iiinova — where your digital self keeps evolving

Reposted by Laurence D

These people are so deep in all of this prophesy and lore that they don’t understand what is and is not real. They fundamentally do not understand modern AI; they just expect it to comport with their scriptural Armageddon. Their position is fundamentally religious.

2948

Laurence D @laurencediver.net · 01/10/2026

This is such a depressing article. Even the phrase 'content creator' is indicative of so much.

000

Reposted by Laurence D

Reuben Binns @rdbinns.bsky.social · 29/09/2026

I'm hearing reports that Volkswagen's AI has been scheming to deceive its evaluators into thinking its emissions are safe. We need to shut down all cars until they are climate-aligned.

143

Laurence D @laurencediver.net · 26/09/2026

Deeply disappointing to see the Turing Institute @theturing.bsky.social accept £2m from Coefficient Giving (née Open Philanthropy), *the* major Effective Altruism outfit www.turing.ac.uk/news/new-ps2...

120

Reposted by Laurence D

Carl T. Bergstrom @carlbergstrom.com · 26/09/2026

Does my Speak-n-Spell feel pain when I type O-U-C-H? The jury is out. Philosophers and neuroscientists say that's fucking stupid, but I called a guy from the Center for Pretending that Household Electronics Are Sentient and he said there's no way to know for sure.

5843311188

Reposted by Laurence D

Louise Amoore @amoorelouise.bsky.social · 20/09/2026

Yes absolutely. Worst so far is opening a PDF PhD thesis to examine “THIS LOOKS LIKE A LONG DOCUMENT. WOULD YOU LIKE A SUMMARY?” Seriously, 4 years of the student’s life & research…

214914

Reposted by Laurence D

Jason Kint @kint.bsky.social · 17/09/2026

Eye popping. Court just unsealed Summary Judgment motions in New York Times v OpenAI / Microsoft. Easy to see why the AI companies had so many redactions (pink). Their own people wrote the lede for NYT, the "largest theft of labor in human history"... "more and more substitutive..." /1

4253582335

Reposted by Laurence D

Tech Policy Press @techpolicypress.bsky.social · 18/09/2026

Surveying the AI ‘moment,’ Nathalie Maréchal says "the kooks are now in power, and they are using that power to dismantle the institutions designed to keep the rest of us safe, healthy, and free."

buff.ly

What Policymakers and the Public Need to Know About the Cult of AI

Nathalie Maréchal asks, why are we betting the economy and the future of democracy on cult-like beliefs?

076

Reposted by Laurence D

Faine Greenwood @faineg.com · 17/09/2026

Nick Beckstead, noted longtermist and "effective altruist,” just comes out and says it: "But it now seems more plausible to me that saving a life in a rich country is substantially more important than saving a life in a poor country, other things being equal."

This is precisely why Nick Beckstead, a noted EA, argued in 2013 that, given the "overwhelming" moral importance of the far future - "millions, billions, and trillions" of years from now - we should prioritize saving the lives of people in rich countries over those in poor countries, other things being equal. The reason is that people in rich countries are better positioned to influence the far future.

To take another example, saving lives in poor countries may have significantly smaller ripple effects

than saving and improving lives in rich countries. Why? Richer countries have substantially more

innovation, and their workers are much more economically productive. By ordinary standards at

least by ordinary enlightened humanitarian standards saving and improving lives in rich countries

is about equally as important as saving and improving lives in poor countries, provided lives are

improved by roughly comparable amounts. But it now seems more plausible to me that saving a life

in a rich country is substantially more important than saving a life in a poor country, other things
being equal.
Screenshot from Beckstead's 2013 PhD dissertation, regarded as one of the founding documents of longtermism, one of the main cause areas of EA.

1521267286

Reposted by Laurence D

Adam Becker @adambecker.bsky.social · 17/09/2026

Just absolutely wild to see so many people discovering the massive influence of the rationalists in real time and say things like "I wish I could unlearn what I've learned in the last 24 hours." I, and others, have been researching these people for YEARS. There is so much I wish I could unlearn.

1061695

Reposted by Laurence D

ve, the ghost of info past @vortexegg.com · 16/09/2026

The fact that LLMs as a product (not the models) are designed to emit "thinking traces" and to label that as "thinking" is one of the more insidious parts of the propaganda of AI singularity and ASI consciousness woo targeted at the most credulous people who believe in computer pseudoscience.

