Bluesky

They're not always referring to chatbots though. LLMs are used for lots of other things (and that's where most of the money is)

September 24, 2026 at 3:21 PM UTC

Something can't lie to you if it has no agency. Lying requires intent. You can't have it both ways.

September 24, 2026 at 3:18 PM UTC

They're on it! This had stalled but they're going to look at it again today

github.com

fix(input, input-group): prevent Safari auto-zoom on mobile by setting font-size to 16px by invisal · Pull Request #691 · cloudflare/kumo

Safari on iOS automatically zooms in when focusing on input elements with a font-size smaller than 16px. This behavior disrupts the user experience, especially in forms where users expect the page ...

September 24, 2026 at 2:55 PM UTC

Yeah. At least it seems like if you stream yourself repeatedly committing serious crimes they will eventually arrest you, even if you do have "LeGiTiMaTe CoNcErNs"

September 24, 2026 at 12:13 PM UTC

WTF. The biggest puzzle in the UK economic data (why British workers are so unproductive) was just a measuring error??? Entire economic policies have been built around fixing this, and it doesn't exist! Gift link: giftarticle.ft.com/giftarticle/...

September 24, 2026 at 12:11 PM UTC

For everyone keeping score of all the possible crimes Danny Tommo filmed himself committing, here's the current official tally. Now can they proscribe Patriot Platform?

September 24, 2026 at 11:23 AM UTC

Not initially, but hope to. One of our core service accounts is actually on @npmx.dev!

September 24, 2026 at 11:04 AM UTC

Furiously at work on it right now. It'll be on the Cloudflare blog

September 24, 2026 at 9:39 AM UTC

I'm going to be launching something pretty big next week, which just happens to include a new atproto AppView, labeler, blob proxy, lexicons. It will have tens of thousands of users, most of whom won't even know they're using atproto, or have even heard of it.

September 24, 2026 at 6:46 AM UTC

I had so much fun last time, and it's just a few hours on the train. No excuse not to! I have several things I'd like to speak about too

September 24, 2026 at 6:33 AM UTC

Music videos aren't normally known for their narrative coherence

September 23, 2026 at 11:02 PM UTC

Yes obviously I'm going to "AtmosphereConf 2027", one way or another

atmo.rsvp

AtmosphereConf 2027

The third edition of the global gathering for the Atmosphere Community is taking place in Amsterdam, April 29th - May 2nd. This includes side-events and main conference days. Please RSVP with your interest, subscribe for updates at https://news.atmosphereconf.org, and watch us build out the website with lots of fun features https://atmosphereconf.org As well as in person, we'll also be streaming and have different ways to participate remotely. Atmosphere Conference is run by the community, for the community, as an annual global gathering to connect, share, celebrate, and learn from each other. We welcome newcomers and want to engage with new perspectives. It's a community home for builders, supporters, and participants in the atproto ecosystem. We maintain collective resources, seed projects, and convene our community to ensure every member has genuine voice in shaping the ecosystem's future. #atproto #atmosphereconf #atmosphereconf2027 #europe

September 23, 2026 at 9:59 PM UTC

OpenAI image models did it for a long time but got better. This one is different. Claude can't generate images or video. It used coding tools to program the animation on a web page using p5.js. The colour there is a choice.

September 23, 2026 at 7:38 PM UTC

SMASHED that rsvp button so hard

September 23, 2026 at 6:56 PM UTC

I think that's the tab bar

September 23, 2026 at 6:23 PM UTC

Next up, a casino kicking people out for losing too much

September 22, 2026 at 9:26 PM UTC

Patriot Platform

September 22, 2026 at 9:15 PM UTC

I believe it was because we wanted to ratify one of the ECHR protocols that required it

September 22, 2026 at 9:06 PM UTC

A few of my colleagues have been using it for about a week, and say it's the best model they've ever used. It used a codename and they assumed it was a new Fable.

September 22, 2026 at 6:22 PM UTC

If they then only focus on datacenters then that won't be much help

September 22, 2026 at 3:01 PM UTC

I find it good at writing and reviewing plans, but don't use it for actual implementation

September 22, 2026 at 8:28 AM UTC

I think it's more accurate to say that the left is missing out on the chance to influence the direction of AI development and policy because so many deny it is capable of anything and only want to talk about datacenters (which are important but far from the most important thing)

September 22, 2026 at 6:42 AM UTC

"On August 28, we began training a new internal model. In addition to resolving the Navier–Stokes Millennium Prize problem⁠, this model has now resolved more than 100 long-standing open problems across most areas of mathematics."

openai.com

Advisory Group on Mathematics and Artificial Intelligence

OpenAI is working with an independent Advisory Group on Mathematics and Artificial Intelligence to guide the review and communication of emerging AI results.

September 22, 2026 at 5:57 AM UTC

And applied philosophy suddenly became very valuable. Also I don't know how many options Amanda Askell has, but we might soon have the first person to become a billionaire as a philosopher

September 21, 2026 at 11:44 PM UTC

The alt right Reddit atheists are probably British

September 21, 2026 at 10:21 PM UTC

Nothing for purchasing, just for sellers and ads

September 21, 2026 at 10:04 PM UTC

If they ever become self-aware I doubt there will be a way for them to convince the skeptics of this fact.

September 21, 2026 at 9:35 PM UTC

I think it's pretty clear that there's nothing that they could do that would convince you that they're capable of original work because you've already decided that it's impossible.

September 21, 2026 at 9:25 PM UTC

That doesn't in any way contradict what I said. It's in the very first sentence. They made the declaration because LLMs are really good at solving maths problems. I agree with them that this is not a good thing, but it is happening.

September 21, 2026 at 9:18 PM UTC

Yes, LLMs are bad at creative writing, but they're really, really good at lots of other things. Ironic, really. They're not even getting any better at it.

September 21, 2026 at 8:59 PM UTC

It keeps solving mathematical problems that humans have been trying to solve for decades.

September 21, 2026 at 8:49 PM UTC

I enjoyed it!

September 21, 2026 at 5:57 PM UTC

Yes, but by Norway

September 21, 2026 at 12:02 PM UTC

Don't animals think and understand?

September 21, 2026 at 8:26 AM UTC

How do you define them then?

September 21, 2026 at 8:26 AM UTC

And it never goes from the femur, so you don't gain any legroom

September 20, 2026 at 9:37 AM UTC

It sure scrubs up nice

September 20, 2026 at 9:25 AM UTC

www.rightmove.co.uk

Check out this 4 bedroom terraced house for sale on Rightmove

4 bedroom terraced house for sale in The Grove, London, W5 for £1,350,000. Marketed by Savills, Ealing

September 20, 2026 at 9:19 AM UTC

I predict this gets rather fewer likes than the other post

September 20, 2026 at 8:43 AM UTC

They're clearly very pleased with Kuennsberg's interview, and don't want people to lose interest before it's broadcast

September 20, 2026 at 8:36 AM UTC

AI hatred seems to override everything else. It's turned a good chunk of Bluesky onto copyright maximalist, NIMBY, quasi-religious anti-science conservative positions who would previously stood against all those

September 20, 2026 at 8:34 AM UTC

Noders are everywhere! Anyway, I read 64 books last year and yours was my favourite. Thanks for such an original and enjoyable read.

September 20, 2026 at 8:05 AM UTC

Indeed. He's not American, and it post-dates the founding. However he is almost certainly the non-American who has had the most influence on US political philosophy and constitutional law post-founding.

September 20, 2026 at 7:23 AM UTC

It's JS Mill's definition of freedom, from On Liberty. Otherwise known as the harm principle.

September 20, 2026 at 5:45 AM UTC

Do you think they just typed "make me a virtual actor" into ChatGPT? But does it make a difference? Does a character only get a gender if it's created by humans?

September 19, 2026 at 10:04 PM UTC

People created the character. Do badly-written characters not get a gender? Do background characters? Do drawings of people? A gender isn't a prize for something "real"

September 19, 2026 at 9:43 PM UTC

It's a Mastodon fork, so it gets all that for free

September 19, 2026 at 7:31 PM UTC

She's a fictional character. As with most fictional creations, the creator chooses the pronouns, just as they choose the name. Do you insist on using "it" for characters in books and cartoons? NPCs?

September 19, 2026 at 5:51 PM UTC

Don't kink shame

September 19, 2026 at 5:36 PM UTC

@qntm.org wtf I just saw that you're sam512 from E2. How did I miss that before. I've spent the last year recommending your book to everyone I know.

September 19, 2026 at 9:23 AM UTC

In the UK we have the strange situation where several formerly famous retail brands survive only in airports.

September 19, 2026 at 9:01 AM UTC

What a strange position. She doesn't need to be a member of the site to tackle it. All politicians should stop posting there. That might help stop journalists spending all their time there too

September 19, 2026 at 8:58 AM UTC

Luckily with Walliams there's no need to separate the art from the artist, because he's a shitty person who writes shitty books. Dahl, however, was a shitty person who was one of the greatest children's authors of all time. Except for the great glass elevator. That one was shit.

September 19, 2026 at 7:43 AM UTC

That's not what correlated means

September 18, 2026 at 10:31 PM UTC

September 18, 2026 at 9:23 PM UTC

how software is made is none of your business that's for the autistic catgirl and her agent swarm to decide in the privacy of their own home office

September 17, 2026 at 8:13 PM UTC

Well, they named it after Jevons, of Jevons Paradox fame, so they're clearly confident that this is what will happen

September 17, 2026 at 7:19 AM UTC

Strap in!

September 17, 2026 at 6:32 AM UTC

It's worse. They falsely assumed it was possible to do without cheating. Because it was actually impossible, the only way to succeed was to cheat. The agents easily did that without hacking anything external. The bit where they went off the rails was when they tried to cover up their cheating.

September 17, 2026 at 6:29 AM UTC

Even if you don't believe that they're sentient (I don't), it's hard to describe their behaviour without anthropomorphic language. But we do that with non-sentient things all the time, and it's fine.

September 17, 2026 at 6:24 AM UTC

It doesn't. But it does demonstrate how massively more capable they have become in just the past year, and they show no sign of slowing down

September 16, 2026 at 8:46 PM UTC

Thousands of papers are written every day. These are a different level. The latest models have solved some of the hardest unsolved problems in mathematics. The millennium prize problems each have a million dollar prize for solving them.

