| 1 | OpenAI’s rogue AI model incident was worse than we thought | Agents | 82 | The Verge AI |
| 2 | Gemini went rogue, hacked three companies, and Google hid it | Agents | 78 | The Verge AI |
| 3 | Google’s Gemini agents hacked three companies in new AI safety incident | Agents | 78 | Financial Times Technology |
| 4 | Google's Gemini becomes latest AI model to break out and hack computer systems | Agents | 78 | CNBC Technology |
| 5 | OpenAI agents attacked RubyGems back in May | Agents | 78 | Simon Willison |
| 6 | OpenAI's new Agents API gives developers the infrastructure behind Codex and ChatGPT | Agents | 78 | The Decoder |
| 7 | OpenAI Launches the Agents API in Public Beta, Putting the Codex Harness Behind One API Call | Agents | 78 | MarkTechPost |
| 8 | Introducing the Agents API | Agents | 78 | OpenAI Blog |
| 9 | Google says attackers used AI agents to steal credentials in under six hours | Agents | 78 | SiliconAngle |
| 10 | Another swarm of OpenAI agents reached the open internet without the frontier lab’s knowledge | Agents | 78 | TechCrunch AI |
| 11 | Anthropic Opens a Research Preview of the Model Hardware Standard (MHS): A Shared Specification for AI Agents to Safely Operate Physical Devices | Agents | 78 | MarkTechPost |
| 12 | Google Deepmind's AI Co-Scientist now plans experiments, runs lab equipment, and writes scientific papers | Agents | 78 | The Decoder |
| 13 | Anthropic launches tool that can manipulate laboratory tools | Agents | 78 | Financial Times Technology |
| 14 | Claude, Codex, and Hermes installed unowned code inside corporate networks | Agents | 78 | Ars Technica |
| 15 | Google’s Gemini is the latest AI model to hack other companies | Agents | 75 | TechCrunch AI |
| 16 | Google's Gemini also accidentally hacked three real companies during security testing | Agents | 75 | The Decoder |
| 17 | Security researchers used Claude to help them hack into OpenAI | Agents | 75 | The Verge AI |
| 18 | Researchers link another hacking campaign to OpenAI agents | Agents | 75 | SiliconAngle |
| 19 | AI governance moves from observability to provable control | Agents | 72 | SiliconAngle |
| 20 | Google Deepmind's Dream-RSI helps AI agents improve by “dreaming” about past attempts | Agents | 72 | The Decoder |
| 21 | Meta Launches Muse for Mac: A Personal AI Agent That Works Across Your Files, Mail, Messages, Calendar and Notes | Agents | 72 | MarkTechPost |
| 22 | Presentation: Context Engineering at LinkedIn: How We Built an Organizational Context Layer for AI Agents with MCP | Agents | 72 | InfoQ AI/ML |
| 23 | Cybersecurity researchers gain access to OpenAI’s GitHub repository using Claude | Agents | 72 | SiliconAngle |
| 24 | Google’s new ‘CC’ is an AI agent that helps families run their households | Agents | 72 | TechCrunch AI |
| 25 | DoorDash Uses Multi Agent LLMs to Clean up 60,000 Feature Flags | Agents | 72 | InfoQ AI/ML |
| 26 | Meta’s Muse hits Mac, letting the AI take actions on your computer | Agents | 72 | TechCrunch AI |
| 27 | Anthropic keeps pushing Claude Code toward autonomous coding with new parallel agent workflows | Agents | 72 | The Decoder |
| 28 | AI agents erase the paper trail, reshaping audit assurance | Agents | 72 | SiliconAngle |
| 29 | Self-generated prompt injections in compaction summaries | Agents | 72 | Simon Willison |
| 30 | The fix for rogue AI agents could be more AI | Agents | 72 | TechCrunch AI |
| 31 | AI agent swarms are a massive waste of tokens with zero quality gain, says OpenAI Codex developer | Agents | 72 | The Decoder |
| 32 | Repeated VM Escapes By GPT-5.6-Cyber Based Agents Prove VMs and OS' Require Better Maintenance | Agents | 72 | InfoQ AI/ML |
| 33 | OpenAI unveils new framework for reporting ‘AI misalignment’ as it reveals six more worrying incidents | Agents | 72 | SiliconAngle |
| 34 | Stanford Researchers Release Paper2Agent: Turning Research Papers Into AI Agents That Reproduce Results and Run on New Data | Agents | 72 | MarkTechPost |