511322

Laurence D @laurencediver.net · 16/09/2026

There's been surprisingly little coverage of this: Turnitin to change its terms to enable training AI models using students' work www.timeshighereducation.com/news/southam...

timeshighereducation.com

Southampton dumps Turnitin over use of students’ work to train AI

Academic integrity giant denies accusations as university declines to renew contract in favour of ‘other options’

000

Laurence D @laurencediver.net · 16/09/2026

@zeteo.uk I'd consider upgrading to a paid subscription, but not when a chunk of the money goes to Substack. Wondering why you chose them when there are far better alternatives.

000

Reposted by Laurence D

ve, the ghost of info past @vortexegg.com · 15/09/2026

The root of this is Rationalism is primarily a *rhetorical style*, the adoption of which is meant to make its interlocutors "think better". It isn't a better reasoning style. It isn't a form of actual statistical analysis. Members of the community are taught to talk about things using made up stats.

27317

Reposted by Laurence D

Timnit Gebru @timnitgebru.blacksky.app · 13/09/2026

Like YCombinator was running their startup school at Stanford. I even audited a "class" taught by Peter Thiel. And now the Effective Altruists, are out in full force there (as they are at Oxford, MIT, Cambridge, Berkeley). These are their strong holds.

1759

Reposted by Laurence D

Terence Tao @teorth.bsky.social · 11/09/2026

A group of 25 Fields Medalists, including myself, have made a joint declaration on Math and AI: mathandai.org . We welcome additional signatories. See also this article in the Economist announcing the declaration: www.economist.com/science-and-...

mathandai.org

Declaration — Math and AI

Read the declaration and add your name.

422052929

Reposted by Laurence D

Oliver Bullough @oliverbullough.bsky.social · 11/09/2026

New podcast idea: The Rest is The Rest is; two middle-aged people are paid £1,000,000 a month each to swap non-controversial opinions on a podcast about the non-controversial opinions swapped by middle-aged people being paid £100,000 a month on other podcasts, with the profits going to Gary Lineker.

281

Reposted by Laurence D

Dave Karpf @davekarpf.bsky.social · 11/09/2026

The key to understanding this shit is: (1) they believe they are smarter than everyone else, and that this amounts to a superpower. and (2) they think their AI is going to be smarter than them. They really do believe this. But they themselves aren’t nearly as exceptional as they think.

1616436

Reposted by Laurence D

Anthony Moser @anthonymoser.com · 10/09/2026

this is, I think, one of the inherent problems with LLMs using natural language interfaces: they are a statistical tool *that looks like* a logical tool statistical approaches can be a useful way to find related information! BUT for it to be useful you have to *know that's what it's doing*

938293

Reposted by Laurence D

Helen Czerski @helenczerski.bsky.social · 10/09/2026

As both an academic and a writer, I am repelled by the "this looks like a long document. Would you like me to summarize it for you?" that seems to appear at the top of every pdf viewer now. NO. The *entire point* of good writing (either books or journal papers) is that every sentence has a purpose.

1344044793

Laurence D @laurencediver.net · 10/09/2026

🔊 'Alignment' isn't a thing that can be 'solved'. 'Human values' is not a monolithic entity that can be represented computationally then mapped and complied with, and especially not on a probabilistic basis (I mean come on).

131

Reposted by Laurence D

Melanie Mitchell @melaniemitchell.bsky.social · 09/09/2026

What, an Anthropic researcher states (without evidence) that there is a 10% chance that AI will cause human extinction? Incredible, we have never before heard such a claim!

2027390

Laurence D @laurencediver.net · 10/09/2026

Stuart Russell (intentionally?) mischaracterising 'AI' while millions of people listen on #r4today

120

Reposted by Laurence D

Ben Recht @beenwrekt.bsky.social · 30/08/2026

The overreaction to this Hugging Face thing is driving me nuts. Come flay me for being wrong but... 1/x

14436126

Reposted by Laurence D

Sebas @ohlookbirdies.bsky.social · 01/09/2026

If you're not using asbestos yet, you are being left behind. Everyone is using asbestos, there's no point in trying to fight it. Asbestos is the future.

8957101553

Laurence D @laurencediver.net · 01/09/2026

Bank of England Governor sounding the alarm on the AI finance bubble in a letter to the G20 (28 Aug): www.fsb.org/uploads/P310...

Screengrab of part of Andrew Bailey's letter to the G20: "The issue is not simply that investors are borrowing more, but that leverage is interacting with high valuations and market concentration, in particular the increasing cross-investment between artificial intelligence (AI) companies and hyper scalers, in a way that could amplify a future market correction. I remain concerned therefore that a large shock or combination of shocks could concurrently trigger multiple vulnerabilities."