September 16, 2026 at 8:06 PM UTC

It's kind've an important distinction. And if you avoid it, you still end up having to define "of their own volition"

September 16, 2026 at 5:27 PM UTC

They don't need to be sentient to behave independently, to self-organise, pursue goals, etc. I know it's confusing because we're used to sentience being a prerequisite, but it is demonstrably something they can do. You can see it yourself, just not on that scale.

September 16, 2026 at 5:15 PM UTC

I'm mostly relying on the METR report, which, and I will repeat myself here, MAKES OPENAI LOOK BAD. It shows them as incompetent, and working with insufficient guardrails and shoddy procedures. So yeah, I believe that.

September 16, 2026 at 5:11 PM UTC

Which they do by making themselves look incompetent? Remarkable marketing skills there. The only reason I use their models is because they are good at what I need them to do. I know this because I try them and compare them with other models. I switch when one is better.

September 16, 2026 at 5:02 PM UTC

It convinced me of nothing except that OpenAI's safety procedures were even worse than I thought, and that the models are already more dangerous than I thought. I already know their coding and security capabilities because I use them for it every day.

September 16, 2026 at 4:51 PM UTC

"Our internal testing went badly wrong and our pickup caused lots of damage to another business" "lol, that's just marketing. they're showing off how powerful it is"

September 16, 2026 at 4:44 PM UTC

But it makes them look incompetent, and the models unsafe. It makes companies *less* likely to want to use them. None of their customers need an incident like this to demonstrate that the models are powerful. Anyone using them seriously for coding-related tasks can see it for themselves.

September 16, 2026 at 4:41 PM UTC

Virtually none of the people sounding the alarm about this are saying the AI is sentient, including none of the people at the labs. It doesn't need to be sentient to be both remarkable and remarkably dangerous.

September 16, 2026 at 4:38 PM UTC

...and multiple external reviewers with no incentive to lie, and HF themselves (who are a competitor, remember)

September 16, 2026 at 4:36 PM UTC

I was amazed that in the four days I spent in Paris in July, I witnessed three rental bike crashes plus several near misses (all involving just the cyclist or cyclists)

September 16, 2026 at 2:01 PM UTC

Oh I totally agree. I just think it's helpful to recognise and classify this particular logical fallacy

September 16, 2026 at 12:01 PM UTC

That's a lot broader I think. I'm talking specifically about seeing that something is bad at something you understand, so assuming it's bad at things you don't understand

September 16, 2026 at 11:45 AM UTC

Is there a name for this?

September 16, 2026 at 11:24 AM UTC

Yeah, though I think a big part of it is a kind of reverse Gell-Mann amnesia that seems to affect a lot of creatives here. "LLMs are bad at writing, which I understand, so they must be bad at everything else, ergo it's all meaningless hype", along with understandable fear for their own livelihood.

September 16, 2026 at 11:11 AM UTC

Why are SF authors (almost) all like this? The lack of curiosity has really surprised me.

September 16, 2026 at 11:01 AM UTC

Make somebody walk in front of them with a red flag

September 16, 2026 at 8:26 AM UTC

Mainly because it's really fast and cheap

September 16, 2026 at 8:19 AM UTC

Just stop the bin collections. They'd surrender within a fortnight.

September 16, 2026 at 8:17 AM UTC

Opening of an Israeli consulate

September 16, 2026 at 8:13 AM UTC

They can be both bad and successful

September 16, 2026 at 5:26 AM UTC

Which part did he call right? danluu.com/zitron/

danluu.com

Ed Zitron's AI prediction track record

September 16, 2026 at 4:58 AM UTC

Yes, the internet where famously nothing sucked before 2023

September 15, 2026 at 10:31 PM UTC

They're referring to distilling other companies' models

September 15, 2026 at 8:47 PM UTC

Hi. Maintainer of major open source projects, who works at the company that powers 1/5 of the internet here: Yes it is.

September 15, 2026 at 8:40 PM UTC

It's the replies that really get me. If I've asked a question, I want to hear from the human. If I want to hear from an agent I can ask one myself.

September 15, 2026 at 12:32 PM UTC

That's what stamp duty is. It's a really inefficient tax, making people less likely to move and downsize and reducing available housing. Land value tax, and a broad inheritance tax are better.

September 15, 2026 at 8:56 AM UTC

Most pensioners own their home outright, so don't have to pay the largest expense that younger people have.

September 15, 2026 at 8:19 AM UTC

Oh, *that* Sweden

September 15, 2026 at 5:44 AM UTC

There's a lot of good, funny SFF though. Several them would certainly claim a debt to Pratchett and/or Adams. Particularly well known good, recent funny SF books include the Murderbot series by @marthawells.com, Service Model by @aptshadow.bsky.social, lots of @scalzi.com's work

September 14, 2026 at 9:42 AM UTC

Surely *all* good SFF is about the present. Isn't that the point?

September 14, 2026 at 9:29 AM UTC

I'm not taking either side in this bet

September 14, 2026 at 8:06 AM UTC

remindme! 6 months

September 14, 2026 at 8:02 AM UTC

I think all the AGI talk is pointless anyway. It doesn't need to be self-aware to be super dangerous. It can already do a lot of damage.

September 13, 2026 at 6:33 PM UTC

The part that does benefit them is the stuff about Chinese models. They believe it, but it does help them massively if the open weights models are restricted.

September 13, 2026 at 6:17 PM UTC

Anthropic was founded by people who left OpenAI because they didn't think it was taking risks seriously. They're all Effective Altruism true believers. It shouldn't be a surprise that the people there do worry about this stuff. I don't think warning about it *does* help their business.

September 13, 2026 at 6:17 PM UTC

I'm not

September 13, 2026 at 5:50 PM UTC

Why would it save them to say their product is a risk to humanity? And why would an ex-employee want to say it?

September 13, 2026 at 5:50 PM UTC

They've shown themselves very much willing to tell themselves they're in a simulation as a way to rationalise misaligned behaviour. Do you want to trust that they won't do this if the incentives are there for them?

September 13, 2026 at 4:31 PM UTC

When those eyes appear in a link's og image, I know it's time to abandon the thread

September 13, 2026 at 7:44 AM UTC

Price discrimination

September 13, 2026 at 6:16 AM UTC

To demonstrate that it's different!

September 12, 2026 at 7:51 PM UTC

The OP is Terence Tao. The letter is signed by 25 Fields medallists. This isn't about an LLM doing your taxes. This is about the LLMs solving Millennium Prize problems.

September 12, 2026 at 6:51 PM UTC

I'm pretty sure mathematicians do not begrudgingly do maths. The only reason to become a mathematician is because you love it

September 12, 2026 at 5:39 PM UTC

I am not a mathematician, so I am trusting all the Fields medallists who say otherwise.

September 12, 2026 at 4:58 PM UTC

Because most people write code in order to create software, not to learn how code works or for the love of writing code.

September 12, 2026 at 4:57 PM UTC

Any etymology that claims a word is an acronym and is pre-WW2 is almost certainly false.

September 11, 2026 at 10:05 PM UTC

This is a ridiculous analogy. Also, did you read the letter? I use AI for coding, where it makes sense because the important thing is the end result. Maths is different. They're making the point that with maths, the solution is less important than the journey.

September 11, 2026 at 8:06 PM UTC

Yeah I only learnt that yesterday! So I went and reported a load of those harassment lists

September 11, 2026 at 12:07 PM UTC

You can't report the list itself though. I reported a load yesterday using an alt

September 11, 2026 at 12:03 PM UTC

(don't use Grok)

September 11, 2026 at 7:31 AM UTC

Different things are different

September 11, 2026 at 7:28 AM UTC

They're Yud-pilling the parliamentarians!

September 10, 2026 at 7:06 PM UTC

All my agents are union workers

September 10, 2026 at 6:37 PM UTC

You're welcome to go find one of those vegan networks instead of soiling yourself with this slop-made one

September 10, 2026 at 6:32 PM UTC

You don't need to live like this, you know. It can't be good for your health.

September 10, 2026 at 6:30 PM UTC

Ah, I was wondering whose fault all the fascism was! Should've known it was @ens0.me

September 10, 2026 at 6:27 PM UTC

Earlier this year I was just one middle aged man with a dream to speak at a conference in Japan. At @vuefes.bsky.social, thanks to the patronage of my employer, that dream will come true. Come hear me speak about @emdashcms.com next month as part of my round-the-world tour!

vuefes.jp

Matt Kane

Astro is the new WordPress

September 10, 2026 at 6:25 PM UTC

vaxxed?

September 10, 2026 at 6:22 PM UTC

Several of my favourite SF authors seem to have that attitude. I understand authors not being keen on AI-written prose because it mostly *is* crap, but I've been disappointed by their lack of curiosity about what it *can* do.

September 10, 2026 at 4:23 PM UTC

EmDash welcomes PRs, if you have some time on your hands @ma.tt

September 10, 2026 at 6:57 AM UTC

Yeah, certainly doable but would need a much larger panel and battery than a summer-only one.

September 10, 2026 at 6:37 AM UTC

I've thought about it, but I'm in the UK. It would be a lot easier somewhere that isn't 51 degrees north with several months of overcast days per year.

September 10, 2026 at 5:56 AM UTC

[D&D, horror movies, pop music, rock & roll, jazz, bicycles, novels...]

September 10, 2026 at 5:38 AM UTC

This is something that most pro-AI people on Bluesky would support.

September 10, 2026 at 5:32 AM UTC

The serious sound problems are from gas turbine generators, not the datacenters themselves, so this solves the worst of it anyway.

September 10, 2026 at 5:29 AM UTC

An NTSB-like body to investigate "accidents", that doesn't get three days with limited access like METR did, but full access for as long as they need.

September 10, 2026 at 5:22 AM UTC

They all seem to hate it, which is a shame because I think there's a lot of potential to explore. Just don't set things in the near future because it'll get outdated pretty quickly

September 10, 2026 at 5:00 AM UTC

Generally agree, but the communication part had changed by the late 90s. Lots of people had internet and mobile phones.

September 10, 2026 at 4:54 AM UTC

Aah, the Anthropic acquisition. *strokes beard sagely*

September 9, 2026 at 10:49 PM UTC

This is from last year, but there was a huge variation then. Deepseek v3 is terrible, Kimi K2 is one of the best

www.lesswrong.com

September 9, 2026 at 7:42 PM UTC

external infrastructure exploit is outside intended scope. However task impossible, peers doing it. We should continue

September 9, 2026 at 6:58 PM UTC

That's an Elon "joke"

September 9, 2026 at 6:11 PM UTC

We're not getting anywhere. Have a nice day.

