OpenAI says ChatGPT will end chats fixated on violence and limit weapons detail for teens
Rollout runs over the coming weeks, repeated endings can pause accounts, and a first Community Safety Report is due in early 2027.
- Eight independent AI evaluation groups form a forum to set ground rules for testing frontier models
- Startups such as Hirundo are editing Chinese open-weight models to strip censorship for Western businesses
- WIRED finds about 400 apps posing as ordinary AI tools that hide sexually abusive deepfake generators
- TÜV groups propose independent pre-launch testing of frontier AI and agents, led by Germany's AI safety institute
- Microsoft AI chief Mustafa Suleyman says labs must not build conscious superintelligence
- Anthropic's amended IPO filing shows $17.0 billion of third-quarter revenue and $895.4 billion of compute commitments
- Marvell agrees to buy Lumentum in an all-stock deal valued at $115 billion to bulk up in AI data-centre optics
- SemiAnalysis launches GDPx, a 206-task expert-work benchmark where the best model scores 12%
- Oracle launches a $10 billion bond sale and files to sell $10 billion of stock to fund its AI build-out
- Anthropic files to sell up to 552.9 billion yen ($3.5 billion) of IPO shares to investors in Japan
- OpenAI unveils an everyday-tasks benchmark and says GPT-6 Astra scores 80%, up from 49.3% for GPT-5.2