Anthropic与五角大楼之争及其Claude Mythos揭示了关于人工智能的什么
人工智能领域出现了一个令人鼓舞的趋势：一些开发者开始意识到其潜在危险性。
Anthropic公司宣布，将限制其先进模型Claude Mythos Preview的发布，因为它发现了包括主要操作系统和浏览器在内的“数千个高危漏洞”，仅向亚马逊、苹果等大型科技公司和基础设施建设者提供。
Anthropic与五角大楼的公开冲突，体现了公司对AI军事使用的限制，以防止大规模监控和完全自主武器的使用。
尽管这种行为可能存在利益驱动和反垄断风险，但在缺乏政府监管的情况下，大型AI公司的自律行动可能为AI的健康发展提供临时的最佳方案。
人工智能的发展正日益与军事规划、情报和网络行动交织，其潜在影响值得关注。
查看原文开头（英文 · 仅前 3 段）
One of the more encouraging developments in artificial intelligence is that some of the people building it have started acting like it might be dangerous. Not in the Skynet sense or the HAL 9000 sense or even the "oops, it deleted all my emails" sense, though AI might be dangerous in all of those ways too. The question is whether the latest models are dangerous to infrastructure, dangerous to privacy, dangerous to security, and dangerous to the blurry line between public and private. For years, Big Tech has been heavy on the gas, light on the brakes—and we have all benefited tremendously, even as angry debates about the downsides have raged. But with AI, at least in a few notable cases, the companies themselves have begun doing something unusual. They have started saying no.
Anthropic has announced that it would not broadly release Claude Mythos Preview, a frontier model that it says has already found "thousands of high-severity vulnerabilities," including in every major operating system and web browser. Instead, it is confining access to a consortium that includes Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, NVIDIA, Palo Alto Networks, and some other organizations that build or maintain critical software infrastructure. Anthropic says the point is defensive: to use the model to find and patch catastrophic flaws before less scrupulous actors get their hands on similar capabilities.
There is no shortage of self-interest in that decision to launch Project Glasswing. These companies would rather be seen as stewards than as Visigoths. Waiting for general release will slow the cycle of copycatting by rivals. But in an industry that spent years insisting that every new capability had to be shipped immediately because progress was inevitable, it is genuinely notable to see a major player conclude that a sufficiently powerful model should not simply be tossed into the public square.
※ 出于版权考虑，仅引用前 3 段。完整内容请阅读原文。