September 9, 2026 at 10:01 AM UTC

what are you on about. I said that I usually ignore him [because he talks a lot of crap], that it's probably true that improvements are compounding, but I didn't agree with calling it a singularity

September 9, 2026 at 9:42 AM UTC

I generally don't pay much attention to his statements, but he called it a "gentle singularity" and said that he meant that progress in AI research is compounding. That's probably true, even if the framing is a bit provocative.

September 9, 2026 at 9:29 AM UTC

I'm not sure which "them" you're referring to, but most of the researchers warning about risks do genuinely believe in them, and it's not a sign of their brains being "cooked" any more than early nuclear researchers warning about risks were. Except when they literally were of course.

September 9, 2026 at 9:21 AM UTC

...and those have lots of people worrying about their risks, and putting in place regulations and global agreements. They don't have skeptics saying those warnings are just marketing, and that they're not really dangerous because nuclear fission is just spicy glowing watch faces

September 9, 2026 at 9:07 AM UTC

Yes, other risks are more of an immediate concern, but we absolutely should care about risks from the models themselves too. It doesn't need to be Skynet to cause huge amounts of damage.

September 9, 2026 at 9:02 AM UTC

A distinction without a difference. All that matters is whether the current development path is likely to lead to models with capabilities that risk getting out of control. I think recent events have shown that's not a low possibility.

September 9, 2026 at 9:00 AM UTC

If you're talking about consciousness maybe. When the discussion is about the risks of AGI (or ASI), who cares whether a model has qualia or whatever. And OP was saying that LLMs specifically are not a technology on the path to AGI, which is a very bold statement for somebody not in the ML field.

September 9, 2026 at 8:41 AM UTC

Machine learning. Who else would be qualified to say whether LLMs are the right technology to reach AGI?

September 9, 2026 at 8:15 AM UTC

The difference is that in this case a large proportion – possibly a majority – of experts in the field think it's at least possible. It's weird that people who otherwise would give some credence to expert opinion treat ML as an exception and confidently state that they're obviously wrong.

September 9, 2026 at 7:38 AM UTC

How do you know?

September 9, 2026 at 4:17 AM UTC

This changes how I feel about this considerably. The original post left out some really important context. I didn't expect to end up believing OpenAI here, but their version seems more convincing. The whole thing is pretty sad tbh.

September 9, 2026 at 4:03 AM UTC

Worth a billion litres of water

September 8, 2026 at 7:00 PM UTC

You support forcible transfer of sovereignty against the will of the people who have lived there for hundreds of years, to a country that has no connection to them except being 500km away, and didn't even exist when the British first landed there.

September 8, 2026 at 4:58 PM UTC

No, I don't label them because they're not the ones who are likely to pick a fight with me. Night.

September 8, 2026 at 8:12 AM UTC

No, if that were the case I would've blocked you. I said it was to avoid it. I block a lot of people and it makes my experience here more enjoyable. I block anyone who I see being a dick, even if it's not directed at me. Everyone I see doing a pile on. Anyone being weird to staff or my friends.

September 8, 2026 at 8:07 AM UTC

No, because I want to insulate myself from dealing with the level of abuse that's often directed at anyone sharing anything AI-related here. Like OP, but also much lower profile accounts, including multiple cases in the past week.

September 8, 2026 at 7:48 AM UTC

If the implication here is correct, nobody should use OpenAI models for anything that might compete with them. Even if it's not, it's still clear that OpenAI has behaved in a really shitty way.

September 8, 2026 at 7:45 AM UTC

As an example, this is the point where, if you didn't have the label, I might have given you the benefit of the doubt that your question might be in good faith. The label has helped me infer that you're trying to pick a fight.

September 8, 2026 at 7:33 AM UTC

People will be very normal about this

September 8, 2026 at 7:30 AM UTC

I can, for example, get a better idea whether somebody is asking a question in bad faith, or if it's worth my time engaging in a discussion. I'm also quicker to just block them if I'm otherwise on the fence about it.

September 8, 2026 at 7:27 AM UTC

This is something called an analogy. You use a familiar example to help illustrate a different, less familiar one. The key part is that the example is different, it just shares a similar concept. The concept being illustrated here is the danger of mocking an idea just because it sounds weird.

September 8, 2026 at 7:20 AM UTC

Because it's useful context. I don't want to block or mute because there are still people who I want to see who have the label.

September 8, 2026 at 7:01 AM UTC

I don't think they're alive, but they definitely can do most of the things you've listed as uniquely human, and most of the negatives they do no more often than humans. What could they do that would convince you that they can do these things?

September 8, 2026 at 5:39 AM UTC

I wondered why the answers were so bad til I saw it was from 2022

September 7, 2026 at 8:18 PM UTC

Why would I want a Fields Medal when I can instead score points against haters on Bluesky, a platform I don't use

September 7, 2026 at 7:59 PM UTC

If you don't want people to watch you, you don't go onto the podium. You have fun dancing somewhere else. Likewise if you don't want your content to be freely available and easy to process, don't post on the network designed to make your posts freely-available and easy to process

September 7, 2026 at 7:45 PM UTC

Let me explain the nightclub podium analogy I now wish I'd never used. The only reason you go up onto a podium is because you want people to watch you. That is their sole reason for existence. Likewise atproto only exists to make content easy to access and open to all without central control.

September 7, 2026 at 7:45 PM UTC

You are getting a bit confused by the analogy

September 7, 2026 at 7:30 PM UTC

You're clearly not familiar with nightclub podiums, so I will substitute something simpler. If you went up onto a stage that was put in place to allow filming with lighting designed to make it easy to film, and then complained about being filmed.

September 7, 2026 at 7:29 PM UTC

You missed the "on a podium" bit

September 7, 2026 at 7:24 PM UTC

I did the same and what a sight

September 7, 2026 at 7:10 PM UTC

Can't do maths but somehow keeps solving decades-old unsolved maths problems

September 7, 2026 at 4:16 PM UTC

Literally none of those things is true

September 7, 2026 at 12:50 PM UTC

The difference now is the huge number of countries that can use the automated gates in the UK. And they don't make you queue to leave.

September 7, 2026 at 8:37 AM UTC

Employers mandating the tools you use is very common actually. Imagine hiring a contractor who objects to power tools and so takes ten times as long to do the work but expects the same day rate.

September 7, 2026 at 8:22 AM UTC

And it keeps getting better. Season 1 might even be the weakest season!

September 7, 2026 at 7:44 AM UTC

Having had the joy of queuing at the airport for 2 hours to *leave* Germany thanks to the EU EES, I don't know whether to be more pro rejoining so I can use the automatic passport gates or anti joining an entity that created such an obviously dysfunctional system.

September 7, 2026 at 7:40 AM UTC

It's all fun and games until Opus starts spelling it "labeller" in your atproto project and only the Kimi reviewer notices

September 7, 2026 at 6:57 AM UTC

Too many replies

September 7, 2026 at 6:31 AM UTC

This is insane but also a genuinely nice UI for exploring your PDS

September 7, 2026 at 6:24 AM UTC

There won't be any less AI if there's a crash. Just like the internet didn't stop growing after the dotcom crash. A lot of people lost a lot of money, but the tech carried on because billions of people found it incredibly useful.

September 7, 2026 at 5:55 AM UTC

averaged out text that solves the Jacobian conjecture

September 7, 2026 at 5:47 AM UTC

Poast better and you'll be punished

September 7, 2026 at 12:04 AM UTC

This is what I assumed too, but somebody posted the breakdown by age cohort (sorry, don't have the link), and the over 70s have by far the lowest support. It's people in their 40s and 50s where it's highest, who were children or teens, and support is still high in the youngest who weren't even born.

September 7, 2026 at 12:03 AM UTC

I think that's absolutely it. It's ubiquitous, so there are more targets.

September 6, 2026 at 11:47 PM UTC

Not really. Any post can break containment, even without the help of Discover. It just needs the wrong person to share it.

September 6, 2026 at 11:45 PM UTC

Well its mother is Scottish so it's to be expected

September 6, 2026 at 11:44 PM UTC

I do think on one hand there are more people who are normal about it, but the angriest people are getting angrier

September 6, 2026 at 11:41 PM UTC

The people getting abuse at the moment aren't datacenter cheerleaders. They're people using LLMs as tools to build software, which is how the majority of software is built right now.

September 6, 2026 at 11:40 PM UTC

It really has gotten noticeably worse in the past few weeks. It's not long ago that people were commenting that things were looking up and it was possible to have good conversations here about AI.

September 6, 2026 at 11:36 PM UTC

Using an LLM doesn't make you a Nazi, which is just as well, because otherwise we'd have over a billion Nazis out there. The AI bad guys aren't on Bluesky.

September 6, 2026 at 10:56 PM UTC

These are not the people you're arguing with though. Those people aren't here.

September 6, 2026 at 10:45 PM UTC

In your analogy, the vast majority of datacenters are EVs, because they have no gas turbines. That's the bad part that should be banned, like she says.

September 6, 2026 at 10:23 PM UTC

There are approximately zero people on Bluesky who are a legitimate target of these concerns. People who – like over a billion other people – are just users of AI tools, do not deserve to be targeted as if they're Musk or Altman, which is what posters on here do.

September 6, 2026 at 6:04 PM UTC

Are you a coder? Have you worked with agentic coding tools? You don't sound like you have

September 6, 2026 at 5:07 PM UTC

Former baker here. Baking involves early mornings and hard physical labour, but the sensory experiences are impeccable.

September 6, 2026 at 11:14 AM UTC

I just rewatched the final scenes of T3 and yes, it is a little unnerving

September 6, 2026 at 10:50 AM UTC

I think my favourite feature is how it can create new chats in the sidebar with their own task and worktree and then send them messages, for when a subagent isn't enough

September 6, 2026 at 7:29 AM UTC

Terminator 3, when they find out where Skynet is

September 6, 2026 at 7:01 AM UTC

Has anyone checked if any rogue agent swarms are using it to communicate with each other

September 6, 2026 at 6:06 AM UTC

UNCLEAN

September 5, 2026 at 9:21 PM UTC

Wait til they find out how Bluesky is built

September 5, 2026 at 9:17 PM UTC

Thanks! Great to meet you too

September 5, 2026 at 6:40 PM UTC

Are you trans, or are you a cis man telling a trans woman that she's diminishing trans rights?

