Sign in

Kate Devlin

@drkatedevlin.bsky.social
16K followers 1.9K following 2.2K posts

Academic; writer. Professor of AI & Society, Chair-Director @kings-dfi.bsky.social, King’s College London. Eye-rolling at AI nonsense on a daily basis. #academicsky . Norn Irish in Norwich.

PostsRepliesMedia
Kate Devlin @drkatedevlin.bsky.social · 19h
They were so close to describing me; just that missing L.
A corner of a billboard. The text on it reads "Devin fuses intelligence with efficiency."
2180
Reposted by Kate Devlin
Dan of the Year @downwithdan.bsky.social · 28/09/2026
Ah yes the circle of life
• Dispatched
Sep 25 23:03 pm
In Transit
Sep 26 09:52 am
Delayed
Sep 26 10:57 am
Held By Carrier
Sep 26 15:07 pm
Destroyed
Sep 26 15:08 pm
12581251360
Reposted by Kate Devlin
Katie Martin @katie0martin.ft.com · 25/09/2026
I have blamed The agents That hacked Your website And which You probably Reasonably thought I controlled Forgive me I am rich And unaccustomed To taking responsibility
5776082089
Kate Devlin @drkatedevlin.bsky.social · 26/09/2026
To clarify: the ELIZA effect (1960s), the Computers Are Social Actors (1990s) thing... most people interact as if the machine is like us, and *that's okay* — convenient, and what we social creatures do — as they also simultaneously know it's a machine. Hyping the idea that it's more is unhelpful. ⬇️
2203
Kate Devlin @drkatedevlin.bsky.social · 26/09/2026
Hey you guys, Weizenbaum called and he said: "how easy it is to create and maintain the illusion of understanding, hence perhaps of judgment deserving of credibility. A certain danger lurks there."
04915
Reposted by Kate Devlin
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.
5742901171
Kate Devlin @drkatedevlin.bsky.social · 25/09/2026
I mean... we've just surveyed a representative sample of the UK and I have bad (good) news for you about Gen Z's attitudes to AI
1346
Reposted by Kate Devlin
Smooth Dunk @smoothdunk2.bsky.social · 25/09/2026
the future tech bros are dreaming of
4 panel comic

Panel 1:
Purple Guy panicking in a room full of flames. He screams “Oh God! I’m going to die!”

Panel 2:
Purple Guy is now on the phone (still surrounded by flames). He yells “Fire Service!! My house is burning down”

Panel 3:
Purple Guy (still burning to death) is now hollering “My address? Shit I don’t know! Lemme check my phone”

