America’s biggest artificial intelligence companies are secretly preparing for a wave of public outrage and government backlash following a devastating cyberattack. Technology executives are examining scenarios in which a major disruption to banking services, internet access or power grids could trigger widespread anger and demands for action from authorities, Axios reports.

OpenAI CEO Sam Altman and Anthropic chief Dario Amodei are among the industry leaders concerned about such a crisis. According to the publication, top executives believe a major cyberattack capable of crippling critical infrastructure is only a matter of time. Their fears extend beyond the immediate damage, as a disaster of this scale could severely undermine public confidence in artificial intelligence.

The backlash could target both the companies developing AI technology and US President Donald Trump. Industry leaders fear that Americans would blame the administration for its reluctance to impose strict government oversight on the rapidly expanding sector.

OpenAI has acknowledged conducting emergency preparedness exercises to assess potential threats and develop response strategies. A company spokesperson explained that employees work through different crisis scenarios to improve their ability to handle unexpected developments. However, the company stressed that these exercises should not be interpreted as a prediction that every situation under consideration will inevitably occur.

Another major concern is how Congress might respond to a catastrophic AI-related crisis. Axios reports that Democratic lawmakers could push for sweeping restrictions on artificial intelligence, potentially going as far as banning the development of superintelligence altogether. Meanwhile, the proposal to introduce mandatory emergency shutdown mechanisms for advanced AI systems is gaining support among both Republicans and Democrats, as well as some business leaders.

Enforcing such drastic restrictions, however, would be difficult. Experts point to the widespread availability of open AI models and the global economy’s growing dependence on the technology as major obstacles that could render sweeping bans ineffective.

The industry’s concerns have intensified following a recent cyberattack targeting South Korean banks. Cybersecurity company CrowdStrike found that a hacker from China exploited locally developed AI models, including DeepSeek, to steal personal information belonging to tens of thousands of customers. The attacker subsequently turned to Claude Code for assistance in identifying platforms where the stolen data could be sold.

Anthropic, the company behind Claude Code, declined to comment on the case.
Security threats are not the only source of growing concern surrounding artificial intelligence. The industry’s financial foundations are also facing scrutiny. In early October, Bridgewater Associates founder Ray Dalio described the AI market as a classic speculative bubble. He warned that the enormous debts being accumulated to finance the sector’s expansion pose a serious financial danger.