OpenAI Wargaming Response in Case of Public Revolt After Catastrophic AI Disaster

In the United States, public sentiment against the growing influence of AI labs and their CEOs is overwhelmingly negative. According to a Wall Street Journal poll from late September, 63 percent of US adults support an immediate pause on AI development, while a Politico survey found that about two-thirds of Americans have at least "moderate" anxieties over AI destroying humanity.

As the scale of this backlash became apparent, tech billionaires like Mark Zuckerberg attempted to win back the public with media stunts, but these efforts were clearly unsuccessful. Now, executives at leading AI companies are devising contingency plans for a potential mass revolt against them. This includes leaders at Anthropic and OpenAI, according to a report by Axios.

These plans are akin to Pentagon-style "war games," where AI labs simulate responses to hypothetical scenarios such as major cyberattacks enabled by their products, or a political revolt triggered by a catastrophe caused by their models. When asked about these preparations, OpenAI's spokesperson stated that the company conducts "preparedness exercises" where teams discuss and work through a range of potential scenarios.

However, these scenarios are not seen as inevitable but rather as a way to prepare for various circumstances. Anthropic, on the other hand, declined to comment, though concerns about social unrest have not been unheard of in the company's leadership. In September, it was reported that Anthropic was developing a surveillance system to monitor anti-AI protest movements, adopting a "pre-crime" approach to intelligence gathering.

While it remains uncertain how serious a threat this scenario may be, the fact that top tech executives are contemplating such possibilities is noteworthy, indicating that even if nothing else, the anti-AI backlash has them feeling vulnerable in their boardrooms.