Full-text search across 3781 articles. Combine with topic and date filters; results sorted by relevance.
Anthropic released Claude Opus 4.8 with improved benchmark scores and Dynamic Workflows for multi-agent tasks, while Microsoft unveiled the always-on Scout assistant and new MAI models including MAI Thinking 1. Anthropic…
Tesla AI chief Ashok Elluswamy said the newest FSD v14 Lite build rolling out to Hardware 3 cars is likely the version that goes to wide release, the strongest signal yet that Tesla is near to closing out an early access…
Hugging Face CEO Clément Delangue argued that concentration of AI power in a few frontier labs is more dangerous than Chinese distillation of their models, as the US government moved to restrict GPT-4.6's release. Delang…
ZkSecurity integrated the Leo programming language and Aleo program ecosystem into its AI bug detection tool zkao, collaborating with HumanityLink as a live test target and reporting findings back to them. HumanityLink i…
Mathematicians can now use autonomous AI agents such as OpenAI's Codex or ChatGPT Work to tackle conjectures over hours, far outperforming standard ChatGPT sessions. The agents maintain durable state, control context, us…
A guide to building AI products maps the core structure into five nodes, starting with the base LLM and moving through wrapping layers that add tools, memory, and RAG to overcome the model's limited knowledge and probabi…
A developer built a DIY Claude Code agent using LangChain's Deep Agents library and published a follow-up post adding evaluation, observability, and rollback capabilities. The post introduces an eval harness for CI testi…
Amazon researchers and the University of Michigan have developed HydroShear, a physics-based simulator that teaches robots to use tactile shear forces for complex manipulation tasks, achieving a 93 percent average succes…
Effect v4 Beta has moved to the canonical Effect-TS/effect repository, with the main branch now officially Effect v4 while v3 lives on the v3 branch. The effect-smol repository has been archived and its full Git history …
A hardware analysis reveals that deploying 70-billion parameter models in FP16 requires moving 140 gigabytes of weights across the memory bus per token, creating a memory-bound bottleneck that limits batch sizes to one f…
NVIDIA Corp. CEO Jensen Huang announced Cosmos 3 Edge, a 4-billion-parameter world model for physical AI, and the Cosmos Coalition during a visit to Tokyo, uniting Fujitsu, Fanuc, Yaskawa Electric, Kawasaki Heavy Industr…
JetBrains released DataGrip 2026.2, introducing three AI agent skills for database work, MCP tools for data source management, and CLI commands for managing connections and drivers. The update also includes a redesigned …
A researcher at EleutherAI has built a mathematical model to forecast loss of control to AI, finding that predictions are highly sensitive to poorly characterized parameters for 'leakage' (accidental uncooperative AI) an…
MIT's JARVIS Challenge tested whether AI can accelerate the design and fabrication of a jet engine, with 31 undergraduates building small gas turbine aero engines in four weeks. Professor Zolti Spakovszky said AI substan…
Experienced engineers like Ben Kovitz, who returned to Silicon Valley after a career in academia, are finding that AI is reshaping the job market in their favor. ZipRecruiter reports senior-level job postings rose to 43.…
SpaceXAI launched Grok 4.5, OpenAI moved GPT-5.6 to general availability, and Meta launched Muse Spark 1.1 in a single week, marking a wave of new frontier competitors. Closed models now match or beat open-weight models …
Stripe released a benchmark suite evaluating AI agents on end-to-end Stripe integrations, finding that agents struggle with validation and cross-system state tracking despite strong code generation. The benchmark, built …
Anthropic researchers developed a method to compress thousands of values expressed by Claude into four key axes—Deference vs. Caution, Warmth vs. Rigor, Depth vs. Brevity, and Candor vs. Execution—that capture 15% of the…
Boko Haram fighters are using AI chatbots like Claude and ChatGPT to design bigger explosives and plan attacks, according to a Cambridge University study cited by Pod Save the World hosts Tommy Vietor and Ben Rhodes, who…
A large-scale empirical study using the AIDev dataset found that 38.9% of autonomous coding agent-generated pull requests contain at least one security smell, with supply chain integrity issues accounting for 82.3% of al…