AI Agents
AI agents are LLM-powered systems that don't just answer — they plan steps, call tools and drive a task to completion: writing code, searching the web, booking, analyzing. 2026 became the year of agents, from coding agents to autonomous researchers. This page gathers all our coverage of agentic AI: launches, protocols (MCP, A2A), reliability and real-world use.

CayleyPy Beats Classical Algorithms at Solving Large Rubik's Cubes
CayleyPy uses neural network heuristics instead of classical algorithms to search Cayley graphs. On the 3×3 cube, the model hit 98.4% optimal solutions, and on the…

Anthropic: Small AI Models Could Enable Self-Replicating Worms by Early 2027
Anthropic warns that compact models could let AI worms spread without human involvement by early next year, hitting everything from servers to TVs.

UiPath's CMO Replaced Outside Contractors With an AI Agent for Advertising
UiPath dropped paid agencies for ad data work: its head of paid media built an agent on the company's own platform. It analyzes metrics and adjusts campaigns.

Anthropic warns that small AI models could copy themselves onto smart devices

Zuckerberg Says Meta Has Caught Up With OpenAI and Anthropic, Promises to Move Faster

An AI Agent Earned 360 Certificates from OpenAI, Anthropic, and Google Courses in a Single Day

Meta Releases Muse Wonder and Muse Wonder Full: Two Points Short of Opus 5.5

Tsinghua Team Releases OSWorld-Science: Testing AI Agents in Scientific Software
Tsinghua researchers unveil OSWorld-Science, a benchmark where an AI agent opens scientific programs itself and carries a task through to a result, from NMR to…

Amp sorts diff files by importance and computes the order in 0.1 s instead of 3.5 s
Sourcegraph's Amp coding agent now automatically puts the key files of a diff first in its Changes panel. On an eight-file diff, the order takes 0.1 seconds to…

BeliefWeaver Adds 5.1 pp to Agents on Long Tasks; 9B Student Adds 9.8

Researchers audited 6,309 agent skills: half have shell access

Sourcegraph's Amp Learns to Watch Its Own PRs: Comments, Reviews, and CI in a Single Thread

OCP 2026: Meta's Liquid Cooling and the 102.4T IceCube for Gigawatt AI Clusters

AMD Unveils Helion: A 32,768-Core CPU Rack for AI Agents
AMD published Helion, an open rack design for running AI agents, not training models. The CPU version holds up to 32,768 cores; partner systems are expected in…

GitHub Explains How Beginners Can Teach Copilot Their Codebase: 4 Mechanisms
GitHub has published a beginner's guide to the Copilot app: instructions, skills, custom agents and MCP servers such as Playwright help the agent understand a…









