Limits & Risks
What the technology cannot do, and the failures that follow from assuming it can.
20 articles · page 2 of 2
Limits & Risks
Interpretability has produced real findings and not the thing people wanted
· 3 min read
Limits & Risks
Using one model to grade another is convenient, and it inherits every bias of the grader
· 4 min read
Limits & Risks
People attribute a mind to these systems reliably, and the design does not discourage it
· 4 min read
Limits & Risks
A model can be corrupted during training in a way that shows up only on a chosen signal
· 4 min read
Limits & Risks
The expertise needed to supervise a system is built by doing the work it replaced
· 4 min read