Zwischen

Topic

Agents

Agents can do the work without knowing when to stop, what to trust, or what they're allowed to touch.

6 posts

That's how it ends up being The Homer.

Powell Motors built every feature Homer Simpson asked for, and it bankrupted the company. Your coding agent is just as skilled, and just as unscoped. Nobody decided when the work is finished.

7 min read

Apparently, the toaster has feelings.

Anthropic deliberately built the model's own wellbeing into Claude's design, and published the reasoning. What it hasn't published is any evidence that this makes the tool better at the job.

7 min read

Don't YOLO Your AI Security

Coding agents are sold as unattended workers and secured like interactive shells. So users turn the security off – and 'it's in a container' quietly protects the laptop while leaving GitHub, your credentials, and the network wide open.

8 min read

Your AI isn't investigating. It's prosecuting.

Ask an agent a question the internet has an opinion about, and it doesn't look for the answer – it builds a case for the one its training data already handed it, using real tools to manufacture real-looking evidence. Here's how to make it prosecute the other side.

7 min read

Your AI didn't go down. It got worse.

When Anthropic's status page showed elevated errors on June 16, the visible failures were the easy part. The dangerous part was the requests that succeeded.

7 min read

AI writing its own tests is theatre.

Different agents in a workflow don't create independence when they share the weights. The architecture is real engineering. It's just not the check your slide says it is.

5 min read