是整个计算生物学领域最大的一场豪赌。

一个人把生物学翻了1000倍

接下来几年发生的事,只能用「离谱」来形容——

2018年,AlphaFold在蛋白质结构预测竞赛CASP上首次亮相,碾压传统方法。

2020年,AlphaFold 2横空出世,困扰生物学家50年的蛋白质折叠问题,被一个AI模型直接「解了」。

2021年,Jumper带队算出了几乎所有5万多种人类蛋白质的3D结构。并在最终实现了约100万个物种、近2亿种已知蛋白质结构的生成。

在AlphaFold之前,人类花了几十年,用X射线晶体学、冷冻电镜等实验手段,一共解出了大约20万个蛋白质结构。

Jumper的团队,一次性翻了1000倍。

毫不夸张地说,生物学家过去一百年没干完的活,AlphaFold几个月就干完了。

2024年5月,AlphaFold 3发布——不只预测蛋白质了,DNA、RNA、小分子药物之间的相互作用,全能算。蛋白质-配体对接准确率76.4%,比前代方法提升1.8倍。

5个月后的斯德哥尔摩,John Jumper和Demis Hassabis一起站上了诺贝尔化学奖的领奖台。

那一年Jumper 39岁,是70年来最年轻的化学诺奖得主。

从一个连深度学习都要现学的博士毕业生,到站在斯德哥尔摩的聚光灯下,他只用了7年。

至此,Hassabis当年那个赌注的回报率,怕是人类科学史上都排得上号。

所以今天他走,Google DeepMind的痛,不只是少了一个Director那么简单。

谷歌到底怎么了?

消息炸开之后,X上的评论区直接沸腾了。

网友Chubby直呼:「这对Google来说是巨大的损失,对Anthropic来说简直是疯了!」

有网友感慨「Anthropic迎来了一位诺贝尔奖得主,人才正持续向OpenAI和Anthropic集中」,还有人直接喊出来:「先是Karpathy,现在又是AlphaFold背后的人,Anthropic这是在组AI复仇者联盟。」

Logan Kilpatrick开玩笑说期待Jumper「再拿一个诺贝尔奖」。语气是调侃,但仔细想想,还真不算夸张。

而震惊过后,所有人都在问同一个问题——谷歌到底怎么了?

Jumper没说,Anthropic没说,谷歌也没说。

或许,投资人Lior Alexander的一条评论,是目前最接近答案的——

「前沿AI实验室在兜售一种Google给不了的东西:一个人就能改变公司轨迹的感觉。」

27亿美元买回来的人,也没留住

就在Jumper官宣的两天前,Noam Shazeer宣布离开谷歌,加入OpenAI,任「架构研究负责人」。

2017年那篇现代AI的奠基之作「Attention Is All You Need」,他是核心作者之一。多头注意力是他设计的,第一个跑赢SOTA的可用实现是他一行一行敲出来的。

而谷歌为了把他从Character.AI请回来,砸了27亿美元。

回来后Shazeer出任Gemini联合负责人,成了谷歌大模型反攻的头号功臣。

结果不到两年,又走了。隔了两天,Jumper也走了。

他们既不是第一个,也不会是最后一个。

过去8年,超过20位在里程碑论文上署名的顶级研究者陆续离开DeepMind/Brain。

2025年一年,至少11名高管离职。DeepMind联合创始人Mustafa Suleyman本人,也在一轮6.5亿美元的acqui-hire中被微软挖走。

生命科学,AI三巨头的下一个战场

回到Anthropic这边。早在两个多月前,布局就已经开始了。

4月3日,Anthropic用4亿美元股票收购了生物科技公司Coefficient Bio。团队不到10人,但已经在AI驱动的抗体设计领域做出了业内顶尖的成果。

与此同时,Anthropic也在建自己的湿实验室,去年10月推出了帮助研究人员加速药物发现和生物实验设计的Claude for Life Sciences,今年1月又上线了面向医疗机构的Claude for Healthcare。

他们说,目标是把生命科学的研发周期压缩10倍。而现在,一个诺贝尔奖级别的蛋白质科学家来领这件事了。

事实上,不只是Anthropic在押注生命科学。

OpenAI今年4月发布了专门面向生物医学的推理模型GPT-Rosalind,主攻药物发现、基因组分析和蛋白质工程,已经和Amgen、Moderna、Thermo Fisher等头部药企达成合作。

OpenAI基金会更是直接表态:未来一年在生命科学方向的投入不低于10亿美元。加上刚刚挖来的Shazeer坐镇架构研究,OpenAI在这条赛道上同样来势凶猛。

而Google DeepMind这边,Hassabis旗下的Isomorphic Labs去年融了6亿美元,和礼来、诺华签下了总里程碑价值高达30亿美元的合作协议,AlphaFold的技术底座依然是行业标杆。

三家实验室,同时把筹码压向了同一个方向——用AI重写生命科学。

Jumper的选择,只是这场大棋的最新一步。

参考资料:

https://x.com/JohnJumperSci/status/2068001285173834106

编辑:摩西

本文来自微信公众号“新智元”,作者:ASI启示录

Trending Cryptos

Related Questions

Q哪位诺贝尔奖得主新近入职了Anthropic,以及他此前在哪里工作?