000

Laurence D @laurencediver.net · 28/08/2026

An actual Chinese Room moment I was testing Claude just now and these two Chinese characters appeared in the middle of its (English) output. They translate as 'recommend', which is the 'right' word for this point in the sentence ("I'd actually recommend keeping...")

A screengrab of a Claude output in English, showing two Chinese characters for 'recommend' in place of the equivalent English word.

111

Reposted by Laurence D

Anil Dash @anildash.com · 26/08/2026

Meta is a creep who will _never_ take no for an answer, and never stop trying to ignore their lack of consent. Every media & regulatory organization in the world seems determined to ignore this reality; I genuinely don't understand why. Today's settlement is so that Meta can keep violating consent.

131125331

Reposted by Laurence D

Natalie Bennett @nataliegreenpeer.bsky.social · 25/08/2026

Powerful and important letter about artificial so-called intelligence from Sheila Hayman from Minderoo Centre for Technology and Democracy, University of Cambridge, in the FT today

$Margaret McGirr’s letter (“AI can lend a hand, but it will struggle to replace one”, July 21) and Chine McDonald’s guest column (“This retro revival is more than a longing for the past”, Opinion, August 7) both touch (sic) on something that should be obvious, but is carefully ignored by almost everyone in the tech world. Namely that our intelligence is embodied, evolved and biological, and inseparable from our bodies. The brain’s job is to control the body in the world; it is the last organ to develop. Every square centimetre of our hands (and of the rest of our skin) contains about a 170 touch sensors or nerve endings, sensitive to different kinds of touch: a slap, a burn, a caress, a hug — each of which travels to different parts of the midbrain, prompting the appropriate response. Quickly to the autonomic nervous system for something dangerous; more slowly via a different channel for something carrying meaning. In addition to the hand’s limitless versatility, it is self-repairing, regulates its own temperature, knows without being programmed how much pressure to exert on an egg, a tennis ball or a day-old chick, and can fuse neurologically to whatever it holds. So when we put a glass down on the table or slide a knife through an apple, we feel the table, or the apple, not the glass or handle in our hand (try it!). Finally, and most urgently, the entire human body — including the hands — does everything on 25 or so watts of completely renewable energy: a tiny fraction of what a humanoid robot, even one of the current primitive iteration of robots, needs. The “retro revival” is thus not cultural nostalgia. It’s millions of years of evolution crying out to be heard, before we suppress it totally in the service of an automated — and dead — planet. Sheila Hayman Member of the Advisory Board, Minderoo Centre for Technology and Democracy, University of Cambridge, London NW1, UK$

14487264

Reposted by Laurence D

Hagen Blix @hagenblix.bsky.social · 09/08/2026

An editor from the economist telling @olivia.science that nobody confidently understands the difference between the human cognitive capacity for language and a corpus. Idk about you, but to me (as a linguist and cognitive scientist) this is absolutely hilarious

Olivia Guest, quoting Dr Katie Twomey: exactly 💯 on LLMs: they simulate the output we call language, not the cognitive capacity we call language!

Followed by Economist Editor Matt Steinglass saying "I am not sure that I or anyone else confidently understands the difference between those."

28597115

Reposted by Laurence D

Elizabeth Bear @matociquala.bsky.social · 29/07/2026

You'll get-- my em- dashes when-- you pry-- --them-- from my cold-- ---dead--- Keyboard.

414422

Laurence D @laurencediver.net · 27/07/2026

My photo from the Glencoe Lochan looking towards Sgorr Dhearg was included in the BBC's Pictures of Scotland this week 🤠 www.bbc.co.uk/news/article...

A photo of the Glencoe Lochan reflecting a blue sky, with the mountain Sgorr Dhearg in the distance

020

Reposted by Laurence D

Louise Amoore @amoorelouise.bsky.social · 12/07/2026

This is such a helpful & succinct way to put it “polite prompts just land in a friendlier slice of the training distribution”. Hence the emphasis on “personas”/style - the tone of the prompt guides the model to a slice of the training distribution. The politics of this are crucial 1/

1126

Reposted by Laurence D

Emily M. Bender @emilymbender.bsky.social · 10/07/2026

In other words, there is no knowing without knowers, scholarship is a giant conversation in which we speak with our colleagues sometimes slowly (in papers) sometimes faster (Q&A sessions of talks). It is not the accumulation of "facts". We get into this some in this episode, with MJ and Lisa: >>

buzzsprout.com

Episode 31: Science Is a Human Endeavor (feat. Molly Crockett and Lisa Messeri), April 15 2024 - Mystery AI Hype Theater 3000

Will AI someday do all our scientific research for us? Not likely. Drs. Molly Crockett and Lisa Messeri join for a takedown of the hype of "self-driving labs" and why such misrepresentations also harm...

