Lightnews — Scholar-powered news

Reposted by Robin Thomas

The Transmitter @thetransmitter.bsky.social · 5h

Astrocytes get less attention than their neuronal neighbors, partly because they’re harder to study. But the glial cells could help explain how memories take a more stable form after recall.

By @lauren-schneider.com

#neuroskyence

www.thetransmitter.org/memory/engra...

Engrams in amygdala lean on astrocytes to solidify memories

Disrupting the astrocyte-neuronal dynamic in mice destabilizes their memory of fear conditioning.

www.thetransmitter.org

13 26

Reposted by Robin Thomas

Damian Koevoet @dkoevoet.bsky.social · 1d

Spatial attention and working memory are popularly thought to be tightly coupled. Yet, distinct neural activity tracks attentional breadth and WM load.

In a new paper @jocn.bsky.social, we show that pupil size independently tracks breadth and load.

doi.org/10.1162/JOCN...

1 12 26

Reposted by Robin Thomas

Chris Geidner @chrisgeidner.bsky.social · 18h

Here is the students' @idsnews.bsky.social report out of Indiana University, where the student media director was fired today. www.idsnews.com/article/2025...

LETTER FROM THE EDITORS: IU fires student media director after he refused to censor the IDS

Indiana University directed the IDS to stop printing news.

www.idsnews.com

16 180 330

Robin Thomas @drrobinthomas.bsky.social · 10h

as am I

1

Reposted by Robin Thomas

Politico @politico.com · 1d

EXCLUSIVE: Thousands of leaked messages show leaders of Young Republican groups joking about gas chambers, slavery and rape in a private Telegram chat.

Inside rising GOP leaders’ racist chats — obtained by POLITICO and spanning more than 7 months👇

‘I love Hitler’: Leaked messages expose Young Republicans’ racist chat

Thousands of private messages reveal young GOP leaders joking about gas chambers, slavery and rape.

www.politico.com

540 3.3K 5.8K

Reposted by Robin Thomas

The New Yorker @newyorker.com · 1d

The neurologist Oliver Sacks found that the social-media era resembles a neurological catastrophe on a gigantic scale.

The Machine Stops

The neurologist Oliver Sacks on steam engines, smartphones, and fearing the future.

www.newyorker.com

8 57 160

Reposted by Robin Thomas

Olivia Guest · Ολίβια Γκεστ @olivia.science · 23h

Really enjoyed & honoured to be part of the Critical AI Literacy Symposium yesterday. Also @lucyavraamidou.bsky.social & @miquelpt.bsky.social (and others) spoke about their wonderful work too. Big thanks to @irisvanrooij.bsky.social, Leo & Barbara for organising. ✨

www.youtube.com/watch?v=Fxyg...

19 34

Reposted by Robin Thomas

Iris van Rooij 💭 @irisvanrooij.bsky.social · 1d

We know — from first principles — that LLMs can only create AI slop.

Why do people keep needing empirical tests?

Iris van Rooij 💭 @irisvanrooij.bsky.social · Aug 17

The intractability proof (a.k.a. Ingenia theorem) implies that any attempts to scale up AI-by-Learning to situations of real-world, human-level complexity will consume an astronomical amount of resources (see Box 1 for an explanation). 13/n

Box 1 in the paper, intuitively explaining the implications of the intractability result.

4 14 56

Reposted by Robin Thomas

Olivia Guest · Ολίβια Γκεστ @olivia.science · 1d

What's funny — unbearably sad and pathetic — is they mostly seem (these days and since recently, circa 2019) totally bought out by the tech industry. theintercept.com/2019/12/20/m...

How Big Tech Manipulates Academia to Avoid Regulation

“AI ethics” is a field that barely existed before 2017. It’s become a Silicon Valley-led lobby to avoid legal restrictions of controversial technologies.

theintercept.com

1 7 17

Reposted by Robin Thomas

JazzReads📚 @jazzreads.bsky.social · 17d

So exhausting..

210 4K 9.9K

Reposted by Robin Thomas

Dr Abeba Birhane @abeba.bsky.social · 2d

you use llms to:
brainstorm an idea
summarise your lit review
analysis your data
evaluate your findings
write your paper…

a reviewer then uses llms to:
summarise your paper
assess the value of it
check for plagiarism
write-up review…

do you see how this destroys academic integrity

3 20 110

Reposted by Robin Thomas

Aparna Nair @disabilitystor1.bsky.social · 2d

This sentient disposable napkin should stop talking so cavalierly about other people’s jobs while he grifts billions for his pet LLM

Futurism @futurism.com · 2d

So that's where we're at.

Sam Altman Says If Jobs Gets Wiped Out, Maybe They Weren't Even "Real Work" to Start With

Worried that AI will destroy work? Well, Sam Altman asks if you've considered what a farmer from half a century ago thinks of your job, first.

trib.al

6 36 120

Reposted by Robin Thomas

Des Fitzgerald @desfitzgerald.bsky.social · 2d

Seeing people nonchalantly post stuff like makes me feel like I'm going crazy honestly.

A post from an academic on blue sky that says: "I'm learning about "prompt engineering" when asking LLMs to extract data. I now add this to all instructions:

Style
Be concise, analytic, and specific (cite page/figure if provided). If essential info is missing, keep going but flag Unclear and exact data needed. Never invent data.

#academicsky"

58 110 770

Reposted by Robin Thomas

Stephan Lewandowsky @lewan.bsky.social · 1d

The trillion dollar flowchart (and several gigatonnes [?] of CO2): www.bloomberg.com/ne... One may well ask if AI is different from tulips?

