Giorgos Tsetis calls for personal liability for AI executives

Tsetis writes from experience: for almost a decade he led a company whose products people swallowed, and he says personal legal responsibility kept him awake at night—and that was a good thing. That accountability, he argues, was essential to building the consumer trust that made his business a success. Now an investor in frontier AI companies such as Anthropic and xAI, he sees a gap: executives are shielded behind their companies, and corporate penalties are often a fraction of turnover and valuations.

He acknowledges that rules are moving. States across the US are introducing their own AI laws, and Congress is weighing legislation that would impose safety requirements on frontier AI companies. The White House accord and published safety frameworks are welcome, he writes, but they are voluntary, and voluntary isn't enough.

His concern is that AI is unlike other industries: it is embedded in power grids, hospitals and classrooms, and a single failure could ripple across a city or country. He does not argue for punishment on every failure, only when leaders know their systems pose serious risks and ship them anyway, underfund safety teams or hide problems from regulators and the public. The details of thresholds and audits, he says, are for legislators and ethicists to decide.