Anthropic在过去五年中不断警告世人先进人工智慧可能带来的巨大危害，但同时它也成为推动AI技术发展的最强大力量之一。该公司深信AI的到来不可避免，并认为自己作为「好人」留在AI竞争的前沿，才能负责任地引导这项变革性技术的安全过渡。（关键数字： five years）

该公司于2021年由一群因对OpenAI（特别是执行长Sam Altman）安全理念失去信心而离开的前员工创立。虽然Anthropic以公共利益机构自居，强调技术与商业力量是实现人类长期福祉的手段，但其内部的意识形态单一性以及对自身使命的强烈信念也引来了缺乏外部监督和盲点的批评。

随著公司规模扩大，Anthropic逐渐面临与其负责任形象相冲突的争议，例如与Palantir合作向美国国防及情报部门提供AI服务，甚至其模型被用于军事目标定位。尽管执行长Dario Amodei承认AI企业掌握过多权力存在风险，但他提出的解决方案仍局限于自我承诺和外部观察，并未根本解决权力集中的问题。

Anthropic has spent the last five years warning the world about the grave dangers of advanced artificial intelligence, yet it simultaneously acts as one of the most powerful forces driving AI capabilities forward. The company operates on the belief that the arrival of AI is inevitable, and that the world will be better off if they, as the "good guys," remain at the frontier to guide its safe transition.

Founded in 2021 by former OpenAI employees who lost faith in their former employer's safety approach, Anthropic structures itself as a public benefit corporation to prioritize humanity's long-term benefits. However, its intense belief in its mission and the homogeneity of thought among its staff have drawn criticism regarding potential blind spots and a lack of accountability.

As the company grows, Anthropic faces controversies that challenge its responsible image, such as partnering with Palantir to serve US intelligence agencies and having its models used for military targeting. Although CEO Dario Amodei publicly acknowledges the risks of concentrated power in AI labs, his proposed remedies focus on self-commitment and observation rather than redistributing that power.