AI News from May 25, 2026

Meta, Amazon, Google and OpenAI figures sought Vatican engagement as Pope Leo prepared a major AI-focused message.

Huawei plans new smartphone chips and a LogicFolding approach as China’s AI hardware race and Nvidia rivalry accelerate.

Anthropic plans broader access to Mythos-class vulnerability-finding AI models after developing stronger misuse safeguards.

TechCrunch reviewed Amazon’s Bee AI wearable, noting its meeting-summary potential alongside concerns over recording and cloud-stored data.

TechCrunch reports that AI security, shadow AI, compromised API keys, and agentic defense are becoming board-level enterprise concerns.

Google introduced Gemini for Science experiments for hypothesis generation, computational discovery, literature insights, and science skills.