saw someone use grok today to correct the pope

this paper is the best thing I've read thus far on LLMs-in-practice; there is presently little substantive work in this area but this paper is a great start

(authors Gabriel Alracas and Donato Ricci)

hal.science/hal-05421828...

To be entirely clear: every "AI" company should have bright red ethical lines against things like genocide and weapons systems and surveillance and consent violations and shepherding an encroaching fascism, and those lines should be placed way earlier than anthropic and openAI's seem to be.

i think lots of people underestimate how badly these “AI” companies want people to be unable to afford computers. they view personal computing as a roadblock to their desire for complete end-to-end centralization. the fact that their tech is putting hardware out of reach is not just a happy accident

Image of the Yeti in the skiing Windows game SkiFree eating the player

Breaking: Tragedy at the Winter Olympics

Feck's sake - you can tell the people who make decisions about public transport in Cork city never have to actually use it.

And haven't for years.

“People don’t understand what this job is,” he said, adding, “You cannot pick and choose, because among the very rich is a higher percentage of unpleasant and not very attractive people. Capitalism is a rough system.” In defending his pursuit of Mr. Epstein and candidly describing the burden of raising money, Dr. Botstein gave insight into how the need to attract dollars can appear to run headlong into an academic institution’s stated values. Mr. Epstein gave prolifically to many charities and universities, including Harvard and M.I.T., and the Wall Street Journal report showed that his network of contacts was wider than had been thought, including prominent figures like the linguist Noam Chomsky and Lawrence Summers, the former Treasury secretary and president of Harvard. Dr. Summers sought money for a poetry foundation, led by his wife, Elisa New, a Harvard literature professor. Mr. Summers declined to comment. “Would we accept money from Jeffrey Epstein today? No,” Dr. Botstein said, describing the former donor as a “monster” and “truly evil man.” “We had no idea, the public record had no indication, that he was anything more than an ordinary — if you could say such a thing — sex offender who had been convicted and went to jail.”

As more academic names keep coming up in the Epstein files, I keep thinking about this NYT interview with the Bard president from a few years ago, and wondering how many people were rationalizing in similar ways.

www.nytimes.com/2023/05/05/u...

This pretty much nails what underlies all the hype about sentient AIs.

read online that someone set-up a premium rate phone number where callers get charged $3.99/ min. then they posted flyers in their neighborhood that said "call this number to report illegal aliens," but the number just puts them on hold with music. so far, they've made $17,000

Ed tech is profitable. It is also mostly useless Independent research identifies few learning gains

Blistering piece on ed tech in @economist.com.

‘Although ed-tech companies tout huge learning gains, independent research has made clear that technology rarely boosts learning in schools—and often impairs it.’
economist.com/united-state...

When Elon changed the name of Twitter to X everyone was like "what the fuck, that sounds like the name of a website for extremely illegal pornography made by and for nazis"

And behold what it became.

Nominative determinism go ew

UCC launches new toolkit to combat engagement with AI deepfakes A programme developed by researchers at University College Cork has found that educating individuals about the harms of AI identity manipulation can help to stop this problem at source. Rachael spoke ...

Reducing harmful engagement with AI-generated explicit images

UCC researchers have launched the world's first educational tool which aims to curb the watching, sharing, and creation of AI-generated explicit imagery.

UCC researchers unveil 'world-first tool' aimed at persuading Grok users to avoid creating explicit images Free 10-minute intervention dubbed ‘Deepfakes/Real Harms’ has been designed to reduce the willingness of users to engage with harmful uses of deepfake technology, like creating non-consensual explicit...

"The free 10-minute intervention dubbed ‘Deepfakes/Real Harms’ has been designed to reduce the willingness of users to engage with harmful uses of deepfake technology, like creating non-consensual explicit content", research by John Twomey supported by Lero
www.irishexaminer.com/news/arid-41...

UCC launches toolkit to combat AI deepfake engagement Researchers at University College Cork (UCC) have developed a new online educational toolkit designed to reduce engagement with AI-generated explicit images.

PAT research led by John Twomey on addressing engagement with deepfake tools is featured in the media today! @uccresearch.bsky.social @ucc-apppsych.bsky.social @lerocentre.bsky.social www.rte.ie/news/munster...

Politicians: don't use AI to make your maps.

Do You Want Your Food Made by a Robot? E-commerce mogul Marc Lore's Wonder is leaning into automation in a $2 billion effort to do to restaurants what Amazon did to shopping.

Every future imagined by a tech company is worse than the previous iteration.

Are you thinking about your 2026 CAO choices? Have an interest in Psychology? Students from UCC PsychSoc have put together an extremely informative introduction to the School of Applied Psychology in UCC! #caochoices #cao #psychology #leavingcert

A very long and important THREAD -

The real reason the US is invading Venezuela goes back to a deal Henry Kissinger made with Saudi Arabia in 1974.

And I'm going to explain why this is actually about the SURVIVAL of the US dollar itself.

Not drugs. Not terrorism. Not "democracy." 1/

If you are thinking of following suit, here's a list of EU tech alternatives:

european-alternatives.eu

if you are talking today about how bad maduro was and were alive, at all, during the run-up to the iraq war you are potentially the dumbest person to ever live

Under Irish law Musk can be jailed for 12 months for being a director of a company distributing CSAM. Section 9 of the Child Trafficking and Pornography Act 1998. 'X' is a company registered in Ireland.

A lovely gentle and accessible dialogue on the limits of computationalism.

We Reject the Use of Generative Artificial Intelligence for Reflexive Qualitative Research - Tanisha Jowsey, Virginia Braun, Victoria Clarke, Deborah Lupton, Michelle Fine, 2025 Four hundred and nineteen experienced qualitative researchers from 32 countries invite readers of Qualitative Inquiry to consider their position on use of gener...

Now published…

We Reject the Use of Generative Artificial Intelligence for Reflexive Qualitative Research - Tanisha Jowsey, Virginia Braun, Victoria Clarke, Deborah Lupton, Michelle Fine, 2025

journals.sagepub.com/doi/10.1177/...

The most generous reading of this is as art intervention, giving AI researchers an opportunity for reflection by applying AI slop to their own work and reputation and seeing the damage it does to

#TwinPeaks #ChatGPT #LogLady
@logladyfilm.bsky.social

Screenshot of an ACM Digital Library search result showing an AI-generated summary where an abstract of the article would usually appear.

What the absolute shit is this? ACM using AI summaries instead of abstracts in search results?! So, if I understand correctly, instead of the author's own summary of the work (i.e. an abstract) we are better off with some regurgitated slop version of that summary instead? @chi.acm.org

why is the ACM stabbing authors in the back with generative AI text effluent?

Apply for my old job as Lecturer in HCI at Bristol. It's a lovely group and a good place to work. I only left because my wife lived in a different country.

www.bristol.ac.uk/jobs/find/de...