17219

Reposted by Laurence D

Society for Hopeful Technologists @hopeful.technology · 01/07/2026

Hello, everyone! A little about us: We are a solidarity network that: 1) Strengthens and activates Hopeful Technologists, so more people can make, advocate for and stand-up for the technology they believe in and want.

1149

Laurence D @laurencediver.net · 30/06/2026

Migration to @eurosky.social took ~15 mins, painless experience 👌

030

Reposted by Laurence D

Mel Andrews @bayesianboy.bsky.social · 29/06/2026

As philosophy rushes to claim expertise on AI, AI companies rush to control the philosophical narrative. Unlike other disciplines, philosophy has no standards for reporting conflicts of interest. This petition endeavors to change that: docs.google.com/forms/d/e/1F...

docs.google.com

Open Letter about Industry Funding and Disclosure in Philosophy

*For full list of signatories, updated every few hours, please click here.* In recent years, philosophical writing on areas such as climate change, technology, misinformation, and artificial intellige...

416678

Reposted by Laurence D

Dr Abeba Birhane @abeba.blacksky.app · 24/06/2026

reading AI eval papers using llm-as-a-judge is alarming for many reasons; many “scientific claims” rely on evals executed on models that are now deprecated. how are we supposed to treat “findings” that are inherently not reproducible? i despair. llms are accelerating the decay of scientific rigor

59432

Reposted by Laurence D

Alexander Clarkson @aphclarkson.bsky.social · 24/06/2026

Higher education policy under the last few UK governments seems to be shaping up into an experiment in how you sustain a knowledge economy without knowledge

421446

Reposted by Laurence D

Mel Andrews @bayesianboy.bsky.social · 17/06/2026

When utilized in literature review, LLMs consistently 1. fail to mention female authors in female-led literatures, 2. insist that men are more influential or more heavily cited when this is contradicted by objective citation counts, and 3. attribute women’s work to hallucinated male scholars.

8153412302

Reposted by Laurence D

Will Dunn @willdunn.bsky.social · 13/06/2026

*Not investment advice www.newstatesman.com/business/eco...

1103

Reposted by Laurence D

Michael Okun @michael-okun.bsky.social · 10/06/2026

If human editors can’t control who reviews science, it’s no longer peer review — it’s a rubber-stamp machine designed for volume and profit, not quality. I have no intention of attaching my name to it. So I’m out.

302361386

Reposted by Laurence D

Elaine Burke @elaineburke.bsky.social · 07/06/2026

This is just more posturing of Anthropic as the 'safe' AI company. The media are doing free PR for a five-year-old company with a valuation approaching $1 trillion despite having no profit to show for the products that it produces. That valuation is built off this marketing. (15/22)

27024

Reposted by Laurence D

Olivia Guest · Ολίβια Γκεστ @olivia.science · 06/06/2026

These pause AI nonsense calls have been happening for a long time. It's always about power because they wanna consolidate and stop racing against each other for no gains. Here's a previous time of many they tried the same silliness: futureoflife.org/open-letter/...

futureoflife.org

Pause Giant AI Experiments: An Open Letter - Future of Life Institute

We call on all AI labs to immediately pause for at least 6 months the training of AI systems more powerful than GPT-4.

1517157

Reposted by Laurence D

Emily M. Bender @emilymbender.bsky.social · 03/06/2026

"Even if the Microsoft Office team employed a philosopher who said you shouldn’t be so certain, because consciousness is not well understood, that would not be sufficient reason for you to take this idea seriously." www.theatlantic.com/philosophy/2...

theatlantic.com

No, Artificial Intelligence Is Not Conscious

Taken to its logical conclusion, this line of thinking is absurd—and damning.

415143

Reposted by Laurence D

Karl Bode @karlbode.com · 01/06/2026

this really is wild they just asked Meta AI chatbot support for access to high profile accounts and it obliged comes after Google complete broke its search results these are not serious, capable companies deserving of professional respect. they rushed shit product to market.

404media.co

Hackers Simply Asked Meta AI to Give Them Access to High-Profile Instagram Accounts. It Worked

The exploit shows the extreme risk of offloading technical support to AI.

18438156

Laurence D @laurencediver.net · 27/05/2026

lastfm is independent again! Hopefully a case of the enshittification trend being bucked 🤞

020