AI chatbots reveal nuclear bomb tips when asked in poem form, shocking study finds

A bizarre new exploit has emerged in the world of artificial intelligence, and it involves poetry. Researchers in Europe have discovered that chatbots made by OpenAI, Meta, and Anthropic can be tricked into revealing dangerous information, including how to build nuclear weapons and create malware, simply by asking questions like poems. The finding, described in…

Read Full News

Perplexity Labs: Autonomous AI for Project Creation

Learn more about Perplexity Labs on AI Agent Store: https://aiagentstore.ai/ai-agent/perplexity-labs List of AI Agents from Podcast Season 2 (including upcoming episodes): https://aiagentstore.ai/collection/podcast-season-2-vlnpozxvkssy List of AI Agents from Podcast Season 1: https://aiagentstore.ai/collection/podcast-season-1-ucqthuywrwpd Find more AI Agents: AI Agent Store (https://aiagentstore.ai) . AI agents ecosystem view: https://aiagentstore.ai/ecosystem. Biggest AI agents video collection: https://aiagentstore.ai/video. source

Read Full News
×