iiinova — where your digital self keeps evolving

Reposted by Reuben Binns

Alexandra Olteanu @aolteanu.bsky.social · 5h

AI alignment is at the centre of the growing debates about AI systems and their failures. The impulses of AI alignment work are however *not* new. Making sure computational systems do what we want them to do has long been a core goal of computing research.

Explanatory-analytical framework for examining and describing AI alignment work through the lenses of system behavior specification and measurement theory. It foregrounds and makes explicit key constitutive components that AI alignment work commonly relies on to specify what the desired behavior should be, whose behavior is being specified—or who or what is being aligned—and how to assess whether they are aligned. Each gray box describes the role of each constitutive component at the conceptual level, while each corresponding red box describes the instantiation or operationalization of that component. The framework also illustrates relationships between components through both arrows and nested boxes; relationships that need to be preserved not only at the conceptual level, but also at the operational level. Both the conceptual and operational choices related to these constitutive components are informed by a diversity of epistemic considerations.

142

Reposted by Reuben Binns

Will Thorne @wbthorne.bsky.social · 20h

This graph is a good reason to introduce my new paper: By Generation and by Design: A Review of the Empirical Rent Control Literature. This paints a (somewhat) misleading picture of rent control evidence because it groups all rent control together regardless of design feature.

516549

Reposted by Reuben Binns

Ben Green @benzevgreen.bsky.social · 06/10/2026

Algorithmic Realism: Data Science Practices to Promote Social Justice comes out today! I wrote this book with the mission to transform data science and AI into fields that genuinely work in service of the public interest. I hope that you’ll join me! mitpress.mit.edu/978026205570...

A photo of the cover of the book Algorithmic Realism by Ben Green.

410225

Reposted by Reuben Binns

Wolfie Christl @wchr.bsky.social · 02/10/2026

LLM chatbot surveillance advertising is going to be next-level disastrous, on every level.

02013

Reposted by Reuben Binns

Wolfie Christl @wchr.bsky.social · 02/10/2026

The web and mobile apps of major conversational AI engines transmit prompts, titles and conversation permalinks to advertising data firms, alongside device IDs, hashed email and other persistent identifiers. Comprehensive analysis, like always from this team: jorgegarciaherrero.com/wp-content/i...

26667

Reuben Binns @rdbinns.bsky.social · 05/10/2026

I did! Didn't realize you're on bluesky, should have tagged, apologies

110

Reuben Binns @rdbinns.bsky.social · 05/10/2026

The simplest explanation for X may not be the simplest explanation for something that behaves like X, if we have independent reasons to think they have different underlying mechanisms.

020

Reuben Binns @rdbinns.bsky.social · 05/10/2026

The problem with applying Occam's razor to explain AI system behavior - i.e. "human-like outputs are most simply explained by positing a human-like mind" - is that what seems like the 'simplest' explanation is just the one we're used to for most human-like behavior we've encountered until now

120

Reposted by Reuben Binns

Ben Wray @benwray1989.bsky.social · 04/10/2026

Glovo is already facing a labour inspectorate investigation over continuing to operate during a heatwave in Barcelona in the summer. It should also be held to account for operating during a flood in Bilbao. No rider should be pounding the streets during extreme weather. bsky.app/profile/fons...

074

Reuben Binns @rdbinns.bsky.social · 03/10/2026

Of course, regulators can only 'think' whatever their industry lobbyist captors program them to think

021

Reuben Binns @rdbinns.bsky.social · 03/10/2026

It would be mean to pick on specific examples but it feels like for some on the pro side there's kooky messianism (we've created God and must upload ourselves to the silicon heaven), and for some on the anti side there's kooky substance dualism

4111

Reuben Binns @rdbinns.bsky.social · 03/10/2026

"The conditions that major functionalist theories identify as relevant for consciousness turn out to be embarrassingly easy to implement" ... "The true theory of consciousness would not tell us which AI are conscious so long as the relevant philosophical debates remain unsettled"

1100

Reuben Binns @rdbinns.bsky.social · 03/10/2026

While I do think much current talk of AI consciousness is deranged, possibly AI psychosis-induced, I also don't think philosophers have a robust basis to decisively debunk it either. This paper 'On Cheap Artificial Consciousness' lays out a good case for why that is: philpapers.org/archive/MICO...

philpapers.org

7586

Reposted by Reuben Binns

michael veale @michae.lv · 01/10/2026

The proposed social media ban, the EU KIDS Act, is perhaps the most chaotic law I have ever read! New paper: Are EU Kidding Me? Notes from a Poorly Drafted Social Media Ban. files.michae.lv/papers/veale_areEUkiddingme.pdf In the spirit of the proposal I have stayed offline and read no takes.