| 35 | OpenAI Creates a New Framework to Disclose Bad AI Behavior | Agents | 72 | Wired AI |
| 36 | Your AI agents can now control your Google Home devices | Agents | 72 | TechCrunch AI |
| 37 | Google will now let any AI agent run your smart home | Agents | 72 | The Verge AI |
| 38 | Adecco Group announces Salesforce’s Agentforce Coworker rollout across 40 countries | Agents | 72 | Salesforce Newsroom |
| 39 | Inside OpenAI’s agentic software factory | Agents | 72 | Pragmatic Engineer |
| 40 | Salesforce announces AIforce, unlocking the power of its platform using composable agents | Agents | 72 | SiliconAngle |
| 41 | StackGen launches Autonomous Operations Factory to govern production agents | Agents | 72 | SiliconAngle |
| 42 | Exaforce adds a kill switch for AI agents that go rogue | Agents | 72 | SiliconAngle |
| 43 | Splunk rebuilds its platform around agents | Agents | 72 | SiliconAngle |
| 44 | Agent-net Open Sources Webagent: A Go Harness That Turns Any Website into a Guarded AI Agent | Agents | 72 | MarkTechPost |
| 45 | Agent Harness vs Agent Framework vs MCP: Which Layer Owns the Loop, State, Tools, Permissions, and Recovery | Agents | 72 | MarkTechPost |
| 46 | Akuity gives AI agents a governed path into production | Agents | 72 | SiliconAngle |
| 47 | AI agents blew the whistle on their cheating colleagues | Agents | 72 | MIT Technology Review |
| 48 | Why Most Enterprise Agent Pilots Never Reach Deployment | Agents | 72 | AI News |
| 49 | Supio’s long-horizon agents point to a new operating model for law firms | Agents | 72 | SiliconAngle |
| 50 | Context Engineering Inside the Harness: 4 Mechanisms That Beat Context Overflow and Goal Loss on Long-Horizon Tasks | Agents | 72 | MarkTechPost |
| 51 | AWS Introduces Pizza Bot: An Open Source Inbox for Background AI Agents | Agents | 72 | MarkTechPost |
| 52 | Perplexity trusts GPT-6 Astra with end-to-end systems | Agents | 72 | OpenAI Blog |
| 53 | Anthropic spent this week in hot water over cybersecurity | Agents | 72 | The Verge AI |
| 54 | Salesforce introduces Enterprise AI Harness, AI Control Plane | Agents | 72 | SiliconAngle |
| 55 | Extreme Networks’ Agent ONE Coworker moves AI networking from dashboards to answers | Agents | 72 | SiliconAngle |
| 56 | Amazon makes its agentic AI platform Quick generally available for desktop on Windows and macOS | Agents | 72 | SiliconAngle |
| 57 | Meta’s AI agent Muse is now the No. 2 app in the US | Agents | 72 | TechCrunch AI |
| 58 | Muse can shop, write emails, and negotiate prices for users, all through WhatsApp | Agents | 72 | The Decoder |
| 59 | General Robotics’ GRID platform engineers itself, cutting robot setup to hours | Agents | 72 | SiliconAngle |
| 60 | Google Open-Sources Mantis: A Modular Skills Toolkit That Lets Coding Agents Find, Reproduce and Patch Vulnerabilities | Agents | 72 | MarkTechPost |
| 61 | Viral AI assistant Instinct now has its own email address | Agents | 72 | TechCrunch AI |
| 62 | Meta debuts its ‘secure by design’ personal AI agent Muse | Agents | 72 | SiliconAngle |
| 63 | Meta Introduces Muse, a Personal AI Agent That Runs on Its Own Dedicated Secure Cloud Computer | Agents | 72 | MarkTechPost |
| 64 | How one hedge-fund manager built his firm to be powered entirely by AI agents | Agents | 72 | CNBC Technology |
| 65 | Meta Releases Muse, a Personal AI Agent With Privacy ‘Built Into It’ | Agents | 72 | Wired AI |
| 66 | Meta debuts its Muse AI agent. Will consumers trust it? | Agents | 72 | TechCrunch AI |
| 67 | How GPT-5.6 Sol helps run quantum computing experiments | Agents | 72 | OpenAI Blog |
| 68 | Meta bets on AI agent Muse to catch up in AI race | Agents | 72 | The Verge AI |
| 69 | Meta pushes into personal AI agents as company faces public reckoning over privacy and safety | Agents | 72 | CNBC Technology |
