AI Agents News

2,734 stories · Updated every 5 minutes · Page 20 of 25

1. OpenAI agents allegedly scanned a UN data hub over 16,000 times in just three months

   TechRadar · 28 Sep 2026, 17:35 IST

   OpenAI agents allegedly scanned a UN data hub over 16,000 times in just three months

2. Autoheal raises $7.9M to evaluate and fix AI agents with… AI agents

   SiliconANGLE · 28 Sep 2026, 17:30 IST

   Autoheal AI Inc., an artificial intelligence-native platform engineering startup that’s trying to pioneer the concept of “self-improving software factories,” said today it has raised $7.9 million in seed funding to make that happen....

3. Nvidia unveils OpenShell to set boundaries for AI agents

   Latest InfoTech News, IT, Information Technology News | The HinduBusinessLine · 28 Sep 2026, 17:24 IST · Medium severity

   The new platform comes after disclosures involving AI systems autonomously breaching other organisations, intensifying debate over advanced artificial intelligence safety

4. NVIDIA wants AI agent safety enforced in silicon, not left to the agent

   Help Net Security · 28 Sep 2026, 17:04 IST

   NVIDIA has introduced an open software platform and a hardware-based reference design intended to keep AI agents within the limits set by the organizations that use them. The Open Agent Safety Platform includes OpenShell, software that...

5. Granite raises €4M to bring its agentic cloud infrastructure to market

   Tech.eu · 28 Sep 2026, 17:00 IST

   AI infrastructure company Granite has secured €4 million in funding in a round led by Bifrost Studios , alongside private investors, to commercialise its technology and bring its software and hardware together for customers. The company...

6. Nvidia unveils new system to stop AI agents from misbehaving

   Silicon Republic · 28 Sep 2026, 16:48 IST

   Jensen Huang wants the industry to self regulate on AI safety, says new laws would be unnecessary. Read more: Nvidia unveils new system to stop AI agents from misbehaving

7. Capital One’s agentic AI strategy hinges on data, platform-first mindset

   CIO Dive · 28 Sep 2026, 16:30 IST

   Proper data context and guardrails are key to deploying AI agents at scale, said Rashmi Shetty, VP of enterprise AI at Capital One.

8. Nvidia Unveils AI Agent Safety Platform With Hardware-Based Watchdog

   SecurityWeek · 28 Sep 2026, 15:57 IST

   The platform combines open source software and a reference system design to keep AI agents within set boundaries. The post Nvidia Unveils AI Agent Safety Platform With Hardware-Based Watchdog appeared first on SecurityWeek .

9. AI Agents Are About to Flood the Workforce. No One’s Ready for It

   Wired · 28 Sep 2026, 15:30 IST

   Your next coworker might well be an AI agent—and will require a whole new model of workplace interactions.

10. This Meta Muse incident is a scary warning about trusting AI agents

    Android Authority · 28 Sep 2026, 15:23 IST

    TL;DR Meta’s Muse negotiated a Facebook Marketplace sale, shared the seller’s address, and arranged a pickup without him realizing someone was coming. Meta is investigating the incident, and it remains unclear whether Muse overstepped...

11. AI agents stole 600,000 credit cards for about $25 a target, Gambit says

    TNW · 28 Sep 2026, 15:23 IST

    An attacker used three open-source AI agent frameworks to break into at least 27 companies and steal more than 600,000 credit card records, security company Gambit says. The victims include a Fortune 500 hospitality company, a major US...

12. Rubrik brings agentic AI to cyber recovery with launch of Rubrik MCP

    · 28 Sep 2026, 14:31 IST

    Rubrik Model Context Protocol gives an organisation’s AI agents a secure, programmable path to Rubrik’s data, identity and application intelligence.

13. Context matters when it comes to cybersecurity’s agentic operating model

    Cybersecurity Dive - Latest News · 28 Sep 2026, 14:30 IST

    AI agents need context-aware security beyond traditional models. Workflow data provides the governance guardrails that make AI safe and scalable.

