OpenAI Accuses Moonshot AI of Over 10,000 Attacks to Crack GPT Reasoning

This accusation highlights security vulnerabilities and potential escalation risks in AI systems.

Auto-scored: Alleged cyber-attacks on AI models represent a real escalation without mitigation evidence.