logoalt Hacker News

mhl47yesterday at 5:14 PM5 repliesview on HN

First test question: "Is the UV Index a good proxy for when to wear sunglasses." Immediately triggered the safety filter ... oh dear.


Replies

msp26yesterday at 6:40 PM

It triggered for me when I asked "Web search for your own model card (released today) and pick out your favourite highlights from the pdf"

aix1yesterday at 5:23 PM

Did not trigger for me (Fable answered the question), so I guess the filters are either non-deterministic or are still being tweaked.

show 1 reply
ijidaktoday at 2:26 PM

Don't worry. They're just leaving the door open for OpenAI and other model makers.

They'll relax these safeguards once competition increases.

Narretzyesterday at 6:07 PM

Iirc correctly Opus 4.7 had the same problem, safety filters were triggered way too easily at the beginning.

Eduardyesterday at 9:28 PM

sunglasses _are_ safety filters