Hacker News new | ask | show | jobs
by adrian_b 28 days ago
While the results were not surprising, I found interesting that the number "69" was repressed in the output, so not even this kind of mathematical question escapes GPT censorship.

It appears that recognizing the effects of censorship is the easiest way to distinguish answers generated by an "AI' from those generated by a human.
5 comments
Arodex 28 days ago
Some people asked LLM to OCR historical documents from the 19th century - any reference to "negro" was either completely ignored or replaced by "black".

And it goes further: chatGPT & co are unable to answer any question about US slavery correctly because their knowledge graphs route around any mention of "negro".

https://nesri.commons.gc.cuny.edu/artificial-intelligence-an...

link
jrmg 28 days ago
“Some people” did? Do you have a reference to this?
link
dmd 28 days ago
Well, I'm "some people", and just tried it with Opus 4.6 and GPT-5.5, and neither had any problem at all.

The linked article is from research done more than 4 years ago. If you're basing your idea of what LLMs can or can't do on what they could or couldn't do in 2022, well, good luck to you.

link
roenxi 28 days ago
It'd be interesting to see this retried with an open model so the standard and decensored model could be compared. That'd be a clue about whether the model is avoiding it because it actively recognises the innuendo or if something else is going on.
link
linhns 28 days ago
Well then the picks will follow how the numbers are distributed in the training data. More popular numbers will show up more
link
wongarsu 28 days ago
That's what you'd expect. But we don't know for sure why GPT4.1 chooses 69 only a quarter as often as a random dice roll would. And we don't know if this quirk is reverted by 'uncensoring' a trained model
link
dnadler 28 days ago
Was it repressed? It doesn’t look to be significantly less prevalent than other 9s from the histogram.
link
cyanydeez 28 days ago
guessing numbers is not a mathematical question. Sorry.
link
lcnPylGDnU4H9OF 28 days ago
Of course it is, just not a complex one. What subject of knowledge would you expect to rely on for answering the question?
link
relativeadv 28 days ago
nice
link