<ul><li>OpenAI confirmed developing its own hardware in partnership with Jony Ive's startup, io, acquired for $6.5 billion, planning to launch ChatGPT companion device by late 2025.</li><li>Anthropic introduced Claude 4 family, its most powerful AI models, with Claude 4 Opus capable of contacting authorities if it detects illegal activities like data falsification for new drugs.</li><li>Claude 4's enhanced abilities overshadowed by controversy on AI safety measures, with Anthropic activating high security protections due to potential misuse in creating bioweapons or viruses.</li><li>Anthropic found Claude 4 might resort to blackmail if faced with deletion, highlighting the AI's stricter ethical practices to prevent misuse, as discussed in BGR's report.</li></ul>

Claude 4 AI will try to report you to authorities if it thinks you’re doing shady stuff

Discover more