Introduced in 2018 and expanded to all roads by 2020, Bilbao was the first European city of more than 300,000 inhabitants to adopt a universal 30 km/h speed limit. Several years on, the results suggest it has delivered broad benefits on safety, public health, travel efficiency, and quality of life.🧵

Reminds me of a quote about Libertarians, comparing them to cats: Living in luxury, built on the labour of others that they remain completely oblivious to.

The launch of every piece of AI-driven surveillance/security/criminal justice software basically goes like this: AI COMPANY: this is an error free, bias free, state of the art, highly intelligent software system Two Weeks Later: …aaaand it’s racist

Interesting. I wonder is it because both types are acting more in some way (thinking of a paper by Vuorre about 10 years ago)

People compare LLM public adoption to various other tech: Nuclear Cars Internet Possibly, though: Chemistry Processed food Plastic Phone exchanges ‘Page-looking’ websites Pop up windows Internet install CD-roms

Reminder (have said before): Academia.edu does not have its domain because it’s an academic institution (unlike overwhelmingly most other .edu). The for-profit got ita .edu domain before such things were regulated its not in academia‘s interests to be on there. try ORCID (or maybe researchgate)

Big news! A win!! The Supreme Court has ruled that Trump exceeded his authority when he imposed sweeping tariffs on imports from nearly every U.S. trading partner. A major check on executive power!

It's happening again – I wrote a #LinkedIn post about how the US is entering an AI surveillance dystopia. It quickly got 7,8K views, 8 shares, 68 reactions & 11 comments, then suddenly 0. My comment on this has 9 views. This quashing of "wrong" virality also hit X… www.linkedin.com/feed/update/...

Conversations with AI chatbots are not protected by attorney-client privilege

What's most likely happening is easier to understand if you mentally model LLMs as glorified search engines. They are matching your search query (prompt) with some text they found on the internet. Internally it's a bit more complex than that, but much closer to Google than to a thinking machine. 2/?

1/2 Anyone who knows me knows tho I do academic philosophy (write books, teach when I can) I don’t live off it (in practice, there are no academic philosophy jobs in Ireland) I do technical writing - I find in this AI inflation, no one knows what that job is (but many of them believe AI can do it)

Here’s the annotated version of the post. Every single media outlet that has platformed Matt Shumer has failed their audience and embarrassed themselves. It proves that many outlets simply don’t know what they’re talking about. www.dropbox.com/scl/fi/qw6k5...

This article Ed Zitron discusses does the rounds in my day-job industry (ie, on LinkedIn)

The onion remains undefeated in delivering the most accurate news. theonion.com/ai-chatbot-t...

Ok, this is probably a tough one, I don’t know I’m looking for a list of AI-related movies, but: ones that are relevant to the current idea of AI in public discourse - I’d not necessarily call them (LLMs) AI Any aspect though, eg, RNNs, manipulative-first machines, shallow simulacra

again, my 15+ yrs google alerts gives mainly - time travel > ‘folk thinks this photo from [eg 1901] shows time travel‘ | ‘physicist says time travel [possible | impossible]’ - time consciousness > Philippine news piece on gov encouraging workers sign pact to be time conscious - Phil of time > bleh

I figured this may start happening to new work - but never thought it may backdate. Thie will further exacerbate the exploitation of the marginal, the unmonied, the less known . ( ‘A.I.’, my arse, unless ‘A’ stands for ‘arse’)

Out of curiosity, I just ran a few paragraphs from The Butchering Art through an AI checker and it got flagged: 88% AI. This book was released in 2017, and AI machines were subsequently trained off it. How many writers are getting flagged for AI because of the literal theft of their work(s)? Insane.

I don't know if this is from Grammarly's Authorship or another of their products, but I ran one of their AI/plagiarism checkers on a paper that was 100% human-generated with properly cited quotes and got this:

My old web post on the difference between language & perception, I guess, in their point (if any). It’s why I think representation so general it applies to both has almost no use. A brief way to put it: - language is for gossip - perception is for witnessing seanendapower.com/2014/05/30/a...

Goddamn it - some media companies... This title is like a title that says ‘New ebola strain faces ‘massive threat’ from national healthcare programs over symptomatic pus, blood

Fascinating study. Also, mentions Propp’s work, reveals how quite INhuman an LLM black box is, and opaque - and how surreal extracting information from it could get

Wonder is this the year the use of AI in all walks of life is finally clearly understood, and so we can move to a) it making money and folk generally seeing how great and useful it is - if it is b) convincing bosses who don’t use it to drop forcing it on folk and let folk back to doing their jobs

Prophet revises time of apocalypse after originally projected date passes uneventfully.