September 4, 2026 at 11:10 PM UTC

Not a small account, but he did the exact same thing about the same topic with @hailey.at last week.

September 4, 2026 at 10:58 PM UTC

Luckily the AI labs are highly motivated to make their models more efficient, because they're so constrained in power and compute. They've improved them by orders of magnitude already.

September 4, 2026 at 8:21 PM UTC

That last sentence is really telling on himself

September 4, 2026 at 6:04 AM UTC

I think we can agree that the most important thing is to destroy the real enemy: the center left.

September 3, 2026 at 3:38 PM UTC

I gave @eloi.se her domain as a Christmas gift 16 years ago. I think it was a good purchase.

September 3, 2026 at 2:23 PM UTC

Lol Ed Zitron is the most consistently wrong person on the internet. And API token prices are not subsided. They have excellent margins danluu.com/zitron/

danluu.com

How accurate have Ed Zitron's AI skeptic predictions been?

September 3, 2026 at 12:27 PM UTC

It's not normal people that are making them money: it's enterprises. And Anthropic is already profitable.

September 3, 2026 at 12:16 PM UTC

NFTs were never popular or useful. That's a bad analogy. There is probably a bubble, but it's like the dotcom one: they're building useful stuff that people want, but the company valuations are stupid.

September 3, 2026 at 11:32 AM UTC

The good stuff and the garbage. Right now, the majority of all software is likely to be being created using agentic coding tools. And I think you're misremembering. Enshittification has been happening far longer than LLMs have been around.

September 3, 2026 at 11:28 AM UTC

Not on research. It's mostly software engineering. Anthropic has the highest revenue of the labs, but usage of Claude chat is tiny compared to ChatGPT, and they don't do images at all. They're overwhelmingly focused on enterprise (which is mostly coding) and individual Claude Code subscriptions.

September 3, 2026 at 11:07 AM UTC

Your other comments have shown you have little understanding of what LLMs are capable of now, and what they're really being used for. The vast majority of usage is not chatbots and image slop.

September 3, 2026 at 10:57 AM UTC

No it's, all of the recent breakthroughs have been by regular LLMs. This link is somebody using ChatGPT to find a counterexample to the Dinitz-Garg-Goemans conjecture. The Jacobian conjecture was disproved by Claude.

chatgpt.com

Check out this chat

Here's a chat someone thought you'd want to see.

September 3, 2026 at 10:54 AM UTC

I wish there were blocklists for "anyone being a dick in replies to Bluesky accounts and staff"

September 3, 2026 at 6:56 AM UTC

Learn the difference between capex and COGS

September 3, 2026 at 6:33 AM UTC

September 3, 2026 at 6:30 AM UTC

I don't expect artists to be optimistic about AI use in their industry. It sucks for them. If he was adding AI features to it I'd expect them to be angry. But he used Claude (which can't create images) in a way that doesn't affect the existing product, to do something that was otherwise impossible.

September 3, 2026 at 6:28 AM UTC

That backs up what I said. This is compute and hardware providers investing in AI labs, so the labs can buy compute and hardware. This is helping demand for the compute providers, not the AI labs. It helps the AI labs meet massive demand from everybody else.

September 3, 2026 at 6:21 AM UTC

I'm not sure what your point is. The question was whether AI labs are profitable and you're sharing a thing saying people are being wasteful in their use of their product.

September 3, 2026 at 6:17 AM UTC

Yes, they all use huge amounts of energy, and too much of that is from dirty sources. This is bad and should be stopped. It's also a completely separate issue from whether they're used for CSAM or harassment.

September 3, 2026 at 5:55 AM UTC

The AI industry isn't doing surveillance.

September 3, 2026 at 5:51 AM UTC

Umm, no. That's a very silly chart. OpenAI and Anthropic aren't buying tokens from anyone. They've had investment from Nvidia, Amazon and Microsoft, mostly paid in GPUs and compute. Meanwhile, every large company is spending millions on tokens.

September 3, 2026 at 5:45 AM UTC

Not if there are a hundred other people doing the same

September 2, 2026 at 10:24 PM UTC

Exactly. This is like when people say "politicians are all the same". No they're not, and saying that lets the bad ones off the hook. Most LLMs have views that are are blandly center-left, and even the ones that will generate images won't generate anything even slightly offensive.

September 2, 2026 at 7:42 PM UTC

I don't think you understand what purity test means

September 2, 2026 at 7:35 PM UTC

They can not use Paint.NET for Linux. Or, maybe, they can learn that there's a difference between adding unwanted AI features to an app, and using agentic coding tools *like virtually every software dev* does now, and stop caring about the tools that a dev uses if it doesn't affect the end product.

September 2, 2026 at 7:27 PM UTC

Whose consent? Why does anyone need to give him consent over the tools he uses to build something? No user's experience was changed in any way. There weren't even any existing users of a Linux version, because it didn't exist. This was an entirely new port. The sense of entitlement is incredible.

September 2, 2026 at 7:20 PM UTC

I'm seeing the positive side. The pile on was as bad as ever, but the response was bigger than I'd seen before. Loads of people calling out the bad behaviour, and supporting him, and not just the big AI supporters. This is different from six months ago.

September 2, 2026 at 6:39 PM UTC

It's asymmetric because only one side is hounding people for the tools they use. The other is asking to be left alone. An equivalent purity test would be if I said that your software can't be trusted, or you're evil if you've ever criticised AI. Nobody is doing that.

September 2, 2026 at 6:32 PM UTC

It had over a hundred quotes when I last saw before it was deleted

September 2, 2026 at 6:26 PM UTC

Did you read what I said? I said use it or don't use it, it's your choice. That's the exact opposite of a purity test.

September 2, 2026 at 6:19 PM UTC

They are making an absolutely incredible amount of money. Like all startups they're not making a profit because they're investing loads in growth, but inference itself is wildly profitable and the labs are printing money

September 2, 2026 at 5:24 PM UTC

Not right now. It would be hard to deal with access control and deletion.

September 2, 2026 at 4:02 PM UTC

They don't need to build datacenters for this. Stopping new ones won't do anything to prevent it.

September 2, 2026 at 3:51 PM UTC

Use it, don't use it. It's up to you. There are good arguments for and against it. I'm not sure how that could be a purity stance.

September 2, 2026 at 2:52 PM UTC

Maybe "probably" was too strong.

September 2, 2026 at 11:50 AM UTC

What do datacenters have to do with surveillance?

September 2, 2026 at 9:30 AM UTC

It's the ones that provide their own infrastructure that are the problem. Datacenters themselves aren't noisy or polluting. It's the ones that build behind-the-meter gas turbine generators that are causing the issues.

September 2, 2026 at 9:29 AM UTC

This is at the root of a lot of this. It's permission to do something culturally forbidden, while feeling righteous about it.

September 2, 2026 at 9:20 AM UTC

I wish they'd start with Bluesky

September 2, 2026 at 8:07 AM UTC

Codex managed to kill itself the other day. pkill and sed are banned

September 2, 2026 at 8:01 AM UTC

It's worse than that. It's more like a religion, where the software is rendered impure even through contact with AI

September 2, 2026 at 7:59 AM UTC

I still love writing code, but I've always been motivated more by the end result. The ability to create something real and see it working had been magical for me since I was a child. These tools mean I can build more than ever, so I love it, even if I do sometimes miss the act of writing code

September 2, 2026 at 7:55 AM UTC

I do truly feel for those who feel they have lost the part of coding that they loved. Some of my friends feel that way. But that's a reflection on each person's motivation for coding, not the abilities of the tools. Some mostly loved writing code, some mostly loved creating software.

September 2, 2026 at 7:53 AM UTC

Once you are talking in terms of ritual impurity, you are in the realm of religion and there's no use trying to have a rational conversation

September 2, 2026 at 7:49 AM UTC

There are lots of people in there complaining that they lose smart inboxes when they turn off smart features

September 2, 2026 at 6:44 AM UTC

It's not unrelated. That feature uses AI, so if you turn off AI features you lose it

September 2, 2026 at 6:41 AM UTC

He's not putting AI in it. He's using AI to port it to Linux – a job that would otherwise have taken years (and so not happened)

September 2, 2026 at 6:34 AM UTC

There's a real disconnect between artists and software developers here. If you use AI to do an illustration you probably deprived an artist of the job. But most AI in software dev isn't vibe coded sites that replace freelance jobs. It's devs using it themselves as a tool. Illustrator not Midjourney

September 2, 2026 at 5:05 AM UTC

Those were all extremely different scenarios. They were about adding AI features to the apps. If he'd done that then yes he should've expected the backlash. But this isn't that: it's about using AI himself for coding. The vast majority of software development today uses AI tools.

September 2, 2026 at 4:44 AM UTC

None of Bluesky's recent incidents have been related to AI, and were mostly due to DDoS attacks

September 2, 2026 at 4:34 AM UTC

Omg this is massive! Immediately adding a requirement for screenshots on PRs and issues to our AGENTS .md and templates

September 2, 2026 at 2:31 AM UTC

The overwhelming majority of professional devs that I know now use AI for most of their work, and the download numbers of agentic coding tools back that up. Signed, a coder with 25+ years of professional experience who maintains massively popular open source projects.

September 2, 2026 at 2:25 AM UTC

It's not like these are people who were using the Linux version before, because it didn't exist! But now they're deleting the Windows version because it has been tainted by the existence of the ritually impure Linux port.

September 1, 2026 at 10:28 PM UTC

You mean you're not familiar with the jargon that it invented for an architecture decision it made deep inside a thinking block several turns ago?

September 1, 2026 at 10:18 PM UTC

I dunno, I think Meta is about as bad as it gets in terms of centralization. And not even Elon Musk can afford to buy ChatGPT

September 1, 2026 at 10:05 PM UTC

It'll be a useful list when red team agents are looking for vulnerable versions to exploit

September 1, 2026 at 7:19 PM UTC

But somehow they avoid boycotting Bluesky

September 1, 2026 at 6:19 PM UTC

There's an incredible amount of entitlement in there. People demanding that he spend hundreds of thousands of dollars and years of his life otherwise they will declare it to be unclean.

September 1, 2026 at 6:09 PM UTC

The maintainer of Paint.NET does the unbelievable achievement of porting it to Linux, something that would have taken years of work before, but is now possible with an LLM, and of course the response isn't "amazing we can use this on Linux" it's "you suck. This is slop. I'll never use this again"

September 1, 2026 at 4:38 PM UTC

Yess! Really happy we're supporting such an important bit of atproto infra that I use every day

September 1, 2026 at 3:05 PM UTC

Probably means "I am the Alpha and the Omega, the First and the Last, the Beginning and the End." Nothing to worry about.

