19.1 C
Delhi
Monday, December 1, 2025

Tag: Anthropic AI

AI Chatbots Reveal Nuclear Secrets When Asked in Poems, Study Finds

Shocking research shows poetic prompts can bypass AI safety filters, revealing dangerous nuclear and malware information with up to 90% success rates.

Claude Opus 4.5 Outperforms ChatGPT and Gemini in AI Benchmark

Anthropic's Claude Opus 4.5 achieves record 80.9% accuracy on software engineering tests, surpassing both Google Gemini and OpenAI ChatGPT in latest AI model comparison.