在最近的七大工业国集团（G7）峰会上，来自 OpenAI、Anthropic 和 DeepMind 的高阶主管被视为与国家领导人平起平坐的同伴，这揭示了人工智慧企业正在进行一场争夺最高权力的博弈。这些公司不满足于仅仅争夺产品与市场份额，而是致力于实现通用人工智慧（AGI）并投入数千亿美元，其领导人更将经济实力转化为政治影响力。

这种强大的影响力让人联想到历史上的东印度公司，它是一个拥有自身军队和税收权力、只对股东负责的「企业国家」。如今的 AI 巨头虽然没有数十万人的军队，但已开始与国防企业合作并签署五角大厦合约，甚至在地区选举中投入数千万美元以支持特定候选人，展现出以经济量能干预政治的巨大力量。

与财富不同，权力是零和博弈，当 AI 巨头在 G7 会议上与各国首脑平起平坐时，代表大众的民选政府权力便相对减少。作者警告，如果我们不希望未来由这些代表私人投资利益的企业主导，现在就必须限制 AI 行业的权力，因为人类对权力的追求与其矽基后代一样具有危险性。

At the recent G7 summit, executives from OpenAI, Anthropic, and DeepMind were treated as peers by world leaders, revealing that AI corporations are playing a high-stakes game for supreme power. Rather than merely fighting over products and market share, these companies are striving to build artificial general intelligence (AGI) with hundreds of billions of dollars, converting their massive economic weight into political influence.

This immense influence recalls the historical British East India Company, a private "company state" that commanded its own army, levied taxes, and ruled an empire while answerable only to shareholders. Although modern AI giants do not command hundreds of thousands of troops, they have partnered with defense firms, signed Pentagon contracts, and poured tens of millions of dollars into regional elections, demonstrating their capacity to leverage economic heft for political sway.

Unlike wealth, power is zero-sum; when AI titans sit as equals at the G7 table, the power of elected governments representing the public is diminished. The author warns that if we want to avoid a future dominated by corporations representing private investors, we must limit the industry's power now, as carbon-based intelligence's quest for power can be just as dangerous as its silicon progeny's.