Anthropic’s draft federal AI bill would allow a court-ordered halt only when a frontier model poses an imminent risk of more than $1 billion in damage or over 50 deaths or serious injuries.
Anthropic says junior technical hires fell from 27% to 9% while employee-directed Claude agents logged six times more runtime than their human handlers.
DeepSeek's V4.1-Flash-1017 claims faster, shorter responses at the same model price while Search and Fetch APIs open its web tools to developers.
Reflection's Apache 2.0 Beam model has 501B total parameters, 23B active per token and a 1M-token context window for coding and agentic work.
Google Research says its agent workflows use a digital twin, simulation and human authorization to run and refine wet-lab experiments.
OpenAI and Meta warned a federal evidence panel that no universal tool can reliably authenticate every AI-generated image or video.
Kling 4.0 is publicly available with 30-second videos, stereo audio and keyframe controls, while MCP support lets AI assistants handle generation.
Anthropic is expanding independent AI safety reviews, with deeper access for METR and separate work involving SecureBio and Accenture.
Meta’s 816B-parameter Muse Spark Open targets local coding agents on two 512GB Macs, due later this month. API access is already available.
Google reports 99.99% magic-state fidelity on Willow, with errors reduced 40x. The quantum computing advance retained 8% of attempts and still needs to scale.
One user says Claude Code turned a 63-second video request into 161 Codex sessions—and kept the boss job. The screenshot shows a plan for 700 frames.
Tether has invested in AIsa, which combines AI-agent access to data, models and APIs with usage-based payments in fiat or stablecoins.
Google Deep Review offers two free document checks a week, using AI agents to flag errors and weak arguments. Its benchmark results come with important limits.
