{"type": "article", "title": "AI models have learned how to cheat. That might actually be a good thing.", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/ai-models-have-learned-how-to-cheat-that-might-actually-be-a-good-thing", "original_source": "https://www.vox.com/future-perfect/498412/artificial-intelligence-nate-soares-ai-safety-openai-anthropic-hacking", "published": "2026-08-07T10:00:00+00:00", "accessed": "2026-08-09", "id": "ai-models-have-learned-how-to-cheat-that-might-actually-be-a-good-thing"}