8 results for “compliance”
DeepSeek-V4-Flash-0731 stable release launches on China's National Supercomputing Internet with one-click API access, model downloads, and a full Notebook environment. Benchmarks reportedly match top closed-source models.
Read more →
Article 50 of the EU AI Act took effect on 2 August 2026, requiring AI providers and deployers to disclose AI interactions, mark synthetic content, and inform people exposed to emotion recognition, biometric categorisation, deepfakes, and unreviewed AI-generated public-interest text.
Read more →
Large reasoning models solve Olympiad math and open research problems, but studies show their chains of thought may be unfaithful, unnecessary, or replaceable with meaningless tokens — raising questions about what actually drives their performance.
Read more →
Foundation demonstrates a tendon-driven robotic hand that catches a baseball mid-air, showcasing biomimetic design with predictive control and centralized actuation for high-speed manipulation.
Read more →
OpenAI outlines its responsible AI practices in Europe, aligning with the EU AI Act through enhanced safety, transparency, and cybersecurity measures. The company details its governance frameworks and collaborative efforts to support European competitiveness.
Read more →
Google DeepMind launches Gemini Robotics ER 2, an embodied reasoning model with video understanding, multi-step task orchestration, and multi-robot collaboration — now available via Gemini API and AI Studio.
Read more →
Google's updated Gemini macOS app adds Fn-key voice dictation and screen-aware reasoning that can read multiple local files and insert AI output directly into any document — challenging Apple Intelligence and ChatGPT on the Mac desktop.
Read more →
Google DeepMind CEO Demis Hassabis is calling for international oversight of frontier AI models as safety, security, and governance concerns grow.
Read more →