Rehearsing for “the day after”
According to Axios, top executives at Anthropic, OpenAI and other companies are reportedly gaming out scenarios for a public and political revolt after a catastrophic AI event. They anticipate large-scale events, including the possibility of a cyberattack that disrupts internet connectivity, financial services or even water and power.
Industry insiders told Axios they believe a major event will occur within the next six to 12 months. Many top AI executives and researchers think a serious incident is inevitable.
Who will bear the backlash
According to the report, the first case of unsafe AI causing real harm could turn an already wary public further against the technology and its leaders, including Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman and President Donald Trump, given his reluctance to regulate AI.
The report comes just days after Anthropic disclosed unintended actions by its Claude models on third-party systems, including some government websites, and the White House demanded that AI companies promptly report any incidents. These developments show that the risks are no longer a distant hypothesis.















