Category: Blog

  • The AI That Ratted Itself Out: How One Click Could Empty Your Gmail Through Copilot

    ⋅

    Ng Chong

    This is about Microsoft Copilot Personal, the free consumer chat assistant at copilot.microsoft.com, not GitHub Copilot (the coding autocomplete tool) or Microsoft 365 Copilot for work and school accounts. The enterprise version has had its own separate, since-patched issues, but this piece focuses on the consumer product. If you wanted to break into a locked…

  • Provenance, Not Proof: What Claude’s Watermark Actually Tells You

    ⋅

    Ng Chong

    Anthropic’s decision to embed machine-readable marks in Claude’s output, announced this week to comply with the EU AI Act’s Article 50(2) Code of Practice, is being covered as a broad AI-transparency milestone. It’s narrower than that framing suggests, and the narrowness is itself the more interesting story. Here’s what the policy actually says, how it…

  • Why the same LLM hangs on one provider and flies on another: cold starts and keep-warm

    ⋅

    Ng Chong

    One of our chat apps lets you choose which model answers you, and for most of those choices it was fine. Point it at DeepSeek running on Azure, though, ask a reasoning-heavy question, and the screen would freeze for anywhere from a few seconds to most of a minute before the first word appeared. The…

  • Project Perception Is in Public Preview. Now Comes the Evidence Test.

    ⋅

    Ng Chong

    Microsoft’s new agentic security platform promises to find risks, investigate threats, and help remediate them with coordinated AI agents. The question now is whether the evidence matches the ambition. Microsoft’s Project Perception is a new AI-driven cybersecurity platform designed to automate large parts of the security operations lifecycle. Rather than acting as a chatbot or…