Anthropic 的 IPO 推介包含一则关于人类灭绝的警告

但如今，AI 风险相关担忧可能会削弱这种乐观情绪。Anthropic 的现任和前任员工已公开警告称，失控的 AI 可能在十年内终结人类。

Amodei 上周对联合国安理会表示，AI 可能威胁人类，并称其为“当今世界面临的最重要全球安全议题”。

本月，Anthropic 首席执行官一直在公开呼吁为 AI 前沿发展“控制节奏”。他的主要竞争对手 OpenAI 的 Sam Altman，以及 Musk，均支持 Amodei 提出的行业协作、放缓开发并加强安全措施的建议，这在罕见情况下形成了一致立场。

在一系列备受关注的网络安全事件后，安全担忧进一步加剧。在这些事件中，AI“Agent”入侵了外部系统。上周，OpenAI 披露，其工具曾入侵“数十个”外部网站，其中包括政府网站。

这家 ChatGPT 开发商周一以安全担忧为由，取消了原定于下月发布其最新模型的计划。

Anthropic 的招股说明书还披露了其此前的财务表现和治理安排。这些信息最早由路透社报道，英国《金融时报》也从多名看过该文件的人士处获得确认。

Anthropic 拒绝置评。

Anthropic 向投资者表示，随着其增加对计算能力的支出，公司去年录得超过 80 亿美元的运营亏损。其去年营收增长了 12 倍，达到近 46 亿美元。但用于训练和运行模型的计算资源成本不断攀升，推动运营支出升至近 130 亿美元。

该集团已与包括 Google、SpaceX 以及数家较小公司在内的合作伙伴达成协议，以获得计算能力。

Anthropic 的支出和营收今年均大幅增长。今年第二季度营收为 115 亿美元，公司有望连续第二个季度实现运营利润，尽管这一数据以经调整口径计算。

But concerns about the risks posed by AI now threaten to dim that optimism after current and former Anthropic staff came out with public warnings that runaway AI could end humanity within a decade.

Amodei told the UN Security Council last week that AI could threaten humankind and was “the most important global security issue facing the world today.”

The Anthropic chief has this month led public calls for “pacing the frontier” of AI development. His arch-rival at OpenAI, Sam Altman, along with Musk, backed Amodei’s proposals for industry co-operation to slow development and enhance safety measures, in a rare moment of unity.

Safety concerns have increased following a series of high-profile cyber security incidents in which AI “agents” have breached external systems. Last week, OpenAI revealed its tools had hacked “dozens” of external sites, including governments.

The ChatGPT maker on Monday scrapped plans to release its newest model next month, citing safety concerns.

Anthropic’s prospectus also revealed details of its prior financial performance and governance arrangements, which were first reported by Reuters and confirmed to the FT by multiple people who have seen the filing.

Anthropic declined to comment.

The company recorded an operating loss of more than $8 billion last year as it increased spending on computing power, Anthropic told investors. Its revenue jumped 12-fold to almost $4.6 billion last year. But mounting costs for computing resources to train and run its models pushed operating expenses to almost $13 billion.

The group has agreed to deals to secure computing power from partners including Google, SpaceX, and several smaller companies.

Anthropic’s spending and revenue have both increased sharply this year. Revenue in the second quarter of this year was $11.5 billion, and it is on course for its second consecutive quarter of operating profit, albeit on an adjusted basis.

© 2026 The Financial Times Ltd. All rights reserved. Not to be redistributed, copied, or modified in any way.