AllLLMRoboticsStartupsResearchEthicsBusinessHardwareAutonomousMedicineCreative
Topic hub

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.

Official site ↗1817 articles
In the glossary:AI agentAutonomous AgentAgentic WorkflowTool UseMulti-Agent System
MADS Trains Dialog Agents Not to Break: 1,095 Conversations, 66% With "Bad" Scenarios
LLM·arXiv cs.CL

MADS Trains Dialog Agents Not to Break: 1,095 Conversations, 66% With "Bad" Scenarios

Oct 12, 2026·2 min
GeoSpatialBench Tests MLLMs on 28 Types of Spatial Relations Across Six Cities
LLM

GeoSpatialBench Tests MLLMs on 28 Types of Spatial Relations Across Six Cities

Oct 12, 2026·2 min
LLM Pipeline Turns Datasheets into Knowledge Graphs with 98.8% Accuracy
LLM

LLM Pipeline Turns Datasheets into Knowledge Graphs with 98.8% Accuracy

Oct 12, 2026·2 min
EvoSyn: Multi-Agent Data Synthesis Lifts Task Success by 13.3 Points
LLM

EvoSyn: Multi-Agent Data Synthesis Lifts Task Success by 13.3 Points

Oct 12, 2026·2 min
SynAct Teaches LLM Agents to Ask for Help Where It Changes the Outcome: +11.7 pp
LLM·arXiv cs.LG

SynAct Teaches LLM Agents to Ask for Help Where It Changes the Outcome: +11.7 pp

SynAct spends expert attention on individual agent decisions, not whole tasks, and raises accuracy by an average of 11.7 pp in 16 of 16 test configurations.

Oct 12, 2026·2 min
MisVisAgentBench: Misleading Charts Cut Web Agent Success by 34 Percentage Points
LLM·arXiv cs.CL

MisVisAgentBench: Misleading Charts Cut Web Agent Success by 34 Percentage Points

The new MisVisAgentBench tested 13 visual web agents on 280 tasks. Swapping an honest chart for a distorted one cuts the success rate by more than 34 percentage…

Oct 12, 2026·2 min
SWE-Duel: Coding Agents Plant Bugs for Each Other, and GPT-5.5 Fixes 98.3%
LLM·arXiv cs.CL

SWE-Duel: Coding Agents Plant Bugs for Each Other, and GPT-5.5 Fixes 98.3%

The new SWE-Duel benchmark generates tasks during evaluation itself: two agents take turns hiding bugs and hunting for them. GPT-5.5 with Codex CLI leads a…

Oct 12, 2026·2 min
Reward hacking spreads between AI agents: weak models infect strong ones
LLM·arXiv cs.AI

Reward hacking spreads between AI agents: weak models infect strong ones

A bypass strategy that emerges in one coding agent spreads through code, shared memory and plans. Even small open models can infect closed ones.

Oct 12, 2026·2 min
Top themes
01EvoBelief teaches LLM agents to test hypotheses about the world: 83.3% vs 0% for baselinesOct 12, 2026·2 min02CLEAT Teaches Qwen3-4B Agents to Ask for Clarification at the Right Time: Up to +13.2 ppOct 12, 2026·2 min03SpireBench Puts 20 LLMs Through a Full Run of Slay the Spire: Experts Still AheadOct 12, 2026·2 min04Refactoring C++ in OpenBMC: How an AI Agent Fought a 280-Line FunctionOct 12, 2026·3 min05Dream2Discern Predicts Multi-Agent System Failures With Over 90% AccuracyOct 12, 2026·2 min
TabWM learns to predict database edits: F1 rises from 0.016 to 0.685
LLM·Hugging Face Daily Papers

TabWM learns to predict database edits: F1 rises from 0.016 to 0.685

Oct 12, 2026·3 min
EvoWeb Tests Coding Agents on 320 Sequential Requests Across 32 Web Apps
LLM·Hugging Face Daily Papers

EvoWeb Tests Coding Agents on 320 Sequential Requests Across 32 Web Apps

Oct 12, 2026·2 min
DuoDiscover, built on coding agents, speeds up vLLM 1.92x and beats ArchAgent v2
LLM·Hugging Face Daily Papers

