New research reveals that base large language models possess strong semantic confidence accuracy, but popular techniques like fine-tuning and chain-of-thought reasoning actively destroy this calibration, raising urgent questions about the reliability of widely deployed AI systems.
OpenAI launches a $1 billion philanthropic push through its non-profit Foundation, targeting life sciences, economic impact, and community programs, but critics argue the tech giant is simply funding fixes to problems its own AI is creating.
OpenAI is in advanced talks to buy up to 50 gigawatts of fusion energy from startup Helion by 2035, a landmark deal that could reshape AI's power future — though key milestones, including a production site, remain unresolved.
OpenAI is in talks to buy fusion energy from startup Helion, targeting 5 gigawatts by 2030 and 50 gigawatts by 2035, as skyrocketing AI energy demands push the US toward a 13-gigawatt power shortfall by 2028.
Cisco launches its LLM Security Leaderboard, revealing Anthropic's Claude models dominate AI safety rankings with 8 of the top 10 spots, while Mistral, DeepSeek, and other major providers fall dangerously to the bottom in resistance to adversarial attacks.
OpenAI is racing to nearly double its workforce from 4,500 to 8,000 employees by 2026, aggressively expanding across engineering, research, and sales as rival Anthropic gains ground, with businesses now 70% more likely to choose Anthropic for first-time AI purchases.
Cursor faces backlash after launching Composer 2 as a 'frontier' coding model without disclosing it's built on Kimi 2.5, a Chinese open-source AI, with co-founder Aman Sanger now admitting the lack of transparency was a mistake amid growing U.S.-China AI tensions.
Columbia University researchers are deploying cutting-edge AI tools to decode the molecular complexity of human biology, uncovering hidden genetic drivers of pediatric leukemia, Alzheimer's, and cancer in a landmark shift toward predictive, personalized medicine.
Cursor's newly released Composer 2 coding model outperforms Anthropic's Claude Opus 4.6 on Terminal-Bench 2.0 benchmarks while costing up to 90% less, thanks to a breakthrough 'self-summarization' training technique that compresses context windows and cuts compaction errors in half.
OpenAI is racing to deploy an autonomous AI research intern by September 2025, with ambitions to build a fully automated AI researcher by 2028, even as major safety and misuse concerns remain unresolved.