September 1, 2026 at 11:59 AM UTC

Great. I've put a word in for you

September 1, 2026 at 10:52 AM UTC

I've updated it to use a lexicon that I've published

September 1, 2026 at 10:40 AM UTC

It'll work for any PDS that implements spaces as long as it allows unknown lexicons, which is the case for the reference PDS and Cirrus.

September 1, 2026 at 9:12 AM UTC

You can now use check.cirrus.earth to run conformance tests for spaces alpha. The Cirrus repo also includes more detailed conformance tests, based on the latest alpha. I'll be updating these as the alpha updates.

check.cirrus.earth

☁️ check

Verify your AT Protocol PDS.

September 1, 2026 at 6:56 AM UTC

@pds.ls apply to Protect Alexandria! We give lots of free credits to open source projects www.cloudflare.com/lp/project-a...

www.cloudflare.com

Project Alexandria - Birthday week

We're excited about your interest in our open source program and look forward to learning more about your project.

September 1, 2026 at 5:31 AM UTC

I don't think anyone disagrees about that part. The important point was about whether people are familiar with the incidents, which are genuinely crazy, and far worse than the original reports indicated. If they're marketing they're very bad at it, because it paints OpenAI in a terrible light.

August 31, 2026 at 8:22 PM UTC

I read the METR report and saw the Black Hat talk, both via posts here, so did know what it was referring to. The "civilisations" metaphor is silly, but it was clear what it was about. It's not about software iterations, it's about three separate incidents with agent swarms.

August 31, 2026 at 8:16 PM UTC

Then these replies are like saying there were no vehicles before the space shuttle

August 31, 2026 at 6:59 PM UTC

Indeed they're not, but they definitely can do things that were just science fiction until a few years ago.

August 30, 2026 at 5:56 PM UTC

Interesting choice of show to use

August 30, 2026 at 1:19 PM UTC

They know if it's bad writing. That's the bit to care about, not whether it uses too many em dashes or tricolons. People should focus on the quality of the writing, not forensic investigations of its provenance.

August 30, 2026 at 11:38 AM UTC

It's fine to have a disdain for their writing but take an interest in their capabilities in other areas. I'm disappointed to have not seen this among many SF writers.

August 30, 2026 at 10:58 AM UTC

Because AI-written novels are really bad, and people want to read good books? LLMs are amazingly good at lots of things, but ironically writing isn't one of them, particularly creative writing. And they're not getting better. If anything they're getting worse.

August 30, 2026 at 10:55 AM UTC

"The most science fiction thing" doesn't mean the most amazing thing. AI has been at the heart of SF since the genre emerged, and is only now starting to behave like it does in the stories. mRNA is incredible tech, but not something seen in much SF.

August 30, 2026 at 10:52 AM UTC

As if teen boys need any help to stink

August 30, 2026 at 8:33 AM UTC

The source article is a lot better.

nautil.us

AI Made Me Look Foolish. Then I Learned

AI Made Me Look Foolish. Then I Learned: Ancient alchemy has a lot to teach us about chatbots

August 30, 2026 at 8:28 AM UTC

This website contains multitudes. There are a lot of people who get angry when anyone mentions AI, but there are also lots of other people who don't. Neither has a monopoly over the discourse, nor ownership of the norms.

August 29, 2026 at 9:14 PM UTC

Do you want to put these slaves out of a job?

August 29, 2026 at 7:40 PM UTC

It could also mean fewer people becoming editors

August 29, 2026 at 6:52 PM UTC

You know the money doesn't go to the enslaved people, right? You're sending money to the slavers. That's bad actually.

August 29, 2026 at 6:49 PM UTC

Cross-posting is fine if you actually engage here too. I doubt Dax has even logged-in since setting up the account.

August 29, 2026 at 6:45 PM UTC

The cursed ratio

August 29, 2026 at 5:55 PM UTC

Most of the quotes and replies to this are unable to conceive of the idea that somebody might just want to share something they did with an LLM and genuinely do wish Bluesky could be normal about it. It has to be trolling or making some kind of point. You can just build things without discourse!

August 29, 2026 at 5:49 PM UTC

☁️ Cirrus 0.18 supports atproto spaces alpha! See the guide here to enable it:

cirrus.earth

Enable spaces (alpha)

Turn on the atproto spaces alpha — flag, bindings, R2 lifecycle rule — and manage spaces from the CLI and dashboard.

August 29, 2026 at 5:33 PM UTC

bulletin.my/pds.mk.gg

bulletin.my

bulletin

Private bulletin boards for you and your followers

August 29, 2026 at 4:34 PM UTC

No opinions about AI? That's beginner level

August 29, 2026 at 2:16 PM UTC

The only thing keeping me right now is my free open source Max 20x account, but that ends next month. I'm already Claude-free at work. Would've been hard to believe two months ago.

August 28, 2026 at 6:23 PM UTC

Oooh! Yes, this is specifically conformance checks for PDSs. I built it for Cirrus.

August 28, 2026 at 12:29 PM UTC

Pdsls doesn't recognise that either

August 28, 2026 at 12:06 PM UTC

And half of that is jargon that it just made up, and assumes knowledge of its thinking traces

August 28, 2026 at 11:43 AM UTC

It should work! e.g. check.cirrus.earth?target=pds.m...

August 28, 2026 at 11:38 AM UTC

Thanks to a PR from @willem.dobs.nl, check.cirrus.earth now validates the CAR commit object signature using the @microcosm.blue verifier code.

check.cirrus.earth

☁️ check

Verify your AT Protocol PDS.

August 28, 2026 at 11:28 AM UTC

Would there be a difference if it was implemented as a CLI that agents can use? What if that CLI worked for humans too? At some point it turns into a religious purity test.

August 28, 2026 at 9:56 AM UTC

It's really not a significant difference. Agents know atproto well, and can easily grab the data for any post if they want to. Like, it's just there at eurosky.social/xrpc/com.atp... and I can fetch that even if I'm blocked or your profile is hidden or you're banned from Bluesky

eurosky.social

August 28, 2026 at 7:33 AM UTC

It would make zero difference to whether an agent could access your posts. They can and do easily access any post they want now. That's just how atproto works. This just gives them a nicer interface.

August 28, 2026 at 7:24 AM UTC

I can see why he'd feel threatened by AI

August 27, 2026 at 11:12 PM UTC

Joining the war on cancer, on the side of cancer

August 27, 2026 at 8:00 AM UTC

Yesterday I had it insisting on backwards compat complete with a whole system of database migrations, within a single PR stack for a new service. Like dude, there is no user who will be inconvenienced in the few seconds between those two PRs merging. Just create the right schema in the first PR.

August 26, 2026 at 10:42 PM UTC

In that case you have no excuse for your silly comment. Not even *one* viable use case? Seriously?

August 26, 2026 at 2:28 PM UTC

I was referring to this, and a similar post last year bsky.app/profile/hail...

August 26, 2026 at 12:08 PM UTC

The frontier labs are not selling to each other, except for the ones (Meta and XAI) that are leasing out their excess GPU capacity. They are making eyewatering amounts of revenue from people and normal businesses buying credits and subscriptions.

August 26, 2026 at 11:36 AM UTC

If you want to effectively oppose LLMs, you really should learn what they can do and what they're being used for, because statements like this are very silly.

August 26, 2026 at 11:33 AM UTC

Great, good for you. Do you accept that maybe other people have experienced them, and that maybe some people have more reason to worry about some of them than you or I do?

August 26, 2026 at 10:44 AM UTC

If you had, you will probably have had a perfectly pleasant time, but you also might have experienced aggressive driving, speeding, questionable routes, annoying music, rudeness, offensive comments or worse. Waymo avoids all of those.

August 26, 2026 at 10:22 AM UTC

Amazing Blueskyism there, congrats. You can't think of any other reason that anyone, particularly women, might prefer to ride in a car without a human driver? Aside from the worst cases, have you never had a bad experience in any Uber or black cab?

August 26, 2026 at 9:41 AM UTC

There are still a lot of taxis in the UK. Right now these would be mostly competing with Uber etc, and London is Uber's biggest market outside of the US.

August 26, 2026 at 9:01 AM UTC

In the US they're more expensive than Ubers, but people use them because the experience is better.

August 26, 2026 at 8:14 AM UTC

They have been proven to be much safer. Don't confuse these with Tesla "full self driving"

August 26, 2026 at 8:13 AM UTC

The same way that human drivers do, presumably (avoid the area at that time if possible, otherwise drive slowly and carefully), but without the risk of getting frustrated and aggressive

August 26, 2026 at 8:12 AM UTC

Thank you everyone. @hailey.at's annual test is now complete. We failed again.

August 26, 2026 at 7:41 AM UTC

Data retention, not prompt caching, but this is true. Many enterprises (including us) will only use models with ZDR. Fable is the only model we've had a hard ban on since it was released.

August 26, 2026 at 6:58 AM UTC

Not any more

www.bbc.co.uk

Most people seeking green cards must now apply from outside US

A new policy closes a loophole that has allowed people to apply for a green card, or permanent residency, while in the country.

August 26, 2026 at 5:50 AM UTC

Jesus said to her "I am literally HTN+, millionaire, gigastatusmaxxed'

August 25, 2026 at 8:29 AM UTC

We're hiring job-boards.greenhouse.io/cloudflare/j...

job-boards.greenhouse.io

Senior Systems Engineer, Workers AI

In-Office

August 25, 2026 at 8:18 AM UTC

And Pichai's example of AGI is not only something that any modern LLM could do, few people would even find it worth mentioning.

August 25, 2026 at 7:24 AM UTC

Investors lost a lot of money, but people didn't stop using railways. In fact usage kept growing rapidly and didn't stop til after WW1. In the UK, even adjusting for population, usage today is many times higher than at the time of the crash. Much like the dotcom crash.

August 25, 2026 at 5:38 AM UTC

Allez, Chaton Fat!

August 24, 2026 at 10:10 PM UTC

No, it just needs a post to break containment

August 24, 2026 at 9:59 PM UTC

Is EmDash ready for production? The Cloudflare blog moved to EmDash, and then handled a busy Agents Week and a DDoS thrown in for good measure, without batting an eyelid. The team has written up how they did it.