3 10 17

Reposted by Robin Thomas

Laprofmme @laprofmme.bsky.social · 2d

Courtesy of colleagues at Penn

A variation on Benjamin Franklin’s 1754 engraving in the Pennsylvania Gazette, published in Philadelphia, in an effort to foster cooperation among the various colonies against the tyrant king of Britain, with the American colonies replaced by the nine universities selected for special collaboration opportunities

34 840 3.9K

Reposted by Robin Thomas

Judith Beyer @judithbeyer.bsky.social · Jul 7

🧵Thread that gathers studies, reports and news on what I’d like to call #ContraAI : showcasing the detrimental effects of #ArtificialIntelligence, particularly in the realms of #research and #education. If you want to contribute posts on the theme, add them below.
No posting of AI prompts, please! 🤖

a cartoon of a boy reading a book surrounded by books .

Alt: Linus van Pelt (of the Peanuts) doing his own research.

media.tenor.com

4 18 35

Reposted by Robin Thomas

Laura Jedeed @laurajedeed.bsky.social · 3d

Hi hello did you know that Elon Musk's third Neuralink patient (previously nonverbal) is "getting help from Grok" to communicate

and I hope -- I REALLY HOPE -- he's not stuck in his own brain listening as a robot talks for him without his input because this, to me, would be worse than death

This patient’s Neuralink brain implant gets a boost from generative AI

A patient who types with his brain thanks to a Neuralink implant is using AI chatbots to speed things up.

www.technologyreview.com

30 120 480

Reposted by Robin Thomas

derek guy @dieworkwear.bsky.social · 3d

About a month ago, the Trump administration got rid of the de minimis exemption, whereby packages valued under $800 could slide in without import duties. Now there's a backlog as the government can't process all of this paperwork, leading to UPS just destroying packages

Business Insider headline reads: UPS is telling customers that their packages coming to the US are marked for destruction.

220 3.7K 9.5K

Reposted by Robin Thomas

Raider @iwillnotbesilenced.bsky.social · 3d

This Is Fascism

690 13K 30K

Reposted by Robin Thomas

Hypervisible @hypervisible.blacksky.app · 4d

The administration “is currently pressuring OpenAI and other AI companies to make their models more conservative-friendly.”

OpenAI is trying to clamp down on ‘bias’ in ChatGPT

GPT-5 is better at resisting liberal ‘pressure,’ the company says.

www.theverge.com

40 160 360

Reposted by Robin Thomas

John Warner @biblioracle.bsky.social · 5d

I strongly urge everyone to not just read this warning from @marcwatkins.bsky.social, but heed it, and be vocal and forceful pushing back against using AI to grade student writing. This must be anathema if we're going to have a world where learning means something. substack.com/inbox/post/1...

The Dangers of using AI to Grade

Nobody Learns, Nobody Gains

substack.com

7 130 280

Reposted by Robin Thomas

Leah Greenberg ❌👑 @leahgreenberg.bsky.social · 6d

fun fact: I was trying to find an old tweet I made about journalists accidentally quoting me as Leah Ghostberg during halloween season. instead, I got a fully hallucinated anecdote about me dressing up in a ghost costume with my husband that absolutely, 100% never happened

humanity is so cooked

A false google AI summary: "Based on social media activity, the "leahgreenb ghostberg halloween" query appears to reference a humorous, improvised Halloween costume made by political activist Leah Greenberg in 2024.
On October 26, 2024, Greenberg posted on X (formerly Twitter) asking for help with a last-minute Halloween costume she could make with items she had at home.
The final result, which she and her husband shared, was an improvised "Ghostberg" costume. It consisted of two people with sheets draped over their heads, with her husband standing slightly behind her and visible through the sheet. The name "Ghostberg" is a play on her surname.
The post went viral and the joke was widely shared on social media. "

8 48 290

Reposted by Robin Thomas

Stuart Watt @morungos.bsky.social · 5d

I’ve heard people give variants on this argument for 30+ years, and mostly all they say is: we don’t understand either QM or consciousness, so they’re obviously the same thing.

I don’t want your “spooky” unification concepts, you better have a damn good theory, and I haven’t heard one yet.

Daily Neuron @dailyneuron.bsky.social · 5d

New research suggests quantum cognition, a field applying quantum math to decisions, could finally connect microscopic brain processes with our conscious experience. #consciousness #cogsci

dailyneuron.com/quantum-cogn...

Quantum Cognition May Be the Missing Link to Understanding Consciousness - Daily Neuron

New research suggests quantum cognition, a field applying quantum math to decisions, could finally connect microscopic brain processes with our conscious experience.

dailyneuron.com

4 6 41

Reposted by Robin Thomas

Not only in science.

8 15

Reposted by Robin Thomas

JJ @jjodx.bsky.social · 4d

One of my two last (ever) papers on tDCS:
Does anodal tDCS over M1 really enhance motor sequence learning? A non-replication of earlier findings in a double-blind, pre-registered large-sample study in humans
www.medrxiv.org/content/10.1...

Does anodal tDCS over M1 really enhance motor sequence learning? A non-replication of earlier findings in a double-blind, pre-registered large-sample study in humans

Background Transcranial direct current stimulation (tDCS) is one of the most widely used noninvasive neuromodulation methods. Despite its popularity, some recent studies highlighted issues about the r...

www.medrxiv.org

4 16 32