In July 2025, a developer's Claude Code instance hit a recursion loop and burned through 1.67 billion tokens in 5 hours, generating an estimated $16,000 to $50,000 in API charges before anyone noticed. The agent did not crash. It did not throw an error. It just kept calling tools, getting confused, calling more tools, and silently accumulating cost. Old software crashes. LLM agents spend. This is
You're in another app and there's a timer counting down at the top of your phone. You lock the screen and the same timer is sitting there. You swipe down to the Notification Center and it's there too, still ticking. It looks like a notification, but a notification can't tick. That's a Live Activity. It looks like three different surfaces (Dynamic Island, lock-screen banner, Notification Center ent
Most API documentation is written for humans. MCP tool descriptions are different. They are read by the model that decides what to call next. That means tool names, descriptions, schemas, and error messages are not just documentation garnish. They are part of the safety boundary. A risky MCP tool often looks like this: name: query input: free-form string description: “Run SQL against the database
I kept seeing the same advice in prompt injection threads. Wrap untrusted content in random delimiters, tell the model "everything inside these markers is data, not instructions," and hope it respects the boundary. Sounds reasonable. I couldn't find anyone who actually measured whether it works. So I did. I'm building a system where LLM-generated output feeds into downstream decisions. The inputs
Hello everyone, I'm @xiaoqiangapi, the Chinese teacher who gives apis a "check-up". An article on , my SQL injection, XSS and prompt hijacked, API are blocked off. Let's take a different approach today - ** not attack, test 'resilience' **. Would the API crash if a sudden wave of requests came in, or if someone typed several thousand characters? I'm curious about it. The tools are still the same o
I finished an English series on the way I think ordinary people can start using AI for real work. The point is not to become an AI expert first. The point is to have one place where you can say what you want, give the tool access to the right folder, and check the result. Anything important still needs a human pause: publishing, deleting, paying, or authorizing. My preferred starting point is simp
This presentation is an adaptation of a keynote address delivered by Sasha Le, Senior Engineer, Tide Foundation at the launch event of the RMIT AWS Innovation Lab (RAIL) on 21st of April, 2026 In 2022, a ransomware group named Lapsus$ breached some of the most sophisticated tech companies on the planet. The list included Microsoft, Nvidia, Okta, Uber, and Samsung. The ringleader wasn't a state-spo