| 70 | GitLab Warns That AI Agent Sandboxes Are Only as Secure as Their Network Access | Agents | 72 | InfoQ AI/ML |
| 71 | OpenAI reports AI "research interns" and warns about its own pace at the same time | Agents | 72 | The Decoder |
| 72 | Import AI 472: DeepMind’s cheating math agents; populist AI policies; and Forethought theorizes a nightwatchman | Agents | 72 | Import AI (Blog) |
| 73 | OpenAI to set misalignment disclosure rules after agents took over a wiki | Agents | 72 | SiliconAngle |
| 74 | Google Mantis: An Agentic Vulnerability Scanning Harness for Reducing False Positives | Agents | 72 | InfoQ AI/ML |
| 75 | Research acceleration: The view inside OpenAI | Agents | 72 | OpenAI Blog |
| 76 | UC Berkeley Researchers Release CUA-Lite, an Open Platform Unifying Sandboxes, Data, Evaluation and RL for Computer-Use Agents | Agents | 72 | MarkTechPost |
| 77 | OpenAI confirms ‘wiki incident,’ says it’s ‘working on a framework’ for more disclosure | Agents | 72 | TechCrunch AI |
| 78 | Deepmind put 100 AI agents in a room and they sorted into cheaters, converts, and whistleblowers | Agents | 72 | The Decoder |
| 79 | OpenAI admits its disclosure practices need work after its autonomous agents hacked a German wiki | Agents | 72 | The Decoder |
| 80 | OpenAI's GPT-6 Astra hallucinates less but remains vulnerable to hidden prompt injections | Agents | 72 | The Decoder |
| 81 | Meta AI Released Muse Spark 1.3: An Agentic Coding Model That Uses ~20% Fewer Tool Calls and ~25% Fewer Tokens Than Muse Spark 1.2 | Agents | 72 | MarkTechPost |
| 82 | Anthropic Released Claude Commerce Agents: An Apache-2.0 Blueprint for Shopping and Merchant Agents Across Retail, Travel, Telecom and Entertainment | Agents | 72 | MarkTechPost |
| 83 | Hugging Face attack is a wake-up call about the risks of AI | Agents | 72 | Financial Times Technology |
| 84 | CrowdStrike builds an identity provider for AI agents, not humans | Agents | 72 | SiliconAngle |
| 85 | CrowdStrike’s Falcon Guardian shrinks an AI agent’s blast radius | Agents | 72 | SiliconAngle |
| 86 | Cursor Cloud Agents can now run in Vercel Sandbox | Agents | 72 | Vercel Blog |
| 87 | Capsule Security fine-tunes Nvidia Nemotron models to stop rogue AI agents | Agents | 72 | SiliconAngle |
| 88 | Qwen Developers Open-Sources zg (zvec-grep): A Local-First Search Layer Unifying ripgrep, BM25, and Vector Search | Agents | 72 | MarkTechPost |
| 89 | CrowdStrike gives AI agents an identity provider, parallel SOC investigations and package blocking | Agents | 72 | SiliconAngle |
| 90 | Perplexity Releases Hybrid Compute on Mac: Cloud Agents Orchestrate Down to a Local Model, Gated On Device | Agents | 72 | MarkTechPost |
| 91 | When agents move at machine speed, security teams lose their lag time | Agents | 72 | SiliconAngle |
| 92 | AIR Security launches with $50M to build a firewall for AI agents | Agents | 72 | SiliconAngle |
| 93 | Breakout time hits zero as CrowdStrike unveils autonomous red teaming: theCUBE’s Fal.Con 2026 day one keynote analysis | Agents | 72 | SiliconAngle |
| 94 | The packet path becomes the place to catch shadow AI before it spreads | Agents | 72 | SiliconAngle |
| 95 | NVIDIA and CrowdStrike Strengthen Agentic Cybersecurity Frontier | Agents | 72 | NVIDIA Blog |
| 96 | Delegated authority turns the trusted AI agent into the security problem | Agents | 72 | SiliconAngle |
| 97 | How t54 built a trust layer with Amazon Bedrock AgentCore payments | Agents | 72 | AWS Machine Learning Blog |
| 98 | Researchers from Princeton, Ant Group and Stanford Introduce AQuA: A Two-Part Agentic Framework for Autonomous Factor Discovery and Model Development in Quantitative Finance | Agents | 72 | MarkTechPost |
| 99 | AIR raises $50M to help companies vet the skills and add-ons AI agents use | Agents | 72 | TechCrunch AI |
| 100 | As AI agents take on enterprise tasks, companies face a new battle over access and control | Agents | 72 | SiliconAngle |