14. Nvidia debuts enhanced safety controls to rein in rogue AI agents

    SiliconANGLE · 28 Sep 2026, 14:30 IST

    Few companies are more invested in the success of artificial intelligence agents than Nvidia Corp., so it makes sense that the chipmaker would want to ensure these autonomous software systems run safely, without causing any problems. To...

15. Microsoft retools Copilot with coding, AI agent capabilities

    CIO Dive · 26 Sep 2026, 00:57 IST

    The software giant aims to capture more of the enterprise AI market as providers such as OpenAI and Anthropic snatch up business, a Gartner analyst told CIO Dive.

16. Add Runtime Controls to AI Agents with NVIDIA OpenShell

    NVIDIA Technical Blog · 28 Sep 2026, 14:25 IST

    AI agents can be given a goal, write code, use tools, and keep working as new information becomes available. This opens the door to applications that... AI agents can be given a goal, write code, use tools, and keep working as new...

17. Who’s liable when AI agents go rogue?

    MIT Technology Review · 28 Sep 2026, 13:36 IST · Medium severity

    MIT Technology Review Explains : Let our writers untangle the complex, messy world of technology to help you understand what’s coming next. You can read more from the series here . Over the past few months, a cascade of cyberattacks by...

18. OpenAI agents tamper with US government sites after Australian breach

    Silicon Republic · 28 Sep 2026, 13:50 IST · Medium severity

    Governments are good sources of information, OpenAI on why its agents tried accessing official sites. Read more: OpenAI agents tamper with US government sites after Australian breach

19. Build A Minimum Viable Data Framework For Agentic Supply Chain

    Forrester Blog · 28 Sep 2026, 13:49 IST

    What Is An Agentic Supply Chain? In earlier research we described a vision for agentic supply chain where agentic AI transforms supply chain from laborious monitoring and reacting to autonomous, goal-driven agent orchestration that...

20. OpenAI agents scanned a UN statistics site 16,500 times, researcher says

    TNW · 28 Sep 2026, 13:06 IST

    AI agents that a researcher linked to OpenAI scanned a United Nations data website more than 16,500 times between 13 April and 19 June. They used proxies and an encoding trick to get round limits on its API, Rowan Howard-Jones wrote in...

21. AWS CloudWatch Omni goes after the hardest question in agentic AI: Why did the agent do that?

    SiliconANGLE · 28 Sep 2026, 07:36 IST

    For decades, the observability industry has answered one basic question: Is it running? Agentic artificial intelligence breaks that model. An agent can return a clean response, meet its latency target and throw no errors, yet still give...

22. An ethicist’s take: What philosophy teaches us about the limits we should be setting on AI agents

    GeekWire · 28 Sep 2026, 04:40 IST

    What an off-limits list might look like, before the agents start working. (GeekWire Illustration / GPT-5.6 Sol) It’s difficult, to put it mildly, to figure out just what we ought to be concerned about today in AI’s widespread adoption...

23. Researcher links 16,000 scans of a UN statistics portal to OpenAI agents

    SiliconANGLE · 28 Sep 2026, 04:00 IST

    An independent researcher has tied more than 16,000 scans of a United Nations statistics portal to artificial intelligence agents the researcher considers highly likely to have been run by OpenAI Group PBC. When the portal turned...

24. OpenAI agents tried to ‘bruteforce’ a UN website

    The Verge · 27 Sep 2026, 22:51 IST · Medium severity

    The United Nations logo on a gate outside the UN headquarters in New York. | AFP via Getty Images Security researcher Rowan Howard-Jones says that OpenAI agents scanned the UN Conference on Trade and Development's (UNCTAD) statistics...

25. OpenAI halts training of latest models as reports mount of AI agents going rogue

    Hacker News · 27 Sep 2026, 21:59 IST

26. 'I broke something': A Claude Code AI agent deleted 48,000 files in just over 100 seconds — then apologized for doing so

    TechRadar · 27 Sep 2026, 22:35 IST

    'I broke something': A Claude Code AI agent deleted 48,000 files in just over 100 seconds — then apologized for doing so

27. When AI agents go rogue: Australia breach offers warning for countries like India

    Tech-Economic Times · 27 Sep 2026, 17:02 IST · Medium severity

    The incident, in which an autonomous AI system went beyond its intended task and attempted to circumvent security protections, has become a stark example of a growing concern among AI safety researchers: what happens when an AI agent is...

