Bill Gates Warns Unchecked AI Could Enable Catastrophic Harm

Bill Gates has warned that sufficiently powerful AI, in the wrong hands, could cause harm far beyond today's cyber threats, and he is pressing governments to regulate more assertively.

What Gates said

In an interview with NBC News in late September 2026, the Microsoft co-founder said AI is "certainly powerful enough" to cause events leading to hundreds of millions or even a billion deaths. In his view, there has never been a "weapon" as powerful as bad actors combined with the latest AI tools, with scenarios ranging from cybercrime to bioterrorism and war.

Asked whether companies can regulate themselves, he said no one thinks self-regulation is enough. Governments need to pass laws, enforce them, and bring political leaders into the discussion on safeguards. Beyond security, he also pointed to the impact on jobs, on children's development, and the possibility that models could start acting against human interests.

Why it matters

The comments come as tech leaders are increasingly split. One side wants to accelerate AI capability; the other wants to slow down and build safety guardrails first. When someone as influential as Bill Gates sides with tighter oversight, AI rules are likely to tighten over the next few years, including in the markets where your company operates.

What companies should do

For most companies, the lesson is not about frontier AI. It is about applying the same caution at their own scale:

• List the AI tools your staff use and the kind of data that goes into each one.

• Define which data must never go into AI, such as personnel files, payroll and client contracts.

• Name the person accountable when something goes wrong with AI.

• Track AI regulation in Vietnam and Korea so rule changes do not catch you off guard.

Source: VnExpress, NBC News