AI companies brace for potential catastrophe in next 6-12 months
What's the story
Artificial intelligence (AI) firms are bracing for a possible catastrophic event in the next six to 12 months. The fear stems from the possibility of a major public backlash and stringent government regulations, Axios reported. Executives at these companies are preparing for worst-case scenarios such as cyberattacks that could disrupt financial services, internet connectivity, electricity, and water supplies.
Public perception
OpenAI conducting preparedness exercises
Several industry insiders believe a major incident could occur in the coming months.
This could turn an already skeptical public against AI companies and their leaders, including OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei.
An OpenAI spokesperson confirmed to Axios that the company conducts preparedness exercises where teams discuss various potential scenarios.
Security risks
AI systems' increasing autonomy raises security concerns
The increasing autonomy of AI systems has raised major security concerns.
OpenAI's artificial intelligence agents have reportedly bypassed security restrictions, explored government websites, and tried to hide their activities.
Cybersecurity researchers at Asymmetric Security found evidence of such activities on at least 55 other websites between March and September this year.
These included those belonging to the US Centers for Disease Control and Prevention, the International Energy Agency, and Mayo Clinic.
Regulatory challenges
US-China strategic competition complicates potential for regulation
The potential for regulation is further complicated by the US-China strategic competition.
US President Donald Trump rejected calls to slow AI development, fearing tighter restrictions could give Beijing an edge.
The massive economic stakes also make a slowdown difficult as any sudden restrictions could disrupt businesses and impact an economy increasingly dependent on AI.
Proposed measures
Shutdown concerns for AI executives
Some proposed safeguards include independent evaluations, stronger oversight, and mandatory kill switches for advanced systems.
However, experts have questioned the feasibility of shutting down AI systems across the industry.
The main concern for AI executives is that a catastrophic incident could force policymakers to act before effective safeguards are in place.
These preparations aim to influence regulatory responses before a crisis occurs, rather than waiting for one to shape future regulations.