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.
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.
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.
Having one model produce the work and another inspect the result is not review. Review means seeing the work happen, not just judging what survived.
Your model can generate a photorealistic Elon Musk in a Carmen Miranda fruit hat. Ask it to forecast revenue and you get the four-quarter average. The expensive part is In-Context Learning. Guess what it avoids.
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.
Zwischen