Panel 4:
Close up on Purple Guy’s phone screen. It reads: Emergency call detected. Pay £45.50 to view address.
4548161320
Kate Devlin @drkatedevlin.bsky.social · 25/09/2026
I have a 16yo child. This involved 5yrs of totally lonesome single parenting from 2015-2020. We are at the stage where they go unbidden to the local Tesco to get ice cream for us. This feels like an excellent return on investment.
3801
Reposted by Kate Devlin
Fr. Robert Ballecer, SJ @padresj.bsky.social · 25/09/2026
There is so much disinformation about the "Super El Nino" floating around the internet b/c mentioning any sort of mechanism that affects weather seems to start a shouting match. So here's a good video that gives you JUST the info you need in under 2 minutes.
631420756
Kate Devlin @drkatedevlin.bsky.social · 25/09/2026
This whole thing of LLM decisions without consent yet again makes me wonder when we'll see the first "Have YOU been misjudged by an AI?" advert along the lines of mis-sold car finance no-win, no-fee lawyers.
78112
Kate Devlin @drkatedevlin.bsky.social · 25/09/2026
Eh, what now? This is awful. This is unacceptable and goes against UKRI policy, which clearly states: "Assessors...must not use generative AI tools as part of their assessment activities except for language refinement purposes".
47426
Reposted by Kate Devlin
Diane Coyle @dianecoyle1859.bsky.social · 24/09/2026
This is not AI agents going rogue, but OpenAI. This hack took the company months to notice and then they notified a public email address rather than the appropriate government agency. www.ft.com/content/5613... OpenAI ‘agent’ hacked an Australian health service website
ft.com
OpenAI ‘agent’ hacked an Australian health service website
AI lab took months to spot the problem, which Prime Minister Anthony Albanese called ‘obviously unacceptable’
913945
Kate Devlin @drkatedevlin.bsky.social · 22/09/2026
I'll tell you why I don't fear the machines: it is because in this, the year of our lord twenty twenty six, despite us burning the planet for compute, it remains impossible to simply one-click share an event in a Google calendar so it becomes an event in an Outlook calendar and vice versa.
9716
Reposted by Kate Devlin
Cynethryth @cynethryth.com · 22/09/2026
Looks like the video is here youtu.be/kErHiET5YPw
youtu.be
Maciej Ceglowski - Superintelligence: The Idea That Eats Smart People (Keynote)
YouTube video by WebCamp Zagreb
0114
Kate Devlin @drkatedevlin.bsky.social · 20/09/2026
Finished last weekend's ultramarathon with a sore lower back. "Obviously you rested it and are now better, right?" you might ask. No. No I did not. In fact, I overdid things and made it worse and now have a hot water bottle strapped to me while I eat painkillers and attempt gentle stretching.
2120
Reposted by Kate Devlin
Nathan Bulmer @natemorebikes.bsky.social · 18/09/2026
2874841579
Kate Devlin @drkatedevlin.bsky.social · 18/09/2026
There's a woman further up the train from me rambling on and boringly on to her son about every second of her journey today, and I was tolerantly thinking "ahhh, nice that she has someone to listen" but then she described something that she thought was stupid as "a bit Irish" so she can fuck off.
2270
Kate Devlin @drkatedevlin.bsky.social · 17/09/2026
There is NOT a 'big spoon' and a 'little spoon', there is an outer spoon and an inner spoon of the same size because that is how they nestle together best. I will not be taking further questions at this time.
6819
Reposted by Kate Devlin
Kate Devlin @drkatedevlin.bsky.social · 17/09/2026
Us for years: "tech companies should be more responsible" Tech for years: "move fast and break things!" Tech now: "someone please stop us" Us now: "we've been asking you to be more responsible for years" The media now: "why are you criticising these tech guys they are so scared and afraid?"
515246
Reposted by Kate Devlin
Shannon Vallor @shannonvallor.bsky.social · 17/09/2026
Truly astonishing that all any world leader can think to do is keep inviting the same guys into a room to give their considered advice on how we can protect ourselves from all the harmful, unsafe things deliberately done by the guys in the room
bbc.co.uk
King Charles warns of 'existential danger' of AI falling into wrong hands
The comments were made at a summit in Ayrshire with participants from AI giants including Nvidia, OpenAI and Anthropic.
610627
Reposted by Kate Devlin
Philip Ball @philipcball.bsky.social · 15/09/2026
In thisa piece for @thenewworldmag.bsky.social I express the wish for a better level of public discourse around AI that doesn't leave us hanging on every word of the alleged tech gurus. www.thenewworld.co.uk/philip-ball-...
thenewworld.co.uk
Apocalyptic AI fantasies are a stupid, annoying distraction from its real problems
The people who work in AI are not gurus and most of their predictions have been wrong. Working in the industry makes them less reliable commentators, not more
713060
Reposted by Kate Devlin
John Hill @jpjhill.bsky.social · 17/09/2026
Say what you like about 2008, but back in the day, when some creep with no house training decided to rip people off and crash the economy out of greed, arrogance and general contempt for others, at least they didn’t clog up the news cycle every other week to cry about how maybe they were “too smart”
