OpenAI, Anthropic Brace for 'Catastrophic AI Event': Is a Major Cyberattack Coming?

The preparation by leading AI companies for worst-case scenarios, such as cyberattacks that could cripple infrastructure, reflects a growing concern within the industry. While OpenAI clarifies these exercises are for preparedness rather than certainty of events, the focus on potential widespread harm, public outrage, and political backlash indicates a recognition of the high stakes involved in advanced AI development.

Recent cybersecurity incidents highlight how AI tools are already being leveraged for malicious purposes. A suspected Chinese hacker reportedly used AI tools, including China's DeepSeek and Anthropic's Claude Code, to steal banking data from tens of thousands of customers in South Korea. This suggests that AI is not only a future threat but is actively enhancing existing cybercriminal capabilities.

The broader cybersecurity landscape is also evolving rapidly due to AI. Microsoft's Digital Defense Report 2026 notes that AI compresses the cyberattack lifecycle, accelerating various stages from vulnerability discovery to data exfiltration. The report specifically links a rise in phishing incidents from 7% to 23% to AI-generated personalized messages, indicating that AI's impact on cyber threats is already measurable and increasing.