A诺贝尔化学奖得主、AlphaFold的核心领导者John Jumper新近入职了Anthropic。他此前在Google DeepMind工作了近9年。

Q在加入Google DeepMind前,John Jumper的学术背景是什么?

AJohn Jumper在本科期间于范德比尔特大学获得了数学和物理双学位,之后在芝加哥大学攻读理论化学博士学位,研究方向是利用计算方法模拟蛋白质的动力学行为。

QJohn Jumper加入DeepMind后,Demis Hassabis做出了一个什么关键决策?

AJohn Jumper博士毕业仅6个月后,Demis Hassabis决定让他直接领导AlphaFold团队,这是一个大胆的赌注,因为他当时深度学习经验尚浅,但Hassabis认为解决蛋白质折叠问题,理解蛋白质本身比精通AI技术更为关键。

QAlphaFold系列模型在生命科学领域取得了哪些里程碑式的成就?

AAlphaFold系列模型取得了多项里程碑式成就:AlphaFold 2在2020年基本解决了困扰生物学界50年的蛋白质折叠问题;2021年,团队预测了约100万个物种、近2亿种蛋白质的结构,将已知结构数量提升了约1000倍;2024年发布的AlphaFold 3能预测蛋白质、DNA、RNA及小分子药物间的相互作用。

Q除了Anthropic,还有哪些AI巨头在重点布局生命科学领域?

A除了Anthropic(通过收购Coefficient Bio、建立湿实验室、发布Claude for Life Sciences),OpenAI也发布了生物医学推理模型GPT-Rosalind并计划投入巨额资金,而Google DeepMind则通过旗下的Isomorphic Labs与大型药企合作。这三家公司都在积极投入资源,旨在用AI技术重塑生命科学。

Related Reads

Soaring Export Data for Memory Chips, Market Is Redefining the Valuation Anchor for Memory Stocks

Korean storage export data for the first 20 days of June shows substantial year-on-year increases in both value and price-per-kilogram for categories like DRAM, NAND, and SSDs. This signals a potential shift beyond simple demand recovery, indicating rising prices and a product mix shift towards higher-value items, possibly influenced by AI infrastructure needs. A key point is that the surge in price-per-kilogram is not simply a uniform chip price hike. It reflects a combination of actual price increases and, more importantly, an export structure increasingly dominated by high-value-density products like HBM (High-Bandwidth Memory) and advanced DRAM, which are critical for AI servers. This suggests AI-driven demand may be spilling over from just HBM into broader memory markets. SK Hynix stands to benefit directly due to its leading HBM position. For Samsung and Micron, the implication is potential for greater margin elasticity if the tightness in high-end memory spreads to enterprise SSD and NAND prices. However, the storage sector remains cyclical. Risks include supply expansion, inventory changes, and potential slowdowns in broader AI capital expenditure. Ultimately, while the strong export data supports upward revisions for storage company earnings and fuels discussion of an "AI infrastructure bottleneck premium," a definitive valuation shift from a cyclical to a structural story depends on upcoming quarterly reports. Investors need confirmation from SK Hynix, Samsung, and Micron that improvements in average selling prices, product mix, and, crucially,毛利率 are sustained over multiple quarters.

marsbit10m ago

Soaring Export Data for Memory Chips, Market Is Redefining the Valuation Anchor for Memory Stocks

marsbit10m ago

Why Does SpaceX Have Such a High Valuation Ceiling? The Answer Lies in Musk's Business Blueprint

SpaceX achieved a record-breaking IPO on June 12, 2026, with its market cap surging past $2.1 trillion. This valuation reflects its central role within Elon Musk's expansive, interconnected technological ecosystem. The article details how four core components form a synergistic closed-loop system: 1) **The "Brain" (xAI & Orbital Compute):** xAI provides AI models and massive ground/space-based supercomputing for simulation and decision-making across the system. 2) **The "Neural Logistics Core" (Starlink & Starship):** Starlink's low-latency satellite network enables global data transmission, while Starship's low-cost, reusable launch capacity aims to make large-scale space deployment economically viable. 3) **The "Physical Body" (Tesla & Optimus):** Tesla's manufacturing prowess and energy products support hardware production and power, pivoting toward mass-producing the Optimus humanoid robot for terrestrial and potential space-based labor. 4) **The "Human Interface" (Neuralink & X):** Neuralink seeks direct brain-computer communication, and the X platform provides real-time societal data. Together, these elements create three reinforcing "flywheels": manufacturing/logistics, data-driven iteration, and energy/compute/network synergy. This integrated approach promises lower costs, faster innovation cycles, and potential infrastructure-as-a-service offerings. However, it also concentrates technical, regulatory, and corporate governance risks. Ultimately, SpaceX's high valuation stems from its position as the indispensable infrastructural backbone—handling space transport, global communications, and future orbital computing—tying together Musk's entire vision for a self-reinforcing technological empire.

