Anthropic says its AI models could manipulate, blackmail and harm humans安特罗皮克公司称，其人工智能模型可能会操纵、勒索和伤害人类。
Anthropic disclosed that its advanced artificial intelligence technology poses "existential risks to humanity" in an initial public offering document.
A smartphone displaying the Anthropic logo is shown in the foreground with a blurred Claude Mythos themed background in Tunis, Tunisia, on July 11, 2026. — Imen Ben Youssef/Hans Lucas/AFP/Getty Images/TNS
Anthropic disclosed that its advanced artificial intelligence technology poses "existential risks to humanity" in an initial public offering document.
As the company prepares for the largest listing in history at US$2 trillion (RM8.1 trillion), its IPO prospectus has been circulated among investors and viewed by Reuters and other news outlets.
Thank you for your report!
Ask AI: ‘Who should I vote for?’
Analysis-AI's race to transform the world before the money runs out
2026年7月11日，在突尼斯首都突尼斯市，一部智能手机出现在前景，屏幕上显示着Anthropic的标志，背景是模糊的克劳德·米索斯主题图案。——伊门·本·优素福/汉斯·卢卡斯/法新社/盖蒂图片社/TNS
Anthropic 在首次公开募股文件中披露，其先进的人工智能技术对人类构成“生存风险”。
该公司正准备进行史上规模最大的IPO，融资额达2万亿美元（8.1万亿令吉），其IPO招股说明书已在投资者中传阅，并被路透社和其他新闻媒体查阅。
感谢您的报告！
问人工智能：“我应该投票给谁？”
分析人工智能在资金耗尽前竞相改变世界