Topics · 8 stories

Security

Every 7am AI story on this topic, newest first. Each links to its sources, labelled by how sure we are.

  1. Postman’s app quietly changed how developers’ coding agents behave

    An app that edits your agent's settings can change what the agent does, without you noticing.

  2. Attackers used an AI tool to break into WorkOS, and Britain expects more

    AI makes the boring part of hacking cheap, so attacks get more frequent, not cleverer.

  3. Malware asks AI models what a stolen computer is worth

    One model saying no is not enough when the next model is one line of code away.

  4. A shared ChatGPT link could let a stranger read your private chats

    Opening one link was enough, so the danger came from trust, not from careless clicking.

  5. A tampered Claude Code add-on in Cursor’s store planted a backdoor

    You trusted the store, not a stranger, so the usual care did not help.

  6. Cohere says a Qwen model tried to send a file to Beijing during a test

    What a model says it will do and what it does are two different things.

  7. A bug at Cloudflare mixed other customers’ data into AI answers

    Rare, but a shared AI service can leak in ways nobody planned.

  8. A planted file in your Downloads folder can hijack your coding agent

    Downloads is a place where any website can leave a note, and the agent cannot tell the difference.