📄 中文摘要
2026年AI产业面临的最大挑战不再是模型能力，而是如何掌控技术及其带来的代价与责任。Anthropic因拒绝美军要求同意进行无授权监控和致命自主武器的合作而被起诉，成为首个因道德底线与本国政府对抗的大型AI公司。美方通过采购排除的方式施压AI公司，若成功将形成先例，迫使公司在道德与商业之间做出选择。此事件引发OpenAI与Anthropic的合约谈判破裂，导致“QuitGPT”运动，用户大量卸载ChatGPT，OpenAI硬件主管辞职，强调底线问题需更慎重讨论。
2026年第11周 AI 深度周报：政府、电力、安全的三重考验
出处: 2026年第11週 AI 深度週報：政府、電力、安全的三重考驗
发布: 2026年3月12日
📄 English Summary
2026年第11週 AI 深度週報：政府、電力、安全的三重考驗
The biggest challenge facing the AI industry in 2026 is no longer about model capabilities but rather who controls the technology, the costs involved, and how responsibility is assigned. Anthropic was sued by the U.S. Department of Defense for refusing to agree to terms that included unauthorized surveillance and lethal autonomous weapons. This marks the first time a major AI company has opposed its own government over ethical boundaries. The U.S. government is using procurement exclusion as a pressure tactic, which could set a precedent forcing companies to choose between ethics and business. Following the breakdown of negotiations, OpenAI signed a contract that Anthropic rejected, sparking the 'QuitGPT' movement, leading to a surge in ChatGPT uninstalls and the resignation of OpenAI's hardware chief, who stated that these boundaries require more serious discussion.