AI Industry Braces for Cyberattack Crisis as 116 Firms Sound Infrastructure Alarm

Leading artificial intelligence companies are privately preparing for large-scale cyberattacks on critical infrastructure within the next 6 to 12 months, according to industry insiders familiar with contingency planning at Anthropic and OpenAI. On August 27, 2026, 116 organizations—including OpenAI, Google, and Anthropic—issued a joint warning that AI-enabled cyberattacks pose an escalating threat to banking networks, power grids, water supplies, and internet connectivity. The signatories cautioned that "status quo security won't be enough" and that defenders currently hold a narrowing advantage that will "quickly erode."