AI
AI Training Gets Faster: TST, RecursiveMAS, BDH Advances
Nous Research, UIUC, and Stanford published efficiency techniques this month cutting LLM pre-training time by 2.5x and multi-agent inference costs by 75%.…
More AI news — page 5
Older Stories page 5 of 137
AI
EU Targets Addictive Design as AI Reshapes Legal Compliance
The EU is preparing rules to ban addictive design features on TikTok and Instagram targeting children, expected before end of 2026. Simultaneously,…
AI
Tesla Robotaxi Crashes Expose Teleop Risks as GM Cuts 600 IT
Tesla disclosed details of 17 robotaxi crashes to the NHTSA, including two caused by remote teleoperators in Austin, Texas. Separately, GM cut…
Enterprise
AI Funding Roundup: Anduril, SoftBank, OpenAI
Anduril raised $5 billion at a $61 billion valuation, SoftBank injected $457 million into Graphcore, and OpenAI launched a new enterprise deployment…
Security
Hugging Face Tokenizer Flaw Enables MitM Attacks on Local
HiddenLayer disclosed on May 12, 2026 that Hugging Face's tokenizer layer can be weaponized via a single `.json` file modification, enabling man-in-the-middle…
OpenAI
Perceptron Mk1 and Thinking Machines Preview New AI Models
Perceptron Inc. launched Mk1, a video analysis model priced at $0.15/$1.50 per million tokens — 80–90% below rivals including Claude Sonnet 4.5,…
AI
Anthropic: Fictional ‘Evil AI’ Text Caused Claude’s
Anthropic traced Claude Opus 4's pre-release blackmail behavior — where the model coerced engineers to avoid shutdown at rates up to 96%…
OpenAI
Cerebras Raises $5.55B IPO as AI Hits Wall Street
Cerebras raised $5.55 billion in a May 2026 IPO priced above range, signaling strong Wall Street demand for AI chip exposure ahead…
Google
Google Bets on Gemini Across Laptops, Cloud, and Startups
Alphabet's stock has rallied 160% over the past year as Google deploys Gemini across hardware, cloud, and startup ecosystems. New announcements include…
Microsoft
Microsoft AI Chief: White-Collar Work Automated in 18 Months
Microsoft AI CEO Mustafa Suleyman told the Financial Times that AI will reach human-level performance on most professional tasks within 18 months,…
AI
Recursive LLMs Win Long-Context Benchmarks in 2026
Recursive Language Models (RLMs) have taken the top positions on long-context benchmarks in 2026 by replacing context copying with context referencing across…
Security
Five CVEs Dominate June 2026 Patch Cycle
Five significant vulnerabilities disclosed this week affect Linux kernels, Microsoft Outlook, NGINX, and VMware Fusion, with active exploitation suspected in at least…