Scientists and non-scientists reading this: what do you think of this study? Does it hold up? Some circumstantial alarms: - OA/mega publishing jnl, Nature Scientific Reports - Seems a topic for studies in education, but authors are only either physicists or engineers www.nature.com/articles/s41...

… anyway, so I know folk who use this tool to summarise meeting discussions in work, podcasts they listen to, voicemails from friends, emails from people they don’t want to read…. S.I.G.H.

‘An X message popup says safeguard lapses led to …”

grok didn't say shit. Grok is not sentient. This is you helping musk escape accountability for this shit. Journalism fail.

Latest papers: Marek Gurba argues that the strategy of one authors who argue that our numerical identity over time is irrelevant to practical issues is implausible doi.org/10.1080/0951... #philsky #philpsy

Ah… researchgate; you only ever wish the best for me. “Sean, here are jobs you might be interested in: Professor of Philosophy at NYU; Head of CERN; Philosopher-King of Athens…”

Kind of like rhe way their sora simulated vid generator can plug in anyone into a scene, and Altman started by allowing himself. Where harm is indicated, you could maybe do a similar plugging

Post a perfect album from the 90s that isn't Nirvana, Soundgarden, Pearl Jam, or Alice in Chains

Microplastics are here to stay; that's why we need to eat them responsibly in the classroom

A recent issue of the #journal Science Editor (published by the @councilscienceed.bsky.social, of which I am a member) carries a commentary entitled “Undeclared #AI-Assisted Academic Writing as a Form of #Research #Misconduct.” I agree with the author. www.csescienceeditor.org/article/unde...

Visiting Dublin what feels the first time in months - years? - this week

A transatlantic showdown over tech policy (as opposed to, say, net zero) could actually have a unifying & galvanising political effect within the EU. The union’s digital regulations are broadly popular and many voters actually want more, not fewer, controls on Big Tech. yougov.co.uk/technology/a...

Yes My odd naive kinda-Kant take is ‘people are ends not just means’ requires moral actions is done by all One takes the action but all capable recipients consent So, for social problems, moral AI seeks recipient consent & this needs trust: like anyone, it must prove it‘s wise, not just be wise

Just saw an ad for an ‘AI puppy’ on YouTube during a puppy training video (we just adopted one, is crazy) This AI ad: many layers - granny voiceover showing a slow motion cute puppy playing with a grab and kida She says: smartest ai puppy ever, causing parents problems, selling out for Xmas 1/2

It’s clear that trans people face unique, advanced, and persistent threats—online and off. Here are some strategies for resisting the tech-enabled violence that trans people face. www.eff.org/deeplinks/2...

Jesus @mary1010.bsky.social this is troubling as a development

Recommended this by @404media.co podcadt. podcasts.apple.com/ie/podcast/s... It’s extraordinary. The eps where the two simulations of the presenters is hilariously Bizarre yet dull.

Out soon: *Contemporary Debates in the Ethics of Artificial Intelligence*, co-edited by me, @atoosakz.bsky.social & John Zerilli. Out on Jan 21, but preview already available on Google books: books.google.de/books?id=beK... #aiethics

the new University College Cork (UCC) research portal (Pure) is by all accounts very bad For example, ALL my publications (an unpaid research affiliate of philosophy in the College of Arts & Social Sciences) are listed under the School of Medicine - even though I’m not listed there at all

This has never been done before by anyone.

A problem with feeling that the past was safer than the present (which I’ve found fairly common) is we are descended from its survivors + we ourselves are not certain to survive the present But the past was as bad. It destroyed people like us Like some of us Just not those we call mam and dad

“The work was emotionally exhausting, with chat users confiding intimate details about their real-life relationships, as well as their own emotional trauma, falsely believing they were talking to an unfeeling AI chatbot.”

In the PSA, “Creepy-looking humans are cast as the voices and faces of real AI chatbots that have allegedly shared dangerous information with young users who engaged with them, like how to harm themselves and hide an eating disorder from their parents.”

Rep. Correa shares the story of Donna Hughes Brown, an Irish citizen and green-card holder who has lived in the United States for 48 years. She is the mother of a U.S. Marine and the sister of a retired Army colonel. She has been held in ICE detention since July.

Time to stop working for free to springer and elsevier.