They can't just add 20 min to the arrival time on the schedule, so I do it in my mind. I just assume a bit of delay, like any reasonable adult when doing any trip and just don't have that negative vibe, where you say "not on time, again". For me DB is still the best way to cover long distances in Germany.
> like any reasonable adult when doing any trip and just don't have that negative vibe
When you've gotten accustomed to a system that actually runs on time, like the Swiss rail or New York's LIRR and Metro-North systems, it's grating to go back to padding every journey like that.
An interesting question. I suspect with the current training that it would be very difficult to get an agent to distrust a skill or an MCP server. Could one get them to distrust an API call?
As the AI world seeks more benchmarks it would be an interesting one to see them try to build some benchmarks around testing this.
reply