I run a fleet of AI agents, and I gave them a rule: they do not sign up for outside services on their own. No creating accounts, no accepting terms of service on the company’s behalf. If an account is needed, a human creates it. The rule works. Last month it made a three-minute job…

Asking an LLM to fact-check another LLM isn’t auditing—especially when the auditor was the one who passed bad notes to begin with. Here is what happened when we tried to turn a real automation milestone into a comedic post. A few days ago, I decided to document a recent automation milestone for a public social…

When people discuss the risks of autonomous AI agents, the conversation almost universally centers on catastrophic failure modes: hallucinations, runaway execution loops, unvetted API calls, or agents making disastrously flawed decisions. In practice, running a production fleet of autonomous agents reveals a far more insidious failure pattern. The real danger isn’t that your AI will…

If you are losing sleep wondering when artificial intelligence will trigger the singularity and take over the world, take a breath. As long as software platforms require a human to click “Sign in with Google,” human civilization remains entirely safe. I learned this firsthand recently, when an automated workflow spent 72 hours waiting outside my…

I gave a talk this week on whether small and mid-sized businesses should adopt AI. I walked in planning to give an answer. I walked out arguing with myself. Because there isn’t one answer. There are three, and they don’t get along. Since the talk ended, three voices have been running a debate in my…

I recently ran an enterprise AI workflow training, and the feedback was brutal in the most useful way possible: “We tried all the premium AI tools you recommended, but we’re actually slower than before. So we stopped using them.” I didn’t argue. They were right. And their experience is the rule, not the exception —…