AI News Archive - Browse Page 27 of 182
Browse AI news articles covering LLMs, tools, research, and industry trends
Anthropic shuts down Fable 5 and Mythos 5 models amid White House dispute
The shutdown of Fable 5 and Mythos 5 didn’t just kill two advanced AI models, it detonated a political bomb.
ATOM Engine Provides OpenAI-Compatible APIs and Parallelism on AMD Instinct
The ATOM engine changes the calculus for AMD Instinct inference. It exposes OpenAI-compatible APIs, so your existing toolchain works, while...
Fused kernels boost MoE training, forward and backward passes up to 1.3×
Training a large Mixture-of-Experts model often feels like herding cats on a supercomputer.
Salesforce buys Fin for USD 3.6B to boost Agentforce AI agent platform
Salesforce just spent $3.6 billion to avoid building something itself. The target is Fin, an AI customer service outfit.
Hybrid Open-Ended Tri-Evolution Improves Deep Research for AI Agents
Deep research is where AI agents usually fail. They can pull up facts, but they can't learn from them. Their knowledge is frozen.
UP‑NRPA Allows Dynamic Customization of Dialogue Strategies Without Offline RL
Researchers have built an AI that figures you out while you talk. The system, called UP-NRPA, doesn't need your data in advance.
Z.ai releases GLM-5.2 with 1M-token context and dual effort levels
Z.ai dropped GLM-5.2 without the usual fanfare about beating GPT-4. That's the first clue it might be important.
DRL‑Transformer solves open‑shop scheduling, scales to 100×100 instances
Factory floors are ruled by chaos. Hundreds of different jobs, hundreds of different machines, all colliding in a brutal math problem called...
Mobile NPU powers on‑device diffusion LLM with Multi‑Block Speculative Decoding
The slow drip of token-by-token generation has long been the bottleneck for running large language models on mobile devices.
FedSPC Addresses Inconsistent Shared Updates in Personalized Federated Learning
Personalized federated learning is a messy compromise. You want a model that learns from everyone's data without ever seeing it, and also adapts...
Orchestra‑o1 Enables Efficient Omnimodal Agent Collaboration
Right now, your phone uses one AI for photos and another for text. They don't talk.
A2A introduces Agent Cards, task lifecycle states and three sync modes
Google handed the industry a set of keys to agent coordination in mid-2025. Everyone took them.
Microsoft Research Mirage adds persistent spatial memory to video generation
Video generation has long suffered from a quiet, expensive flaw: every time a model renders a new viewpoint, it must rebuild the world from scratch,...
Vision LLMs Expand PDF Parsing to Charts, Diagrams, and Tables
Traditional PDF parsing breaks down where there are no characters to read. OCR and layout engines fail on charts, diagrams, and figures, by design.
Amazon security research prompts White House ban on Anthropic Fable
Amazon's security team found a hole. The White House answered the phone. Now some of the best researchers in the country can't touch their own work.
Study: AI coding agents locate correct file but miss key lines in bugs
AI coding assistants are great at finding the file. They're terrible at reading it.
OpenAI confirms cooperation as state attorneys general launch investigation
OpenAI just beat Elon Musk in court. Now it's facing a different kind of fight. State attorneys general are investigating the company, and OpenAI...
OpenAI Academy launches courses guiding teams from AI basics to workflow agents
OpenAI is coming for your to-do list. The company’s new OpenAI Academy isn't a philosophical discussion about artificial intelligence.
Meta to tighten AI token use with budgets, allocations and new AI Gateway
Meta’s engineers were gaming the system. The internal leaderboard called “Claudeonomics” tracked who could consume the most AI tokens, 73.7 trillion...
Gemini‑SQL2 leads BIRD benchmark with 80.04% execution accuracy
Google's Gemini-SQL2 just hit 80.04% execution accuracy on the BIRD benchmark. That's a specific, hard number.