In September 2026, the European Commission published a proposed EU KIDS Act, responding to the demand for restrictions and further design obligations applying to digital services accessible to children, including social media. This Article argues that regardless of what one believes about the merits of this policy direction, the EU KIDS Act is not the legislative text we need. The scope and drafting are little short of chaotic, full of holes, enormous ambiguities, and unintended consequences. Recitals contradict articles. The use of scope borrowed from other pieces of legislation not designed for this purpose brings services into scope that should not be subject to the relevant provisions, like book-recommendation websites, music services, or exercise trackers. The infrastructural requirement for assurance of age and parental responsibility attestation simply omits enormous issues of complex families and childhoods. Complex design obligations are written in ways that can be plausibly interpreted as implicitly prohibiting entire swathes of products, such as voice assistants. This Article outlines these issues, and many more. The Commission should withdraw this proposal and reissue a workable and realistic text.

38758

Reposted by Reuben Binns

rasmuskleis.bsky.social @rasmuskleis.bsky.social · 01/10/2026

Interested in, or already using (for academic or industry research) digital twins? In this analysis, digital twins' responses "correlate weakly [0.20] with human responses". Not nothing but.. not great if goal is reliable knowledge or basis for consequential decisions www.science.org/doi/10.1126/...

0167

Reposted by Reuben Binns

Jack Stilgoe @jackstilgoe.bsky.social · 30/09/2026

On my way to the government’s expert advisory panel on self-driving vehicles. Passed this guy, stopped in a pedestrian crossing. Oops.

3334

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.

353

Reuben Binns @rdbinns.bsky.social · 28/09/2026

I recently had the pleasure of being interviewed for The Inference podcast, which has dropped today. We talk about the research I've done with @workerinfox.bsky.social and Uber drivers, the concept of a 'human in the loop'(hole), the state of AI, and more: youtu.be/dHJLcSl18RI?...

youtu.be

Your Boss Is an Algorithm (Uber Case) | ft. Reuben Binns | Mian Khubaib Jim | Episode 6

YouTube video by The Inference Podcast

020

Reuben Binns @rdbinns.bsky.social · 27/09/2026

My space ships and gliders are so intelligent they have broken containment and are collaborating and self replicating in the wild. I urgently need a trillion dollars to align them with humanity or there's a 10% chance all our pixels will be killed.

static.klipy.com

conways game of life space ship generator

Alt: conways game of life space ship generator

073

Reuben Binns @rdbinns.bsky.social · 25/09/2026

Ahem, spurs would also like a word about the title in 16-17 and league cup '21

110

Reposted by Reuben Binns

Simon McGarr @tupped.bsky.social · 24/09/2026

I think the relevant caselaw here is shaping up to be Rylands & Fletcher. If you collect up any inherently risky material or item on your property and it then escapes and causes damage, it creates strict liability. ie, you do not have to prove liability, you go straight to quantum of compensation.

35524

Reuben Binns @rdbinns.bsky.social · 23/09/2026

Falling out of my trembling fingers

120

Reposted by Reuben Binns

Lilian Edwards @lilianedwards.bsky.social · 23/09/2026

Very happy to have in our! trembling hands the hard copy of Law, Policy and the Internet 2nd edn( Hart/Bloomsbury), just out in the nick of time for first semester 2026/7! Order or inspection copies at share.google/RCRm2lmLQNAb... & do note a guide to teaching sample "tech law" course with it!

5438

Reposted by Reuben Binns

Joe Bak-Coleman @jbakcoleman.bsky.social · 20/09/2026

I don’t mean this to be cruel, but there’s just too much air being taken up by people with media or tech backgrounds who think their hot take on consciousness with LLMs is new, and lord it has all been done before and we’re past that. Christ, half the news cycle is yud refusing to read a book.

738464

Reposted by Reuben Binns

Dave Karpf @davekarpf.bsky.social · 18/09/2026

Based upon their documented behavior over the past few years, the tech industry is less afraid of the potential ramifications of Artificial Superintelligence than they are of Lina Khan.

91074254

Reuben Binns @rdbinns.bsky.social · 18/09/2026

Have you tried turning it off and not on again?

141

Reuben Binns @rdbinns.bsky.social · 18/09/2026

You can literally unplug a cable! bsky.app/profile/merb...

221

Reuben Binns @rdbinns.bsky.social · 18/09/2026

Lads, what are we doing here? First, we didn't want freely editable text documents, so we made fixed PDFs. Then, we changed our minds and wanted people to be able to edit those PDFs. Then we changed our minds again and put these wonderful little 'lock form' buttons in the PDF 😆

screenshot from a pdf "Click the 'Lock Form' button to prevent further edits after completing the form"

000

Reuben Binns @rdbinns.bsky.social · 16/09/2026

This would actually be a fun game to play for a human (me)

000

Reuben Binns @rdbinns.bsky.social · 16/09/2026

Can any central bank economists explain to me how making mortgages more expensive will reverse the impact of extreme-weather-induced-reduced-crop-yields on food prices, end the US-Iran war and resulting fuel shortages, refill the Panama canal, solve memory chip shortages, and prevent greedflation?