0173
Kate Devlin @drkatedevlin.bsky.social · 17/09/2026
Us for years: "tech companies should be more responsible" Tech for years: "move fast and break things!" Tech now: "someone please stop us" Us now: "we've been asking you to be more responsible for years" The media now: "why are you criticising these tech guys they are so scared and afraid?"
515246
Kate Devlin @drkatedevlin.bsky.social · 17/09/2026
Well, hey, I'm entirely in favour of an AI slow-down too, I'm just not convinced that tech companies are calling for it out of the goodness of their own hearts.
5437
Reposted by Kate Devlin
Brigitte Nerlich @bnerlich.bsky.social · 13/09/2026
Yep, 2023 was the year makingsciencepublic.com/2023/06/15/a...
makingsciencepublic.com
Artificial intelligence and existential risk: From alarm to alignment
1956 was a momentous year: I and AI were born. Ok, I was born and artificial intelligence was defined as a field of research in computer science. A lot has happened since, especially over the last …
022
Kate Devlin @drkatedevlin.bsky.social · 13/09/2026
"News framing shapes our perceptions; done well, it’s an important contribution to public understanding of the technology – and we need that. Done badly, and it perpetuates the dystopia." I wrote this more than three frickin' years ago and it still applies: bylinetimes.com/2023/06/09/r...
bylinetimes.com
Rogue Drones and Tall Tales
Kate Devlin dispels the sudden Science Fiction panic around superintelligence, and looks at the real threats to employment and the environment from AI and machine learning
1113
Reposted by Kate Devlin
Kate Devlin @drkatedevlin.bsky.social · 11/09/2026
Me for the ipaper: "it’s likely they do believe this, viewing themselves as the Oppenheimers of the modern age. They see the race to develop bigger and better models as something that must be done despite potentially lethal consequences" inews.co.uk/opinion/worr...
inews.co.uk
Worried about the end of humanity thanks to AI? Here's what's really happening
It's not superintelligence we need to worry about, it's the people who are recklessly pressing ahead with powerful new systems
65230
Reposted by Kate Devlin
michael lutz @ztul.bsky.social · 13/09/2026
Barber CEO: You Risk Falling Behind If You Don't Get A Haircut "Right Now" Industry Experts Foresee Future In Which Everyone Receives Multiple Haircuts Every Day Today Your Hair, Tomorrow The World: Barbers Say They Are On Track To Build A Pair Of Scissors So Big, It Could End Life As We Know It
142237644
Reposted by Kate Devlin
Emily M. Bender @emilymbender.bsky.social · 13/09/2026
So how about this for a "pause AI" proposal: "pause credulous AI reporting". No more media about "breakthroughs" or "rogue agent swarms" or ... UNLESS and UNTIL that media demands the actual empirical evidence and/or focuses on the motivations of & harms by these companies.
4744196
Kate Devlin @drkatedevlin.bsky.social · 12/09/2026
I did my first ultramarathon today. 51.3K (they rerouted and added extra) and I enjoyed it mostly and didn't break any bones, unlike my last marathon. I'm knackered and desperately in need of something that isn't sugar but worryingly I haven't said that I'd never do it again.
A medal on a yellow ribbon. The medal says "Suffolk Coastal Path 50K". It sits beside a bag of fish and chips.
13440
Reposted by Kate Devlin
WE’VE BEEN HERE BEFORE @priorartdaily.bsky.social · 11/09/2026
The emergency crumb catcher was a spacesuit pocket. In 1965, John Young brought a corned beef sandwich aboard Gemini III. It started breaking up, so Gus Grissom put it away in his suit. At least your laptop has gravity on its side. Source Transcript: NASA
Cropped scan of page 45 of the April 1965 Gemini III flight transcript, showing mission times 01:52:26–01:52:54. C is commander Gus Grissom; P is pilot John Young. Grissom: “What is it?” Young: “Corn beef sandwich.” Grissom: “Where did that come from?” Young: “I brought it with me. Let’s see how it tastes. Smells, doesn’t it?” Grissom: “Yes, it’s breaking up. I’m going to stick it in my pocket.” Young: “Is it?” then “It was a thought, anyway.” Grissom: “Yep.” Young: “Not a very good one.” Grissom: “Pretty good, though, if it would just hold together.” Source: NASA, publicly released mission transcript. Original scan cropped only; text not altered.
121
Reposted by Kate Devlin
Pavel @spavel.bsky.social · 11/09/2026
If Hugo continues to grow at this rate, by 2030 he will no longer fit in the barn. By 2050 he will have grown to the size of Godzilla and could crush humanity. But if we let the Chinese grow their own Hugo first that would be bad. This is why we must invest 100 trillion dollars in Hugo immediately.
1450396
Kate Devlin @drkatedevlin.bsky.social · 11/09/2026
I kinda miss the days when people just wanted to ask me questions about sex robots and then shout at me about that.
3260
Kate Devlin @drkatedevlin.bsky.social · 11/09/2026
Also did a Q&A for @theconversation.com here with a Betteridge's Law headline: theconversation.com/could-ai-rea...
theconversation.com
Could AI really kill off humanity within the decade? Expert Q&A
It’s not the first time AI researchers and companies have issued such warnings.
1117
Kate Devlin @drkatedevlin.bsky.social · 11/09/2026
