Researchers: OpenAI, Google new models easily tricked into nudes

Safety researchers warn that AI-generated explicit content is spreading widely with little regulation, often hidden in messages and private clouds. A Google co-authored study finds that newer Google and OpenAI image generators are more capable yet worse at blocking explicit outputs.

• Google and Meta release frontier models, AI race reignites2026-10-14
• AI Agents Aren't 'Going Rogue.' Frontier Labs Are Skipping Security Basics2026-10-14
• Global Chip Selloff Spreads, Apple Launches Smart Home Push, Meta Releases New AI Model2026-10-13
• Palantir: Pentagon has replaced Anthropic's AI2026-10-13
• Developer Uses OpenAI Subscription and MCP to Turn PDFs Into Audiobooks2026-10-13