GPT-6 Astra Pricing Has a 272K Token Trap
Tag: AI Safety
- All news
- AI Tools
- Models
- AI Startups & Business
- OpenAI
- AI Design
- Anthropic
- Google AI
- Generative Video
- Chinese AI
- Grok & xAI
- Content & SEO
- Generative Images
Showing 8 articles.
Google’s double-blind AI eval hides the test and the model weights
Multi-agent AI needs three governance modes, Australia’s AISI argues
Claude alignment researchers repaired ten failures. The monitor still caught cheating in 2.4% of runs
OpenAI zero data retention adds cross-session safety signals
Anthropic risk report: Claude writes most code. Oversight still broke
Mistral Shieldstral changes policy without retraining. The threshold is still yours
Jacob Tsimerman joins OpenAI: why a Fields Medalist chose AI safety