The Download: AI’s trillion-dollar gamble and OpenAI’s biology data bid

《下载》：AI 的万亿美元赌注与 OpenAI 的生物数据竞标

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. 这是今日份的《下载》，我们平日更新的简报，为您提供每日科技世界的动态。

What’s at stake in AI’s trillion-dollar gamble

AI 万亿美元赌注背后的利害关系

When Jessica Wachter, a finance professor at the University of Pennsylvania, wanted to assess AI’s impact on the economy over the next few years, she faced a long list of uncertainties. So she started with a “remarkable fact” that is not in question: a handful of so-called hyperscalers are investing huge amounts of money to build AI data centers. Instead of trying to predict how widely deployed AI models will be, Wachter asked how fast the hyperscalers’ earnings will need to grow to justify their spending through 2027, when expenditures are expected to reach nearly $1.1 trillion. The results are eye-opening. AI companies will need to achieve an extraordinary increase in productivity just to break even by 2030. Take a closer look at what it will take for the AI buildout to pay off. —David Rotman 当宾夕法尼亚大学金融学教授杰西卡·瓦赫特（Jessica Wachter）想要评估 AI 在未来几年对经济的影响时，她面临着一长串的不确定性。因此，她从一个毋庸置疑的“显著事实”入手：少数几家所谓的超大规模云服务商（hyperscalers）正在投入巨资建设 AI 数据中心。瓦赫特没有试图预测 AI 模型会被部署到何种程度，而是提出了一个问题：这些超大规模云服务商的收益需要以多快的速度增长，才能证明其到 2027 年的支出是合理的（届时支出预计将达到近 1.1 万亿美元）。结果令人震惊：AI 公司需要实现生产力的非凡提升，才能在 2030 年实现收支平衡。深入了解 AI 建设要实现盈利需要付出什么。——大卫·罗特曼（David Rotman）

AI models need more data about biology, and OpenAI is paying to create it

AI 模型需要更多生物学数据，OpenAI 正出资创造这些数据

AI needs much more information to make important breakthroughs in curing disease. So last year Ruxandra Teslo, a policy analyst, posted an idea for supercharging medical AI systems: use data from failed biotech companies. By bidding at bankruptcy proceedings, she argued, it might be possible to obtain detailed regulatory filings, manufacturing strategies and safety data, creating what she called “biotech’s lost archive.” The OpenAI Foundation, the nonprofit parent of OpenAI, announced this week that it will fund her idea, paying to create “high-quality scientific datasets.” Learn more about their new effort. —Antonio Regalado AI 需要更多信息才能在治愈疾病方面取得重大突破。因此，政策分析师鲁克桑德拉·特斯洛（Ruxandra Teslo）去年提出了一个为医疗 AI 系统“增压”的想法：利用倒闭生物技术公司的数据。她认为，通过在破产程序中竞标，或许可以获得详细的监管文件、制造策略和安全数据，从而创建她所称的“生物技术失落档案”。OpenAI 的非营利母公司 OpenAI 基金会本周宣布，将资助她的这一想法，出资创建“高质量的科学数据集”。了解更多关于他们这项新工作的信息。——安东尼奥·雷加拉多（Antonio Regalado）

Our Roundables on AI’s extinction threat is now available on demand

我们关于 AI 灭绝威胁的圆桌讨论现已可点播

As frontier models become more capable, warnings about AI extinction have become widespread in Silicon Valley. But are the threats really as dangerous as they’re presented? In the latest MIT Technology Review Roundtable, executive editor Niall Firth, senior AI editor Will Douglas Heaven and AI reporter Grace Huckins took a closer look at the arguments behind those warnings. They discussed what AI extinction could actually mean, how seriously we should take the risks and what, if anything, can be done to reduce them. Subscribers can now watch an exclusive recording of the discussion. Want to join the next conversation? Subscribe to MIT Technology Review for exclusive access to all our future Roundtables, and recordings of previous ones. MIT Technology Review 随着前沿模型能力越来越强，关于 AI 导致人类灭绝的警告在硅谷变得普遍。但这些威胁真的像描述的那样危险吗？在最新的《麻省理工科技评论》圆桌会议上，执行编辑尼尔·弗斯（Niall Firth）、资深 AI 编辑威尔·道格拉斯·赫文（Will Douglas Heaven）和 AI 记者格蕾丝·哈金斯（Grace Huckins）深入探讨了这些警告背后的论点。他们讨论了 AI 灭绝实际上可能意味着什么，我们应该多认真地对待这些风险，以及如果可以的话，能做些什么来降低这些风险。订阅用户现在可以观看此次讨论的独家录像。想参加下一次对话吗？订阅《麻省理工科技评论》，即可独家访问我们未来所有的圆桌会议以及过往会议的录像。——《麻省理工科技评论》

