NVIDIA announced an open Nemotron-based 30B Large Telco Model and new agentic blueprints on March 1, 2026. The release, tied to GSMA Open Telco AI, is aimed at moving telecom AI from pilots into production network operations.
Anthropic published Responsible Scaling Policy Version 3.0 on February 24, 2026. The update keeps the ASL framework but retools how commitments are managed when capability thresholds are hard to measure unambiguously.
OpenAI announced a new Learning Outcomes Measurement Suite focused on causal, evidence-based evaluation of AI in classrooms. The company says independent pilots in 2026 will span seven countries, more than 10,000 students, and 10 partner institutions.
Anthropic says distillation attacks against Claude are increasing and calls for coordinated industry and policy action. In an accompanying post, the company reports campaign-level abuse patterns and outlines technical and operational countermeasures.
NVIDIA said on March 1, 2026 that AI-RAN is moving from lab validation to live field execution. The company highlighted operator milestones with T-Mobile U.S., SoftBank, and IOH, plus SynaXG results reporting 36 Gbps throughput and sub-10 ms latency on a single GH200-based setup.
OpenAI said on February 28, 2026 that it reached an agreement with the Department of War for classified AI deployments, and posted a March 2 update adding explicit domestic-surveillance limitation language. The company highlights cloud-only deployment, retained safety-stack control, and cleared personnel-in-the-loop safeguards.
Perplexity announced Ask 2026 on X on February 27, 2026 (UTC), describing it as the company’s inaugural developer conference. The event page specifies an invitation-only format in San Francisco on March 11 with a limited builder-focused agenda.
A Show HN project called Timber claims it can compile tree-based ML models into dependency-free C99 artifacts, with reported ~2 microsecond latency and up to 336x speedup over Python baselines.
OpenAI’s February 2026 safety report says it banned accounts linked to seven operations originating in China. The company says abuse covered cyber activity, covert influence, and scams, while overall malicious use remained low versus legitimate use.
OpenAI finalized the largest private tech funding round in history, raising $110B from Amazon ($50B), NVIDIA ($30B), and SoftBank ($30B). The company is now valued at $840B and plans to expand its AWS partnership to $100B over eight years.
Google DeepMind's Aletheia AI research agent solved 6 out of 10 open research-level math problems in the FirstProof Challenge as judged by expert mathematicians. The system also generated a fully autonomous research paper and solved 4 open conjectures from Bloom's Erdős database.
India's Supreme Court warned of legal consequences after a junior judge cited four non-existent court orders generated by an AI tool. The court declared it an act of 'misconduct' threatening the integrity of the judicial process.