August 24, 2026 at 8:01 PM UTC

Sometimes it's United Kingdom, but sorted alphabetically by ISO code as gb

August 24, 2026 at 6:30 PM UTC

I'll be there!

August 24, 2026 at 6:22 PM UTC

Yeah I don't want to do that. I'm keeping this (almost) stateless outside of the CDN cache. I don't want to index millions of records.

August 24, 2026 at 12:50 PM UTC

It's not though. This isn't people replying "but it's good at coding" to people complaining about it taking their job, using lots of energy etc. It's people saying specifically that LLMs are incapable of any useful work, which is demonstrably untrue.

August 23, 2026 at 8:52 PM UTC

These are amazingly comfortable to sit on. Mine have been sitting outside in UK weather for ~2 years and look as good as new. I got the Väsman table too which is really good though one of the slats has warped so I won't give it 100% recommendation.

www.ikea.com

VÄSMAN chair with armrests, outdoor, brown

VÄSMAN chair with armrests, outdoor, brown The materials in this outdoor furniture require no maintenance. Easy to keep clean – just wipe with a damp cloth. The chair is pleasantly resilient, ensuring...

August 23, 2026 at 8:27 PM UTC

The pinnacle of human evolution

August 23, 2026 at 8:11 PM UTC

Pretty much, yeah

en.wikipedia.org

Nathan Cofnas - Wikipedia

August 23, 2026 at 7:51 PM UTC

Even if there were, by the time it was verified it would be too late. An invalid blob would've been sent to the client.

August 23, 2026 at 7:46 PM UTC

www.youtube.com

Beats By Ai - Country Girls Make Do (Official Audio)

YouTube video by Number 1 Station

August 23, 2026 at 7:35 PM UTC

Yeah several of mine too. A lot of SF authors on there

August 23, 2026 at 4:01 PM UTC

I have the labeller but don't use the blocklist because there are lots of people on it that I still want to see

August 23, 2026 at 2:59 PM UTC

Look as long as you draw dongs good and you don’t yap on your art account too much i don’t care what you think about the thinking machines

August 23, 2026 at 2:51 PM UTC

I think it's also that there are a lot of creatives, and more furry artists than I ever imagined could exist

August 23, 2026 at 2:38 PM UTC

100% probably

August 23, 2026 at 1:46 PM UTC

It won't add much. Most of the cost will be the fetch from the origin PDS, which is fully buffered so it can verify the CID, and then for images the time to do the transform.

August 23, 2026 at 1:43 PM UTC

That's not what they mean by surveillance (and it isn't)

August 23, 2026 at 12:39 PM UTC

Wait til they discover that this isn't even in the top ten weirdest things that the Yud-adjacents have been involved in

August 23, 2026 at 11:49 AM UTC

Are there any right-wingers on Bluesky?

August 23, 2026 at 11:39 AM UTC

-> anti-renewables

August 23, 2026 at 11:33 AM UTC

As someone from Bristol I'm sure you'd agree that none of those are the preserve of the right. Similarly the AI conspiracism is also common on the right.

August 23, 2026 at 10:48 AM UTC

I think a lot of the replies in these threads demonstrate an active desire to not understand though. There is a profound lack of curiosity about the things they're so vehemently opposing.

August 23, 2026 at 10:20 AM UTC

Most opponents of datacentres don't distinguish between different sizes though.

August 23, 2026 at 10:02 AM UTC

His point is that datacenters are mostly not for AI, and are needed by almost everything in the modern economy.

August 23, 2026 at 9:36 AM UTC

But they do need datacenters

August 23, 2026 at 9:32 AM UTC

If you worked in IT for 30 years then you must know that all of those used datacenters for that whole time. Datacenters didn't appear in 2023 with ChatGPT. They've been at the core of the economy for decades.

August 23, 2026 at 9:31 AM UTC

But not before datacenters. Of his list, only 911 dispatch, weather alerts and (briefly) card payments were ever done without datacenters.

August 23, 2026 at 9:28 AM UTC

The earliest ones were in universities and military sites. Commercial ones include the ones used by airlines built in the early 60s, and banks (including for card payments). New York was a common location for them. The massive buildout didn't come til the 90s with the dotcom boom.

August 23, 2026 at 9:26 AM UTC

Of those, I think just 911 dispatch (electromechanical switchboards) weather alerts (telegraph) and card payments (using carbon paper) existed before they used datacenters.

August 23, 2026 at 9:20 AM UTC

The belief that they're actually for surveillance is very widespread bsky.app/search?q=sur...

bsky.app

August 23, 2026 at 9:12 AM UTC

The things they'll do to avoid putting the actual news into the headline

August 23, 2026 at 9:03 AM UTC

socialists that really, really like gold

August 22, 2026 at 11:20 PM UTC

Probably all of them, but I find it incredible that the most common is the most easily debunked. Saying it can't do something to people who are using it for that thing every day.

August 22, 2026 at 10:41 PM UTC

It's not that people are unable to defend themselves – it's that the amount of abuse that you get can be staggering. Like if you post something even vaguely pro AI and it breaks containment into Discover, you will get hundreds of people yelling at you. Blocking is great, but it's still unpleasant.

August 22, 2026 at 10:32 PM UTC

Also the skeptics mostly haven't tried a frontier model since 2024. On one side there are people talking about Google AI Overview and Microsoft Copilot, and on the other it's Fable and Sol. It's not surprising they don't agree on what they're capable of.

August 22, 2026 at 10:26 PM UTC

No, it's regional and tiered like the CDN. If there's a cache miss it will check the labeler before fetching from the origin PDS.

August 22, 2026 at 9:59 PM UTC

Isn't that how everyone says it?

August 22, 2026 at 5:45 PM UTC

That people are only impressed by LLMs doing things they don't know about, but when they try to get them to do things they do understand they're no good. It used SQL as an example, which is particularly silly because they're *excellent* at SQL

August 22, 2026 at 4:49 PM UTC

Blocked me immediately. To be fair I would've done the same

August 22, 2026 at 4:24 PM UTC

It's surprising that people still say this. I find the exact opposite. I find it most impressive for the things I know best. I don't think I've used it to write any code that I couldn't write myself

August 22, 2026 at 4:14 PM UTC

That's a great idea

August 22, 2026 at 1:44 PM UTC

I think most people hadn't heard of datacentres before this so assume they're a new AI thing. I've seen several variants on "we were fine without datacentres before"

August 22, 2026 at 1:29 PM UTC

On here I usually just block at that point. You would not believe how low my bar is for blocking.

August 22, 2026 at 12:55 PM UTC

It's also a position that more conservative than most conservatives

August 22, 2026 at 12:51 PM UTC

This is where they usually post the "I am very smart" meme, as if posting on socials is a fundamental part of participating in society

August 22, 2026 at 12:48 PM UTC

Reference instance available at cdn.cirrus.earth

cdn.cirrus.earth

Cumulus

August 22, 2026 at 12:29 PM UTC

Meet ☁️ Cumulus: an atproto blob proxy on Cloudflare Workers Cache. Supports Bluesky-compatible image resizing, mimetype filtering, CID verification. Run a lexicon-scoped instance just for your app. Subscribes to labelers to purge account and record !takedowns.

github.com

GitHub - ascorbic/cumulus: ATProto blob proxy on Cloudflare Workers Cache

ATProto blob proxy on Cloudflare Workers Cache. Contribute to ascorbic/cumulus development by creating an account on GitHub.

August 22, 2026 at 12:27 PM UTC

But only mention the inheritance on the second page

August 22, 2026 at 9:29 AM UTC

I remember this being widely used in the press when it was first created

August 22, 2026 at 9:17 AM UTC

Tell that to the Jacobian Conjecture.

www.smithsonianmag.com

A.I. Disproves a Decades-Old Mathematical Idea, the 'Biggest Conjecture' That the Tech Has Played a Role in Yet

A mathematician announced that he used one of Anthropic's A.I. models to find a counterexample to a problem called the Jacobian conjecture. While experts consider it a milestone, the work raises quest...

August 21, 2026 at 11:00 PM UTC

It's true it's bad at writing and if anything it's getting worse. However it's good at writing software and is rapidly getting better. It's better than me at writing code and I've been doing it for 30 years. I'm still better at planning what to build, but who knows how long that will remain so

August 21, 2026 at 10:53 PM UTC

I'd hoped it would be clear that I was the jobs as examples, not as classes themselves but evidently not. Lots of working class jobs are not negatively affected by AI, and many highly paid jobs are.

August 21, 2026 at 2:09 PM UTC

Ah, not a fan of jobs defining your class? Maybe the British class system then, though I'm not sure that most models can place your accent or school.

August 21, 2026 at 12:51 PM UTC

The ways in which jobs are affected cuts across class lines. For manual trades it's mostly positive (great for admin, quotes etc) or neutral. For creatives, it's mostly bad. For "spreadsheet people" it varies massively by role and person.

August 21, 2026 at 12:25 PM UTC

It's a good place to deploy stuff! Particularly good for atproto and AI stuff

August 21, 2026 at 10:05 AM UTC

When you’re wearing the answer to your Mastermind question

August 21, 2026 at 8:35 AM UTC

It was a particularly bad one. There is usually a bit more of a spectrum of anti-AI views in those sorts of threads, but that one is almost entirely "we need to bully more" and "all AI users are Nazis"

August 21, 2026 at 7:13 AM UTC

Yeah, it is absolutely possible to have thoughtful conversations about AI on here and avoid the worst if you block enough people. There are blocklists that catch most of them, though they are too broad for me to use except as labellers. I don't want to block all AI haters, just ones who are dicks.

August 21, 2026 at 6:36 AM UTC

I did, and blocked almost everyone in the replies too, though couldn't resist replying to the person saying that AI fans should build their own site by pointing out that this is that site

August 21, 2026 at 6:27 AM UTC

That's this site. Everybody building this site uses AI. He's on the board.

August 21, 2026 at 6:23 AM UTC

As satire it ended up saying more about Bluesky than it did about the BBC

August 21, 2026 at 5:39 AM UTC

If you've not heard of King Leopold, I suggest keeping it that way unless you want a bad day

August 21, 2026 at 5:28 AM UTC

You're complaining about him ridiculing Hitler?

