23 points
lazily regex
I’m having a sneaking suspicion that this is what they do for all the viral ‘here the LLM famously says something wrong’ problems, as I don’t think they can actually reliably train the model it made an error.
14 points
That’s the most straightforward fix. You can’t actually fix the output of an LLM, so you have to run something on the output. You can have it scanned by another AI but that costs money and is also fallible. Regex/delete is the most reliable way to censor.
11 points
Yes, and then the problem is that this doesn’t really scale well. Esp as it is always hard to regexp all the variants correctly without false positives and negatives. Time to regexp html ;).
8 points