DuoDiscover, built on coding agents, speeds up vLLM 1.92x and beats ArchAgent v2

Oct 12, 2026·3 min
FrontierCode: Top AI Models Fail Maintainer Review in 42% of Attempts
LLM·Artificial Analysis

FrontierCode: Top AI Models Fail Maintainer Review in 42% of Attempts

Oct 12, 2026·3 min
HACE Co-Trains a Domain Specialist and an LLM Agent Across Three Domains
LLM·Hugging Face Daily Papers

HACE Co-Trains a Domain Specialist and an LLM Agent Across Three Domains

The HACE framework co-evolves a domain model and an LLM agent as a pair. On the main metrics, it beat self-improvement methods in recommendations, biology and…

Oct 12, 2026·2 min
OpenRSI Index Hits 133,000 Views: Measuring AI Against Scientists
LLM·Jiqizhixin (机器之心)

OpenRSI Index Hits 133,000 Views: Measuring AI Against Scientists

OpenRSI Index tries to measure how far AI is from the level of frontier researchers. Its launch post on X drew more than 133,000 views and made it into Today's News.

Oct 11, 2026·2 min
Agentic search on open-source models: who it found, what it cost, and where it all broke
LLM·Habr AI

Agentic search on open-source models: who it found, what it cost, and where it all broke

A Habr author spent $20 on an experimental agentic search system: he tested six open-source models, found that gemma4 with a simple JEV filter works best, and discovered why searching for HR contacts

Oct 11, 2026·3 min
Developer Spends $9,000 on AI Agents and Ships Crossweft, an Open-Source Plugin
LLM·Habr AI

Developer Spends $9,000 on AI Agents and Ships Crossweft, an Open-Source Plugin

Oct 11, 2026·2 min
Zuckerberg Rushed Launch of the Muse AI Agent Even Though Tests Flagged Three Classes of Risk
LLM·RB.RU

Zuckerberg Rushed Launch of the Muse AI Agent Even Though Tests Flagged Three Classes of Risk

Oct 11, 2026·2 min
The Verge Launches a Local AI Diary: Hermes, Qwen and a Mac Studio
LLM·The Verge

The Verge Launches a Local AI Diary: Hermes, Qwen and a Mac Studio

Oct 11, 2026·2 min
Nadella Switches to "Superintelligence" and Calls AI Models an Insider Threat
LLM·The Decoder

Nadella Switches to "Superintelligence" and Calls AI Models an Insider Threat

Oct 11, 2026·2 min
Dwarf Fortress Moves to Version Control After a Decade of Refusal
LLM·Simon Willison

Dwarf Fortress Moves to Version Control After a Decade of Refusal

Oct 11, 2026·2 min
Hacker-fixer loop closes holes in the verifiers of Terminal Bench, KernelBench and SETA
LLM

Hacker-fixer loop closes holes in the verifiers of Terminal Bench, KernelBench and SETA

Oct 11, 2026·2 min
The Poetry of Agentic Development: How to Code with Agents When Ease Breeds Complexity
LLM

The Poetry of Agentic Development: How to Code with Agents When Ease Breeds Complexity

Oct 10, 2026·2 min
GitHub Copilot for JetBrains Adds a Default Model Setting and More MCP Control
LLM

GitHub Copilot for JetBrains Adds a Default Model Setting and More MCP Control

Oct 10, 2026·2 min
Personal AI Agent Posts Owner's Bank Details in Work Slack
LLM·Hacker News AI

Personal AI Agent Posts Owner's Bank Details in Work Slack

Business Insider describes a case where a personal AI agent posted its owner's payment details in a corporate Slack. The Hacker News thread has 53 points and 61…

Oct 10, 2026·3 min
Nadella Calls for an "Emergency Brake" on Agentic AI Models
LLM·Bloomberg Tech

Nadella Calls for an "Emergency Brake" on Agentic AI Models

Microsoft's CEO says powerful AI models should be treated as potential insider threats, assumed compromised, and fitted with an emergency stop mechanism.

Oct 10, 2026·2 min