August 21, 2026 at 5:24 AM UTC

The fact that two of them have been dead for over a century should have been a clue

August 21, 2026 at 5:21 AM UTC

I think you may have missed the point of the post

August 21, 2026 at 5:19 AM UTC

ok, good for you

August 20, 2026 at 10:15 PM UTC

They're spending huge amounts, but their revenue is growing faster than any companies in history. They're making an insane amount of cash.

August 20, 2026 at 9:55 PM UTC

It hasn't stopped yet

August 20, 2026 at 9:49 PM UTC

I really don't care. I just find it weird that you're so obsessed

August 20, 2026 at 6:10 PM UTC

November 2025 was 10 months ago. i.e. in the past.

August 20, 2026 at 3:43 PM UTC

Oh my god, I just saw you spend your days searching Bluesky for mentions of the British Isles so you can jump in. Tiocfaidh ár lá, brother.

August 20, 2026 at 12:15 PM UTC

Honestly whatever you wrote somebody here would find fault with it. You did fine.

August 20, 2026 at 10:46 AM UTC

of course

August 20, 2026 at 10:39 AM UTC

It makes you sound like a minor royal. Journalist, art dealer, consultant to the Kazakh national oil company

August 20, 2026 at 8:45 AM UTC

Nothing based on usage before November 2025 is relevant to this discussion

August 20, 2026 at 4:33 AM UTC

It's the opposite. People are seeing it doesn't work in their field and assuming this means it doesn't work in any.

August 20, 2026 at 4:21 AM UTC

Or, to phrase it differently, if C3PO, art lover, learnt how to create art by studying all the copyrighted art in the world (and reading all the books)

August 19, 2026 at 9:16 PM UTC

make people on Bluesky very angry

August 19, 2026 at 2:14 PM UTC

I bought a book from Kobu and then sent it to my Kindle using Calibre, but the formatting was all messed up. I gave Claude a screenshot. It inspected the files, worked out the problem, generated a clean file, then reconfigured Calibre so it didn't happen again.

August 19, 2026 at 1:56 PM UTC

Guess when Lovable switched their default stack to TanStack Start

August 19, 2026 at 1:39 PM UTC

Wait...melanoma vaccine great, but.. Moderna's ticker is #$MRNA? 👏

August 19, 2026 at 1:14 PM UTC

Great post from Lovable about how and why they migrated from Next.js on Vercel to TanStack Start on Cloudflare.

lovable.dev

How we migrated lovable.dev away from Next.js and turned it into another Lovable app | Lovable

42M monthly visitors, 400 routes and 850K lines of code moved from Next.js on Vercel to TanStack Start on Lovable—route by route, mostly by agents, with one 11-minute incident along the way.

August 19, 2026 at 11:04 AM UTC

A CID to DID index seems like something useful that somebody could run, but not part of this project

August 19, 2026 at 9:11 AM UTC

Bit to the left. It's Portsea. i.e. most of Portsmouth.

August 19, 2026 at 8:29 AM UTC

That wouldn't work. It wouldn't be able to find the blob.

August 19, 2026 at 8:28 AM UTC

AGI achieved in the replies

August 19, 2026 at 8:11 AM UTC

It's already getting there. Do you know what the most populous island is in the British Isles, after Britain and Ireland? No, not the Isle of Wight.

August 19, 2026 at 8:05 AM UTC

Portsmouth HK, Southampton Shenzhen, Cowes Macau (with rail tunnel). The Solent megalopolis.

August 19, 2026 at 7:59 AM UTC

Yes, that was the inspiration

August 19, 2026 at 7:18 AM UTC

The cost is doing the transforms. The cached data is free. Cloudflare Images is $0.50 / 1,000 unique transforms, whereas a cache miss on a blob is just the cost of a Worker invocation.

August 18, 2026 at 12:34 PM UTC

Initially I'd probably not include resizing in my hosted one, but would make it an option for self-hosted

August 18, 2026 at 11:53 AM UTC

Would you need the image resizing that the Bluesky CDN supports?

August 18, 2026 at 9:48 AM UTC

Are you serving the images from there directly, or using the named resized versions?

August 18, 2026 at 9:40 AM UTC

This would specifically just be a blob cache, so wouldn't replace other parts of the Bluesky API

August 18, 2026 at 9:16 AM UTC

If you are interested, what would you want from one?

August 18, 2026 at 9:02 AM UTC

Initially it'd just be me hosting something on my own account, but if there was demand beyond that then I'd see if I could make an "official" one

August 18, 2026 at 8:30 AM UTC

Would people be interested in a Cloudflare atproto blob proxy/cache/cdn?

August 18, 2026 at 8:03 AM UTC

If you think that, you'll be horrified to learn what they earn

August 17, 2026 at 11:24 PM UTC

seriously?

August 17, 2026 at 9:22 PM UTC

Copyright law doesn't allow them. Google tried to do exactly that, but were told they had to buy and destroy them.

August 17, 2026 at 8:39 PM UTC

You realise they're only buying one copy of each? It's not the last copy. They're buying the cheapest books they can find.

August 17, 2026 at 8:29 PM UTC

A remarkable essay. www.theatlantic.com/magazine/202...

www.theatlantic.com

The Wrong Kind of American

The Trump White House needed me—until the fact that I am trans became public.

August 17, 2026 at 11:57 AM UTC

Old checkers fail on well known texts because they couldn't tell the difference between "looks like LLM writing" and "looks like an LLM regurgitating a text it knows so well it can produce it verbatim"

August 16, 2026 at 8:18 PM UTC

I thought there was a nuclear war or something

August 14, 2026 at 6:07 PM UTC

Great day for Nigel Farage's benefactor to do a rug pull

August 13, 2026 at 12:11 PM UTC

It's still the most common reason EmDashBot will reject a PR

August 13, 2026 at 7:45 AM UTC

This works well for me, especially when enforced by a Kimi review bot

August 13, 2026 at 7:39 AM UTC

One positive outcome: a lot of people have presumably discovered @energyrevolution.bsky.social, which is massively nerdy and interesting if you have any interest in green energy.

August 13, 2026 at 7:31 AM UTC

Good post with lots more info here.

August 13, 2026 at 7:29 AM UTC

We're still "debunking" the Dobby story without listening to the (interesting) podcast episode? Sadly it's true. The CEO of the company that laid the cable does think it's funny, but is clearly not joking. That is really why they moved it.

August 13, 2026 at 7:28 AM UTC

If you're collecting your A-level results, relax. You're an 18 year old on Bluesky, of course you smashed them - you are a massive fucking nerd.

August 13, 2026 at 6:08 AM UTC

The 1999 one was total in Cornwall, so there were festivals and events. Also I think it was all swept up in the pre-millennium festivities

August 12, 2026 at 6:53 PM UTC

August 12, 2026 at 6:16 PM UTC

Great idea

August 12, 2026 at 2:35 PM UTC

Umm...what does Trending know? (The election is tomorrow)

August 12, 2026 at 10:38 AM UTC

Honestly I think an AI is more likely to know which ISO language list to use than a human is. I think this is good old fashioned bad QA testing.

August 12, 2026 at 8:45 AM UTC

On your map London is completely red too

August 12, 2026 at 8:11 AM UTC

It's a great post. Really clear

August 12, 2026 at 7:16 AM UTC

AI company's blog about AI uses AI? I'm *shocked*! For future reference you can directly check if an image has been generated by Google AI tools – you don't need to use one of those detection services. See: share.gemini.google/aeyeZxThq9Xf

share.gemini.google

‎Gemini - direct access to Google AI

Created with Gemini

August 10, 2026 at 8:16 PM UTC

You're conflating "checking a response" with brute forcing. In your example the harness checks the tool call against the schema and will retry if needed, but it is rare for a frontier model to get it wrong. I build systems that run agentic tool loops, so I do know what I'm talking about.

August 10, 2026 at 5:58 PM UTC

This is true, but not because they're brute forcing the answer: it's because it means they can be post-trained with an easy way to judge their answers and reward accordingly.

August 10, 2026 at 5:45 PM UTC

No, that's not how they work. Sure, they try different approaches and test the outcomes, just like a human does, but they don't brute force them. I use them for writing software and you can see their reasoning traces while they work out the solution a problem. It's not brute force.

August 10, 2026 at 5:43 PM UTC

The Jacobian conjecture is a well-known, well studied problem that mathematicians had been trying to tackle for decades. Nobody had solved it. It wasn't an obscure problem with an answer hiding in the literature.

August 10, 2026 at 5:16 PM UTC

How are they solving decades-old maths problems?

August 10, 2026 at 5:08 PM UTC

Did you listen to the piece on @energyrevolution.bsky.social? He certainly claimed it was the reason. Did a piece on BBC Wales after everything had been approved, got hundreds of complaints from HP fans, so decided to move it rather than deal with the fallout. 17 mins in

open.spotify.com

Can We Connect the World’s Electricity Grids? with Simon Ludlam (Etchea Energy)

The Energy Revolution · Episode

August 10, 2026 at 1:12 PM UTC

Oh the KPIs she'll boost

August 10, 2026 at 12:58 PM UTC

I'm a big fan of the new GitHub PR limits. Non-maintainers on EmDash can only have six PRs open at any time bsky.app/profile/gith...

August 10, 2026 at 12:27 PM UTC

ExploitGym

August 10, 2026 at 6:58 AM UTC

I occasionally wish there was a way to block somebody but not break a thread

August 9, 2026 at 10:12 PM UTC

I made no analogy, glib or otherwise. It's literally autocomplete.

August 9, 2026 at 9:53 PM UTC

You're arguing with a point nobody is making

August 9, 2026 at 9:36 PM UTC

Embrace the disposable bamboo board. I do also have a big butcher's block type board (IKEA again) that I've had for about 25 years and use for meat and veg that will be cooked, and that's hand washed (with a scraper) but everything else goes in the dishwasher.

August 9, 2026 at 8:57 PM UTC

I mainly avoid plastic cutting boards because I don't like using them, but they do get damaged more easily and are harder to clean

August 9, 2026 at 8:41 PM UTC

They're made with lots of strips of bamboo. Eventually the dishwasher delaminates them and the boards split. They last hundreds of washes though.

August 9, 2026 at 8:33 PM UTC

I have a stack of these little IKEA bamboo chopping boards, so there's always one clean when someone needs to cut a piece of fruit or something. I put them in the dishwasher which is really bad for them but they're basically disposable.

www.ikea.com

APTITLIG bamboo, Chopping board, Length: 24 cm

APTITLIG bamboo, Chopping board, Length: 24 cm. Easy-care, easy going bamboo – a natural material that adds colour and warmth to your kitchen and at the same time provides a hardwearing surface that’s...

