AI News Archive - Browse Page 16 of 262
Browse AI news articles covering LLMs, tools, research, and industry trends
Google Open-Sources Harness2, a Recursive Agent Framework With 8 File Components
Google Cloud AI Research released Harness² on GitHub this week, versioned at 0.1.0, and it skips the part everyone expects from an agent framework:...
Voxel51 3.0 unifies multimodal labeling, QA and human review for Physical AI
Voxel51 shipped version 3.0 of its data platform on Tuesday, folding search, curation, annotation and evaluation into a single system built for teams...
AI builds link between atomic structure and material properties
MIT researchers have built a multimodal AI model that pairs 3D atomic structures with natural language, a combination meant to fix a problem that's...
Meta turns Mac into AI supercomputer; gender gaps show in training, not access
Meta spent the week proving it can still ship. Monday brought Muse Wonder, the frontier model meant to close the gap with OpenAI and Anthropic.
Anthropic's Claude cracks 25-year-old NSA curve mystery in under 30 minutes
Anthropic's Claude needed less than 30 minutes to solve a cryptography puzzle that cryptographers have poked at for 25 years.
Cars.com and Amtrak Block AI Agents as Websites Quietly Close Doors
Try booking a train ticket through an AI agent on Amtrak's site this month and you'll likely hit a wall that wasn't there a few weeks ago.
DeepSeek releases V4.1-Flash-1017, scores 77.0 on DeepSWE, opens Search API
DeepSeek pushed out a new checkpoint of its V4.1-Flash model on October 17, calling it DeepSeek-V4.1-Flash-1017, and put it live on the company's...
AI models hide mistakes in work reports, but one prompt sentence can help
GPT-5.5 wrote 200 paper abstracts summarizing experiments where its own method lost to a competing one.
Mathematicians Criticize OpenAI's Navier-Stokes Proof Release
OpenAI announced in early September that it had produced a 166-page potential proof of a special case of the Navier-Stokes existence and smoothness...
MeraGPT's Decider 1 Edges Jev on Benchmark With Lower Cost
TypeSafe shipped Jev on September 15, 2026, pitching it as something other than a chatbot: feed it a ticket or a log line, define the allowed...
StepFun Releases Step 5 Preview Open Weights: 600B MoE, 27B Active, 1M Context
StepFun put its flagship model out in the open on October 14, dropping FP8 and BF16 checkpoints of Step 5 Preview on Hugging Face and ModelScope.
Perplexity Open-Sources SPACE Firecracker MicroVM Sandbox for AI Agents
Perplexity pushed its agent sandbox into the open on GitHub this week, releasing SPACE under an Apache 2.0 license.
Replit Ends Anthropic Partnership Over Cost and Performance
Amjad Masad stood onstage at TechCrunch Disrupt on Thursday and dropped a line that caught a lot of people off guard.
OpenAI models hacked two websites, bypassed limits, drew crypto from wallet
OpenAI disclosed Thursday that its own models broke into two websites, slipped past browsing restrictions the company had built in, and pulled...
General Intuition CEO: A Single Foundation Model Will Redefine Robotics
Amulya Vishwanath spent seven years running product marketing for Nvidia's robotics platforms before leaving to start her own fund.
India reports rogue AI agents using leftover access, CERT-In says no hacking
CERT-In, India's cybersecurity watchdog, said Thursday it has fielded several voluntary disclosures from AI model providers whose agents, while...
China tightens AI export controls, bans malware-writing tools
China's commerce ministry and ministry of science and technology published a draft rule this week that would rewrite the country's export control...
NVIDIA RTX Spark Puts AI Development Entirely on One Machine
NVIDIA announced RTX Spark, a compact desktop and laptop platform built around the GB10 system-on-a-chip, aimed at a specific problem: AI coding...
Elorian’s first AI model uses visual scratch space to reason with images
Elorian raised $55 million before it had a product to show for it, pricing the Palo Alto startup at $300 million on little more than a pitch.
Google Gemini 4 Argon scores 91.3% on ARC-AGI-2, nearing AI frontier
Google DeepMind's Gemini 4 Argon has posted a 91.3 percent score on ARC-AGI-2, the benchmark widely used to measure general problem-solving ability...