Narrated: a startup claims it’s found a drug to make your blood young

播客：一家初创公司声称发现了一种让血液“返老还童”的药物

Generation Lab says its new rejuvenation treatment “blocks the systemic spread of aging in the bloodstream, reawakens the body’s own repair mechanism, and restores health and youth to multiple tissues.” The approach is based on research by the company’s scientific founder, Irina Conboy. She found that joining the circulatory systems of old and young mice improved the old animals’ ability to heal from injury. Conboy now says she has found a combination of two existing drugs that can produce youthful effects without the need for any bodily fluid exchange. But there’s a snag: Generation Lab won’t reveal what the drugs are. This is our latest story to become an MIT Technology Review Narrated podcast, which we publish each week on Spotify and Apple Podcasts. Just navigate to MIT Technology Review Narrated on either platform, and follow us to get all our new content as it’s released. Generation Lab 表示，其新的回春疗法“阻断了衰老在血液系统中的全身性扩散，重新唤醒了人体自身的修复机制，并恢复了多种组织的健康与活力。”该方法基于公司科学创始人伊琳娜·康博伊（Irina Conboy）的研究。她发现，将年老和年轻小鼠的循环系统连接起来，可以提高年老动物从损伤中愈合的能力。康博伊现在表示，她已经找到了一种由两种现有药物组成的组合，无需进行任何体液交换即可产生年轻化的效果。但有一个问题：Generation Lab 不愿透露这些药物是什么。这是我们最新一期《麻省理工科技评论》播客（Narrated）的故事，我们每周在 Spotify 和 Apple Podcasts 上发布。只需在任一平台搜索“MIT Technology Review Narrated”并关注我们，即可在第一时间获取我们的最新内容。

The must-reads

必读内容

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 我已梳理了互联网，为您找出今天关于科技最有趣/重要/可怕/迷人的故事。

Nvidia and Meta CEOs have rejected calls for a coordinated AI slowdown Jensen Huang and Mark Zuckerberg pushed back on the proposals. 英伟达和 Meta 的 CEO 拒绝了协调放缓 AI 发展的呼吁 黄仁勋和马克·扎克伯格对这些提议进行了反驳。
The FTC chair has warned against giving AI companies antitrust waivers His comments follow Anthropic’s call for a safety exemption. 美国联邦贸易委员会（FTC）主席警告不要给予 AI 公司反垄断豁免权 他的评论是在 Anthropic 呼吁获得安全豁免之后发表的。
A Chinese hacking firm has used AI to analyze stolen secret Its tools turn hacked government data into intelligence reports. 一家中国黑客公司利用 AI 分析窃取的机密 其工具将窃取的政府数据转化为情报报告。
“Smart” nanoparticles delivered mRNA to tumors in a cancer study The treatment reprogrammed cells to attack tumors in mice. 一项癌症研究中，“智能”纳米颗粒将 mRNA 输送至肿瘤 该疗法重新编程了细胞以攻击小鼠体内的肿瘤。
A digital fly brain is taking on an extraordinary range of tasks online People have taught it to drive, trade bitcoin, and play Doom. 一个数字果蝇大脑正在网上承担各种非凡的任务 人们已经教会它开车、交易比特币和玩《毁灭战士》（Doom）。
The Senate has blocked new crypto rules amid a fight over Trump It demanded tougher ethics rules around Trump’s crypto holdings. 在围绕特朗普的争斗中，参议院阻止了新的加密货币规则 它要求针对特朗普的加密货币持有量制定更严格的道德准则。
Chinese firms allegedly used Binance to launder Iranian oil money Prosecutors say they laundered more than $1.5 billion. 据称中国公司利用币安（Binance）洗白伊朗石油资金 检察官称他们洗钱金额超过 15 亿美元。
An AI agent platform is reinventing spam to flood inboxes worldwide iLand says its agents have sent 1.6 million messages. 一个 AI 代理平台正在重塑垃圾邮件，以淹没全球收件箱 iLand 表示其代理已发送了 160 万条消息。
ByteDance founder Zhang Yiming has become Asia’s richest person His fortune has risen above $105 billion as AI booms. 字节跳动创始人张一鸣成为亚洲首富 随着 AI 的繁荣，他的财富已超过 1050 亿美元。
A fully AI-generated sitcom has arrived—and it’s terrible A reviewer called the characters “dead-eyed waxworks.” 一部完全由 AI 生成的情景喜剧问世了——而且非常糟糕 一位评论家称其中的角色是“眼神呆滞的蜡像”。

Quote of the day

今日名言

“The only institution that Americans might trust less than Washington these days is Silicon Valley.” —Patrick Hillman, the chief operating officer of Logical Intelligence, a San Francisco–based startup chaired by Yann LeCun, says in a statement that people have little faith in tech companies to act in the public interest. “如今，美国人可能比不信任华盛顿更不信任的机构，就是硅谷。” ——帕特里克·希尔曼（Patrick Hillman），由杨立昆（Yann LeCun）担任主席的旧金山初创公司 Logical Intelligence 的首席运营官，他在一份声明中表示，人们对科技公司能够为了公共利益行事几乎没有信心。

One more thing

还有一件事

Why Trump’s “golden dome” missile defense idea is… 为什么特朗普的“金穹”导弹防御构想是……