marsbit15m ago

Why Does SpaceX Have Such a High Valuation Ceiling? The Answer Lies in Musk's Business Blueprint

marsbit15m ago

Snap, Unprofitable for Nine Years, and a Decade-Long AR Obsession Without Return

Snap's AR Obsession: A Decade of Betting Against the Odds On June 16, Snap CEO Evan Spiegel unveiled the new AR glasses, Specs, priced at $2,195, causing the company's stock (SNAP) to plummet nearly 10%. The launch was met with intense criticism online, with investors questioning why a consistently unprofitable company would stake its future on an expensive product its core young user base can't afford. Snapchat, known for pioneering features like ephemeral Stories and popular AR lenses (like the iconic dog filter), has a history of innovation often copied by rivals like Instagram and Meta. Despite this, it has struggled to translate first-mover advantage into commercial success. Since its 2017 IPO, Snap has reported annual net losses, with a Q1 2026 loss of $89 million. Its stock is down 94% from its 2021 peak, hampered by iOS privacy changes, competition, and a young demographic less attractive to major advertisers. In this challenging context, Spiegel is doubling down on AR. He calls 2026 a "crucible moment," having recently laid off 16% of staff while reportedly investing over $3.5 billion cumulatively in its AR glasses line over nearly a decade. The new Specs represent a significant leap from the 2016 camera-focused Spectacles, offering true AR overlays, gesture control, and standalone operation. However, at $2,195, it faces tough comparisons. While more advanced than Meta's $799 Ray-Ban smart glasses, critics point to its heavier weight, short battery life, and features largely replicable by a smartphone. Facing pressure from investors to cut losses on the Specs project, Spiegel has refused, framing it as essential to Snap's long-term vision. The company finds itself in a paradoxical position: cutting costs while heavily funding a decade-long, unproven bet. Some see Specs as an awkward but necessary step in AR's evolution, akin to early mobile phones. Whether Spiegel is a visionary outlier or a gambler destined to fail remains an open question, highlighting the tension between long-term ambition and short-term market demands.

marsbit36m ago

Snap, Unprofitable for Nine Years, and a Decade-Long AR Obsession Without Return

marsbit36m ago

Annualized Revenue Exceeds $20 Billion, Kalshi Aims to Become the First Prediction Platform IPO?

Kalshi, a leading U.S. prediction markets platform, is reportedly in early, informal discussions for an Initial Public Offering (IPO). The company's annualized revenue now exceeds $2 billion, fueled by its dominance of over 90% of the domestic prediction market activity. This growth stems from a surge in trading volume—reaching a total of $52.7 billion—and an increase in fee rates, largely driven by sports event contracts like the NBA playoffs and the 2026 FIFA World Cup. Monthly active users are approximately 2 million. Kalshi recently raised $1 billion in a funding round led by Coatue Management, valuing the company at $22 billion. It has also expanded its offerings to include Bitcoin perpetual contracts and plans to launch a dedicated trading platform, Kalshi Pro. However, Kalshi's path to an IPO faces significant regulatory hurdles. The core risk involves jurisdictional conflicts, as multiple U.S. states are challenging its operations under local gambling laws. For instance, Arizona has filed criminal charges against the platform, while states like Kentucky have filed lawsuits. Kalshi and the Commodity Futures Trading Commission (CFTC) argue that its event contracts fall under exclusive federal jurisdiction as "swaps." The outcomes of these ongoing legal battles could critically impact Kalshi's core revenue and its IPO timeline. Analysts suggest that while an IPO could theoretically occur by late 2026, a more likely timeframe is late 2027 or 2028, contingent on resolving legal issues and favorable market conditions. If successful, its fundraising could significantly exceed $1 billion, given its current valuation and revenue multiple.

Foresight News54m ago

Annualized Revenue Exceeds $20 Billion, Kalshi Aims to Become the First Prediction Platform IPO?

Foresight News54m ago

Trading

Spot
Futures

Hot Articles

How to Buy CORE

Welcome to HTX.com! We've made purchasing CORE (CORE) simple and convenient. Follow our step-by-step guide to embark on your crypto journey.Step 1: Create Your HTX AccountUse your email or phone number to sign up for a free account on HTX. Experience a hassle-free registration journey and unlock all features.Get My AccountStep 2: Go to Buy Crypto and Choose Your Payment MethodCredit/Debit Card: Use your Visa or Mastercard to buy CORE (CORE) instantly.Balance: Use funds from your HTX account balance to trade seamlessly.Third Parties: We've added popular payment methods such as Google Pay and Apple Pay to enhance convenience.P2P: Trade directly with other users on HTX.Over-the-Counter (OTC): We offer tailor-made services and competitive exchange rates for traders.Step 3: Store Your CORE (CORE)After purchasing your CORE (CORE), store it in your HTX account. Alternatively, you can send it elsewhere via blockchain transfer or use it to trade other cryptocurrencies.Step 4: Trade CORE (CORE)Easily trade CORE (CORE) on HTX's spot market. Simply access your account, select your trading pair, execute your trades, and monitor in real-time. We offer a user-friendly experience for both beginners and seasoned traders.

5.5k Total ViewsPublished 2024.03.29Updated 2026.06.02

How to Buy CORE

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of CORE (CORE) are presented below.

活动图片