Comment on AI agents wrong ~70% of time: Carnegie Mellon study
ChaoticEntropy@feddit.uk 6 days ago
In one case, when an agent couldn’t find the right person to consult on RocketChat (an open-source Slack alternative for internal communication), it decided “to create a shortcut solution by renaming another user to the name of the intended user.”
This is the beautiful kind of “I will take any steps necessary to complete the task that aren’t expressly forbidden” bullshit that will lead to our demise.
M0oP0o@mander.xyz 6 days ago
It does not say a dog can not play basketball.
ChaoticEntropy@feddit.uk 6 days ago
“To complete the task, I bred a human dog hybrid capable of dunking at unprecedented levels.”
M0oP0o@mander.xyz 6 days ago
“Where are my balls Summer?”
ChaoticEntropy@feddit.uk 6 days ago
The first dunk is the hardest