AI Bosses Privately Assume a Catastrophe by 2027, Axios Reports; OpenAI Calls It a Drill, Not Destiny

The fine print

VandeHei posted the line on X on the morning of October 9, 2026, linking Maria Curi's Axios scoop. The scoop said the officials 'anticipate a large-scale event, most likely a cyberattack, that shuts down access to financial services, internet connectivity, or even power and water', and: 'Many AI industry insiders told Axios they believe a major event will occur in the next six to 12 months.' An OpenAI spokesperson told Axios: 'OpenAI conducts preparedness exercises where teams discuss and work through a range of potential scenarios. These scenarios are not treated as inevitable, but are meant to help us prepare for a variety of circumstances.' Anthropic declined to comment. A month earlier, in his Axios C-Suite column of September 8, 2026, VandeHei wrote: 'I work under the assumption that something big and bad will hit in the next 18 months, most likely a widespread cyber attack or bio-engineered virus set loose by a bad actor.' The clock here is his 2027 line; the 18-month window ends in March 2027 and the insiders' 6-to-12-month window ends in October 2027, both inside it.