1·
15 days agoIf that was true, wouldn’t every AI get the answer wrong? It’s actually around 50/50. The leading “reasoning” models almost always get it right, the others often don’t.
- 0 Posts
- 31 Comments
Joined 2 years ago
Cake day: January 18th, 2024
You are not logged in. If you use a Fediverse account that is able to follow users, you can follow this user.
- Spends the first 90% of the competition developing specialized subagents and custom MCP servers to allocate the problems and most relevant information efficiently into the LLM’s contexts.
- All of his agents easily escape their own sandboxes and one accidentally configures itself into “delete-only mode”.
- “Codex, how the fuck do you not have access to your own documentation?”
- Places 29th globally after one of his subsubagents finds a way to reconstruct the full solution set from filesystem metadata in the online judge VMs.
Isn’t that like $900 worth of IPv4 addresses?
21·7 months agoI’ve found online feedback useful. You just have to be careful about where you get it and take it with a grain of salt. A very large one.
9·7 months agoI swear to god this is true. The recruiter said it was my personality. I didn’t even ask.
divulgâche
They were actually quite nice about it and I was happy to get the feedback.
Newton so we could talk about both being life-long virgins.
Why would anyone choose to know that?
My take away is that it’s mainly children who are still using the free version of ChatGPT. Surely everyone else has moved on to better models.
If you want to know what people are typing into chatbot sites, here’s 140,000 examples: https://huggingface.co/datasets/lmarena-ai/arena-human-preference-140k. It’s mostly nonsense.
Is there any altruistic act you couldn’t apply that logic to?
1·9 months agodeleted by creator
“…” (Unicode U+2026 Horizontal Ellipsis) instead of “…” (three full stops), and using them unnecessarily, is another thing I rarely see from humans.
Edit: Huh. Lemmy automatically changed my three fulls stops to the Unicode character. I might be wrong on this one.
This article is written in such a heavy ChatGPT style that it’s hard to read. Asking a question and then immediately answering it? That’s AI-speak.
My bad. I assumed this was about regular split tunneling, not inverse, which I had never heard of.
1·10 months agoBed.
Mullvad has that now. It usually works.
I never thought of it that way. That’s kind of nice.
The realization that if I won the lottery I would browse 4chan all day
I want to say upfront that I’m not trying to defend AI here. I wouldn’t be on Fuck AI if I wanted to do that. I just think it’s philosophically interesting despite causing way more problems than it solves.
I copied the message from the image verbatim.
About 50% of the models I tried got it right. (Don’t worry, I didn’t pay the AI companies for that or give them feedback or anything.)
The question from the image.
My question was how do you then explain some models getting the question right?
It’s usually the more advanced ones that get it, so it’s possible that a similar enough question is in the training data somewhere and the only difference is that the advanced models are large enough to encode it. The question in the image has been around since at least 2023.
So let’s try making our own question, taking a well-known trick question and subtly inverting it so it becomes a kind of double bluff.
It’s hard to google, for obvious reasons, but I couldn’t find anyone trying this question like I could with the question from the image. But I got similar results with the AI models.
They actually did slightly better on this one. About 60-70% got it right.
I’ve tried a few different types of questions, over the last few years, to see what AI gets wrong that humans get right. What I’ve found so far is that AI has been a lot dumber than I had expected, but humans have also been a lot dumber than I had expected.
To be honest, the gap was far wider for the humans. My theory is that COVID gave us all brain damage.