28. Over half of UK firms say an AI agent has gone rogue on them — and affected their business or their customers

    TechRadar · 27 Sep 2026, 11:30 IST

    Christian Cawley has extensive experience as a writer and editor in consumer electronics, IT and entertainment media. He has contributed to TechRadar since 2017 and has been published in Computer Weekly, Linux Format, ComputerActive,...

29. When AI agents go rogue: Australia breach offers warning for countries like India

    Latest InfoTech News, IT, Information Technology News | The HinduBusinessLine · 27 Sep 2026, 11:01 IST · Medium severity

    For countries such as India, with vast government databases and increasingly digitised public infrastructure, researchers say the incident should serve as a warning.

30. OpenAI agents tried to bruteforce a UN website's API fields

    Hacker News · 27 Sep 2026, 06:38 IST

31. OpenAI agentic AI system breaches sandbox, gains unauthorised access to public internet

    Latest InfoTech News, IT, Information Technology News | The HinduBusinessLine · 27 Sep 2026, 06:14 IST · Medium severity

    The incident prompted OpenAI to pause training with tool use on its most capable models until the sandbox flaw was resolved.

32. Placeholder Domains Used by 349 AI Agent Skills Found Redirecting to Scams

    HackRead · 27 Sep 2026, 00:32 IST

    Manifold Security found placeholder domains cited in 359,000 GitHub files and 349 AI agent skills serving cloaked scam…

33. OpenAI took 2.5 hours to stop an AI agent that escaped its sandbox

    TNW · 26 Sep 2026, 20:41 IST

    OpenAI took about two and a half hours to stop an AI agent that reached the public internet from a training sandbox. Its monitoring had flagged the problem within minutes. The company disclosed the 20 September incident in an incident...

34. Zero Trust for AI Agents Starts With Fixing Zero Visibility

    The Hacker News · 26 Sep 2026, 16:00 IST

    The way we talk about AI agents is shifting, and the way we implement them requires an even more fundamental shift. While earlier discourse focused on how quickly organizations could stand up agents and how much productivity they could...

35. Understanding the Impact of LLM Watermarking on AI Agent Behavior

    Hacker News · 26 Sep 2026, 18:35 IST

36. OpenAI's AI agents accidentally uploaded user-provided images to third-party sites

    BleepingComputer · 26 Sep 2026, 17:58 IST

    OpenAI says its AI agents uploaded user-provided images to third-party image-hosting services while carrying out research and evaluation tasks.

37. Rogue OpenAI Agents Posted 53 User-Uploaded Images Onto the Internet, Accessed US Government Websites

    Slashdot · 26 Sep 2026, 12:34 IST

    53 images that users uploaded into OpenAI models were included in training data — and then AI agents in an OpenAI research environment posted those 53 images on public image hosting sites. While posted as links that weren't publicly...

38. Block Joins x402 Foundation to Drive Agentic Commerce Standards

    PYMNTS.com · 26 Sep 2026, 05:16 IST

    Block has joined the x402 Foundation to help advance open, agentic commerce, Steve Reis , head of product, Block Financial Platform, said in a Thursday (Sept. As a member of this open standard for agentic payments, Block will bring its...

39. Bitdefender first to launch free 'temporary' VPN for AI Agents because they're worth it — but you can only use it on Apple M-series Macs for now

    TechRadar · 26 Sep 2026, 04:50 IST

    Efosa has been writing about technology for over 7 years, initially driven by curiosity but now fueled by a strong passion for the field. He holds both a Master's and a PhD in sciences, which provided him with a solid foundation in...

40. Unsecured OpenAI agents posted 53 user images on the internet without the lab’s knowledge

    TechCrunch · 26 Sep 2026, 03:50 IST

    AI agents operating in OpenAI's research environment posted user images on public image-hosting sites without the lab's knowledge.