070

Reposted by Reuben Binns

Les Carr @lescarr.bsky.social · 15/09/2026

Lots of online discussion about the implications of AI solving the Navier-Stokes Millenium Problem. But spending $15,000,000 to get a $1,000,000 prize is just business as usual for OpenAI.

051

Reposted by Reuben Binns

Rude1 Haunted Badness. ⁂ @rude1.blacksky.team · 15/09/2026

Blacksky was one of the 1st PDSs to allow open sign ups We are also now the largest PDS deployment that doesn't run any of bsky's code Sorry for the turbulence. It was the one-time price to pay for being able to build features like private accounts or sign-in via phone number w/o waiting on Bluesky

9566133

Reposted by Reuben Binns

Abby Gilbert @agkg1.bsky.social · 19/04/2026

⚠️New Report Alert⚠️ Economic Sovereignty and the Question of Post-Deployment Training AI sovereignty is about who captures the value from worker and organisational know how codified in workplace data. This presents a risk not only to workers, but also firms and nation states 🧵

248

Reuben Binns @rdbinns.bsky.social · 14/09/2026

After a brief hiatus, I'm back! This time, how unweighted average commission statistics can hide how much money is being taken from Uber drivers: vm.tiktok.com/ZN8jqSQoF/

vm.tiktok.com

TikTok - Make Your Day

010

Reuben Binns @rdbinns.bsky.social · 14/09/2026

Luckily, the UK already has the world's safest network of kill switches for dangerous AI:

1110

Reposted by Reuben Binns

Daniel Solove @daniel-solove.bsky.social · 14/09/2026

One way to slow down AI is to start strengthening and enforcing privacy laws. Even existing laws if interpreted wisely and enforced by regulators with a spine can make a difference.

517641

Reposted by Reuben Binns

Colin @colin-fraser.net · 12/09/2026

But there is simply no reason that I can see that an “advanced AI” must behave in this way. Humans, for example, famously don’t. Things that act like perfect expected utility maximizers over an open world are actually quite rare, and it’s not clear that we know how to make them.

315910

Reposted by Reuben Binns

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.

1354043793

Reposted by Reuben Binns

Nikhil Venkatesh @nikvenkatesh.bsky.social · 10/09/2026

Why would you do work that you think has a significant chance of wiping out humanity? The answer is capitalism!

271

Reuben Binns @rdbinns.bsky.social · 10/09/2026

Apropos of the latest attacks on RNLI volunteers, I present my distant relative Bert who was in the RNLI. A true patriot and a very fetching chonky lad in his rescue gear

031

Reuben Binns @rdbinns.bsky.social · 03/09/2026

A year on from releasing our audit of Uber's pricing and pay algorithms, our collaborators @workerinfox.bsky.social have filed this class action lawsuit in Amsterdam. Drivers across Europe may win €billions compensation for violations of their rights under GDPR: www.theguardian.com/technology/2...

theguardian.com

Uber drivers launch European class action over ‘soulless’ and ‘scary’ AI algorithm

Landmark lawsuit claims tech company’s pay-setting system has breached privacy laws and pushed down earnings

1209

Reposted by Reuben Binns

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 Reuben Binns

Lilly Irani @gleemie.bsky.social · 26/08/2026

Mechanical Turk may be shutting down, but AI will always depend on data workers whose knowledge trains and tunes models. AI runs on extracted cultural skills and knowledge, and culture is always changing. Less the end of an era, but the end of an early prototype.

215260

Reposted by Reuben Binns

Ketan Joshi @ketanjoshi.co · 19/01/2024

Do I have this right

DAVID AND VIC BECKHAM MEME

WE'LL NEED OIL AND GAS FOR DECADES
BE HONEST
WE'LL NEED SOME OIL AND GAS FOR DECADES
BE HONEST
I WANT TO SELL OIL AND GAS FOR DECADES
THANK YOU

312846743

Reuben Binns @rdbinns.bsky.social · 21/08/2026

Yes! But it's one thing among others mentioned in this post (ethically sourced training data, fully open, etc)

000

Reuben Binns @rdbinns.bsky.social · 21/08/2026

"There’s no risk that CorridorKey is going to tell kids to self-harm in the way that ChatGPT does. Because CorridorKey has a specific job to do. And that’s the way AI should work — solving a specific problem for a particular community" www.anildash.com/2026/04/28/o...

anildash.com

(One) Good AI Is Here - Anil Dash

A blog about making culture. Since 1999.

141