Shutting down conversations about the Bayeux Tapestry by telling them "I'm not really into embroidery".
1486
Kate Devlin @drkatedevlin.bsky.social · 11/09/2026
You call it "lying on the sofa trying not to get Wotsit crumbs on the laptop", I call it "tapering and carb-loading".
2250
Reposted by Kate Devlin
Hetan Shah @hetanshah.bsky.social · 09/09/2026
'Humanities & social sciences research is central to ask questions about how and in whose interests technology is being developed, how it can be adopted, deployed & governed and how to create a world where AI can benefit the many, not the few' My blog for @ukri.org ukri.substack.com/p/why-ai-wil...
ukri.substack.com
Why AI will fail without the humanities and social sciences
AI can transform society. But only if society has a say in shaping it, writes Hetan Shah, Chief Executive of The British Academy.
59943
Kate Devlin @drkatedevlin.bsky.social · 11/09/2026
Me for the ipaper: "it’s likely they do believe this, viewing themselves as the Oppenheimers of the modern age. They see the race to develop bigger and better models as something that must be done despite potentially lethal consequences" inews.co.uk/opinion/worr...
inews.co.uk
Worried about the end of humanity thanks to AI? Here's what's really happening
It's not superintelligence we need to worry about, it's the people who are recklessly pressing ahead with powerful new systems
65230
Reposted by Kate Devlin
Brigitte Nerlich @bnerlich.bsky.social · 11/09/2026
This week's post is about the myths and metaphors that emerged within and about AI models after the Hugging Face incident #AI #metaphors #AIMetaphorObservatory makingsciencepublic.com/2026/09/11/f...
makingsciencepublic.com
From rogues to collectives: Myths and metaphors after the Hugging Face incident
Some weeks ago, I wrote a blog post just after reports about AI ‘going rogue’ had spread in the media. This referred to an incident at OpenAI that one can summarise like this: In July 2026, experim…
0116
Reposted by Kate Devlin
Melanie Mitchell @melaniemitchell.bsky.social · 10/09/2026
I wrote down my thoughts about the last several weeks of AI hell. aiguide.substack.com/p/misleading...
aiguide.substack.com
Misleading Metaphors, Real Risks
What To Fear from AI Agents and How to Reclaim Our Human Agency
1019880
Reposted by Kate Devlin
Gordon Guthrie @foundationsofthedigitalstate.com · 09/09/2026
As everyone is losing their mind over <another> AI is alive and going to kill us scare, it is time to point y'all to this by @rachelcoldicutt.bsky.social again hdsr.mitpress.mit.edu/pub/wz35dvpo...
hdsr.mitpress.mit.edu
AI Safety Is a Narrative Problem
0118
Reposted by Kate Devlin
Dermot Casey @dermotcasey.bsky.social · 10/09/2026
Hadn't seen this paper before but the way our media and political class who are still on Twitter are having their perspectives shifted rightward by a literal evil billionaire with space rockets seems so absurd that its hard to fathom.
02914
Kate Devlin @drkatedevlin.bsky.social · 10/09/2026
Did two very contrasting radio interviews on AI destroying all humans: Nicky Campbell on 5Live, where the discussion was a decent "let's think the issues through" perspective and Radio Ulster's Talkback where they were deliberately leaning into the panic and I hope you could hear my eyerolls.
3312
Kate Devlin @drkatedevlin.bsky.social · 09/09/2026
A lot of news coverage on AI doom based off one tweet on a shitposting platform by a prodigal techbro. Do I fear superintelligence wiping us out? No, I fear incompetence and probably climate disaster, both of which could be somewhat mitigated by making tech companies more responsible & accountable.
18914
Reposted by Kate Devlin
Isilanka @isilanka.bsky.social · 09/09/2026
It's really funny how OpenAI talks like they're running project Manhattan while apparently having no concept of things like "maybe don't connect your testing environment to the Internet" or "what do you mean we are supposed to keep an eye on our chatbots"
224423
Reposted by Kate Devlin
Diami @diami.myatproto.social · 09/09/2026
My school has had 5,999 conversations about AI in the past year and exactly zero about fascism.
1143
Reposted by Kate Devlin
The Policy Institute @policyatkings.bsky.social · 04/09/2026
NEW w/ @ukandeu.bsky.social: Reform UK supporters are imagined to be more hostile to rejoining the EU and keener on low taxes for the wealthy than they really are. They're also viewed as the group of voters most driven by dislike of others and the party people most want to block from power
kcl.ac.uk
Reform UK supporters misunderstood on EU and taxes on wealthy – and seen as most driven by dislike of others, study finds | King's College London
Reform is also the party people most want to block from power
071
Reposted by Kate Devlin
King Algamemnon @adamrothman.bsky.social · 03/09/2026
Sobering, insightful essay on how chatbots mangle historical primary sources. 🗃 www.hnn.us/article/ai-i...
hnn.us
AI Is an Anti-Primary Source Machine
Real voices from the past are threatened by chatbots designed to amplify — and ultimately become — the loudest voices in the room.
9421182
Reposted by Kate Devlin
Jamie McKelvie @mckelvie.bsky.social · 03/09/2026
A reminder that a ban on social media for under 16s is also this inevitably happening to the rest of us.
3296167