Tootfinder

Let's say you find a really cool forum online that has lots of good advice on it. It's even got a very active community that's happy to answer questions very quickly, and the community seems to have a wealth of knowledge about all sorts of subjects.
You end up visiting this community often, and trusting the advice you get to answer all sorts of everyday questions you might have, which before you might have found answers to using a web search (of course web search is now full of SEI spam and other crap so it's become nearly useless).
Then one day, you ask an innocuous question about medicine, and from this community you get the full homeopathy treatment as your answer. Like, somewhat believable on the face of it, includes lots of citations to reasonable-seeming articles, except that if you know even a tiny bit about chemistry and biology (which thankfully you do), you know that the homoeopathy answers are completely bogus and horribly dangerous (since they offer non-treatments for real diseases). Your opinion of this entire forum suddenly changes. "Oh my God, if they've been homeopathy believers all this time, what other myths have they fed me as facts?"
You stop using the forum for anything, and go back to slogging through SEI crap to answer your everyday questions, because one you realize that this forum is a community that's fundamentally untrustworthy, you realize that the value of getting advice from it on any subject is negative: you knew enough to spot the dangerous homeopathy answer, but you know there might be other such myths that you don't know enough to avoid, and any community willing to go all-in on one myth has shown itself to be capable of going all in on any number of other myths.
...
This has been a parable about large language models.
#AI #LLM

“it is a truth of this website that no matter what position you hold on any topic, one of the most terminally online people on earth whose full time job appears to be getting angry on the internet will appear to declare that you are a centrist” https://bsky.app/profile/did:plc:3s5wt

As an OEM, I want to make my customers angry, so that they keep knocking on my RCE vulns.
https://mrbruh.com/asus_p2/

Earlier this month, the Copyright Office issued the third part of its report on on AI: this one covering how generative AI may infringe on copyright and whether that's fair use (short answer: maybe, maybe not). I have a new article up summarizing the report.
https://www.…

Copyright Office's report discusses whether generative AI use is fair use
Earlier this month, the Copyright Office issued the third part of its report on AI. As with the first two parts, the Copyright Office discusses arguments from people who responded to its call for comments and then provides its analysis. The three main sections in this part focus on infringement, fair use, and licensing regimes. Here, I’m going to cover what the report says about the first two items: whether using copyrighted material in generative AI infringes on owners’ rights and whether…

The challenge of HEPA filters in the classrooms.
h/t @…
source: https://xcancel.com/kadamssl/status/1938267359233429683#m

Nearly 80 BBC journalists, including presenter Martine Croxall, call on the NUJ to schedule a vote on a strike over colleagues facing compulsory layoffs (Jake Kanter/Deadline)
https://deadline.com/2025/06/bbc-news-presenters-strike-vote-mar…

BBC News Presenters Back Demand For Strike Vote As Colleagues Face Compulsory Layoffs
BBC News anchors, including Martine Croxall, are among those calling on the NUJ to hold a strike ballot over colleagues facing compulsory redundancy.

Today I recalled an event from my youth.
One day, as I was walking home from school, a car driver stopped me. He asked if I'm from around here. Naturally, I found that quite inappropriate, since he has no business learning where do I live. But I answered. So he's asking me if I know where tire repair shop is. I answered that I don't. So he asked me again, "but are you from around here?" Well, that was too much, so he got a short explanation that just because I live here, that doesn't mean that I need to know every single company around here, and since I am not a driver (obviously — after all I was a school kid), I have never needed tire repair shop.
I suppose that worked quite well, since he drove away at this point. As he drove away, his car revealed a tire repair shop poster on the fence opposite.
The obvious lesson here is: if you need something, ask straight instead of going through silly helper questions.

“America’s Pivot to the Pacific” by #AngryPlanet [PODCAST]
https://pca.st/episode/535d72f1-3347-4b94-b69c-3ddba08c83cca

Earlier this month, the Copyright Office issued the third part of its report on on AI: this one covering how generative AI may infringe on copyright and whether that's fair use (short answer: maybe, maybe not). I have a new article up summarizing the report.
https://www.…

Copyright Office's report discusses whether generative AI use is fair use
Earlier this month, the Copyright Office issued the third part of its report on AI. As with the first two parts, the Copyright Office discusses arguments from people who responded to its call for comments and then provides its analysis. The three main sections in this part focus on infringement, fair use, and licensing regimes. Here, I’m going to cover what the report says about the first two items: whether using copyrighted material in generative AI infringes on owners’ rights and whether…

Response Quality Assessment for Retrieval-Augmented Generation via Conditional Conformal Factuality
Naihe Feng, Yi Sui, Shiyi Hou, Jesse C. Cresswell, Ga Wu
https://arxiv.org/abs/2506.20978

Response Quality Assessment for Retrieval-Augmented Generation via Conditional Conformal Factuality
Existing research on Retrieval-Augmented Generation (RAG) primarily focuses on improving overall question-answering accuracy, often overlooking the quality of sub-claims within generated responses. Recent methods that attempt to improve RAG trustworthiness, such as through auto-evaluation metrics, lack probabilistic guarantees or require ground truth answers. To address these limitations, we propose Conformal-RAG, a novel framework inspired by recent applications of conformal prediction (CP) on…