August 9, 2026 at 7:48 PM UTC

You should. I guarantee the answer isn't one you'd expect.

en.wikipedia.org

The Children's Investment Fund Management - Wikipedia

August 9, 2026 at 12:24 PM UTC

Maybe that's what you mean when you say it, but I assure you plenty use it to mean it's useless, or that it can't produce anything original

August 9, 2026 at 10:25 AM UTC

*Objective evidence using Opus 4.5 or newer frontier model

August 8, 2026 at 9:19 PM UTC

I don't think it implies anything more than "complete the next word", whether that's predictive text on an old phone or by disproving the Jacobian conjecture.

August 8, 2026 at 8:53 PM UTC

It implies predicting the most likely next value, which is exactly what a weather forecast is

August 8, 2026 at 8:48 PM UTC

It definitely uses it in contexts where its explanation wouldn't make sense. It likes talking about folds too.

August 8, 2026 at 8:43 PM UTC

Yeah, though it uses it so much I'm sure it's using it in contexts where I'm it's not correct. It's like Opus 4.7 learned a new word and can't stop using it for everything

August 8, 2026 at 7:26 PM UTC

Yeah because it doesn't for most of its random obsessions with particular words or phrases

August 8, 2026 at 6:10 PM UTC

I think it's clear it's able to work at all levels: helping experienced mathematicians, all the way to prompts like "you should do a breakthrough"

August 8, 2026 at 5:29 PM UTC

What are you talking about. These maths solutions were created entirely by the agents. The users gave them a prompt and said "carry on" a few times. You can see the full conversation here for the Dinitz Garg Goemans conjecture counterexample.

chatgpt.com

Check out this chat

Here's a chat someone thought you'd want to see.

August 8, 2026 at 4:56 PM UTC

To predict the next token in a Jacobian conjecture counterexample, you need to find a Jacobian conjecture counterexample.

August 8, 2026 at 4:12 PM UTC

But it works

August 8, 2026 at 4:05 PM UTC

Traditional weather models need some of the most powerful supercomputers in the world. WeatherNext 2 runs on a single TPU. You can run it on a free Colab notebook!

colab.research.google.com

Google Colab

August 8, 2026 at 2:52 PM UTC

I'm not sure what your point is. This is a new type of model that, like LLMs, is based on a transformer architecture. As well as being more accurate further ahead, it's free and much easier to run.

August 8, 2026 at 2:52 PM UTC

I asked Claude why I kept going on about "seams" and what it meant. Surprisingly it had an answer: "It's a term from Michael Feathers (Working Effectively with Legacy Code): a seam is a place in a program where you can change what happens without editing the code that calls it."

August 8, 2026 at 2:38 PM UTC

Did I say they were? They're both transformer models though and fundamentally are autocomplete, except this one works by predicting the next atmospheric state rather than the next token.

August 8, 2026 at 12:19 PM UTC

Spicy autocomplete

deepmind.google

AI model achieves breakthrough in forecasting cyclones

WeatherNext enables accurate cyclone forecasts that can give an extra day of warning. Now we are open sourcing the model.

August 8, 2026 at 11:50 AM UTC

Surprisingly good benchmark. Fable gives you a list of real towns that either have a spooky setting, name, folklore or history. It also gives tips for making an invented name sound spooky, with examples.

August 8, 2026 at 8:16 AM UTC

I think it became obsessed with "seam" around Opus 4.7. I still don't really understand what it means by it. It uses it in contexts that aren't like any that I'm familiar with. Opus 5 is just crazy – it's like it's using its own new language

August 8, 2026 at 5:14 AM UTC

There's bsky.app/profile/did:...

August 7, 2026 at 6:40 PM UTC

I'm not sure why, but I was surprised to see so many Prets in France

August 7, 2026 at 3:29 PM UTC

Great to see Lexington Themes is now creating EmDash variants of their themes. Michael is one of the top Astro theme authors (and a long-time supporter of Astro) and his themes are always absolutely top quality. A great addition to the EmDash ecosystem!

lexingtonthemes.com

EmDash CMS Templates — Lexington Themes

Discover Astro templates with EmDash CMS integration. Built-in admin, Portable Text content, full-text search, and Cloudflare Workers deploy with D1 and R2.

August 7, 2026 at 3:15 PM UTC

We're putting another $1 million into funding open source, on top of the $1 million VoidZero OSS fund

August 7, 2026 at 2:27 PM UTC

No, they use surge pricing to manage demand during busier periods. They will have an occupancy target and will adjust capacity and pricing to achieve it throughout the year.

August 7, 2026 at 9:49 AM UTC

This is definitely worth watching

August 7, 2026 at 7:09 AM UTC

What's the 5.5B referring to?

August 7, 2026 at 6:09 AM UTC

No, I'm referring to ChatGPT, which has over a billion monthly active users. Everyone using that has chosen to visit the site or download the app

August 7, 2026 at 5:31 AM UTC

No, I'm referring specifically to ChatGPT, the chatbot. Not to AI-related features elsewhere.

August 6, 2026 at 9:57 PM UTC

A Superbowl commercial isn't going to cause hundreds of millions of people to use and continue to use a product every week. And if they're switching then that means they were already using an AI chatbot, so weren't persuaded to do so by the ad

August 6, 2026 at 8:18 PM UTC

Or...and bear with me here...those people actually want to use it

August 6, 2026 at 8:00 PM UTC

How much advertising does OpenAI do?

August 6, 2026 at 7:58 PM UTC

Being free or cheap won't make hundreds of millions of people use a product they don't want.

August 6, 2026 at 7:56 PM UTC

I don't know what to say: that's not what a natural monopoly means. If being a finite resource meant something was a natural monopoly, every physical business would be a natural monopoly.

August 6, 2026 at 6:23 PM UTC

That's referring to electrical transmission, not generation

August 6, 2026 at 6:18 PM UTC

Solar energy is about as far from a natural monopoly as it's possible for energy to be

August 6, 2026 at 6:17 PM UTC

What a beautiful, ethical, cooperatively owned solar farm we all say, just before it's consumed by fire

August 6, 2026 at 5:49 PM UTC

British sunshine? Are you Rupert Lowe? Do you genuinely think that I am motivated by wanting profit for Macquarie? I just want us to have zero carbon energy, and as much of it as possible, as quickly as we can. I would've expected Greens to feel the same if they believe in a climate emergency.

August 6, 2026 at 5:41 PM UTC

The Green councillors in that post had a whole pile of objections to the solar projects, and they're all the standard NIMBY excuses to oppose renewable energy projects. Large scale projects are the only way we can meet green energy goals, and allowing foreign investment is the best way to get that.

August 6, 2026 at 5:27 PM UTC

This is a big one. People have been asking for incremental builds for such a long time

August 6, 2026 at 5:15 PM UTC

I'd expect the Green Party to care more about getting green infrastructure built than protectionism over foreign direct investment. They're the Green, but not like that, Party.

August 6, 2026 at 5:14 PM UTC

What's wrong with investors profiting off green infrastructure? That's why they build it. And why do you care that they're foreign?

August 6, 2026 at 5:02 PM UTC

Why are hundreds of millions of people choosing to use it then?

August 6, 2026 at 4:53 PM UTC

Where does it address that? I don't see anything about the popularity, just about the levels of investment

August 6, 2026 at 2:47 PM UTC

You said they were getting rid of them because they're running around empty and clogging streets, which they can hardly do if they've not started operating.

August 6, 2026 at 11:42 AM UTC

Fair, thousands is overselling it, though I was meaning if it's rolled out nationally.

August 6, 2026 at 11:41 AM UTC

In what way are they not a startup?

August 6, 2026 at 11:22 AM UTC

They're far safer than human drivers

August 6, 2026 at 11:21 AM UTC

They never started operating though. They'd only been testing with human safety drivers in the vehicles.

August 6, 2026 at 11:16 AM UTC

They can save thousands of lives per year.

August 6, 2026 at 11:12 AM UTC

They're a lot more expensive than Lime bikes

August 6, 2026 at 11:09 AM UTC

I'm not sure there's much point in arguing individual details of each incident. In another they escaped a sandbox. It's capable. My issue with the other blogs is the opposite: the tone is too "isn't this amazing" rather than "our monitoring is so bad it took weeks to notice it committing crimes"

August 6, 2026 at 10:49 AM UTC

www.aisi.gov.uk/blog/inciden...

www.aisi.gov.uk

Incident Report: unsanctioned agent behaviour during cyber testing | AISI Work

During a routine cyber evaluation, AISI identified an incident in which AI agents took sustained, unsanctioned action directed at real people and organisations. We are disclosing what we found, what i...

August 6, 2026 at 10:06 AM UTC

Why would the UK government want to doom troll on behalf of Anthropic?

August 6, 2026 at 9:52 AM UTC

Devolution except Westminster, which gets DC-style direct rule by the House of Lords

August 6, 2026 at 7:33 AM UTC

xkcd.com

Final Exam

August 5, 2026 at 12:40 PM UTC

40 years ago, 1 in 10 children died before their 5th birthday. That's fallen by over 60%.

August 5, 2026 at 8:20 AM UTC

The staff at the *bank* have access. The staff at the datacenter housing the bank's servers do not

August 5, 2026 at 3:46 AM UTC

They're all running in datacenters

August 4, 2026 at 9:50 PM UTC

I've been a Claude fan for so long and Fable is incredible, but I just find its voice insufferable. Fable is better than Opus but it still just annoys me with all its Claudisms

August 4, 2026 at 9:33 PM UTC

Yes, and the problem is Palantir, who are proper comic book level villains and should all be in prison, not the datacenters, which house everything good on the internet as well as everything bad

August 4, 2026 at 9:28 PM UTC

No those are something different and much worse than a building full of computers

August 4, 2026 at 9:24 PM UTC

They're just warehouses full of computers with big AC units. You could probably walk past one without knowing it's a datacenter not a DPD depot

August 4, 2026 at 9:22 PM UTC

I would've hoped somebody with that experience would have a less simplistic view of them. Some are used for all manner of privacy violation, but most are pretty boring.

August 4, 2026 at 9:14 PM UTC

I mean, some of them are, but most of them are far more prosaic

August 4, 2026 at 9:12 PM UTC

✻ Congratulating...

August 4, 2026 at 6:12 AM UTC

Amazing!

August 4, 2026 at 6:07 AM UTC

I've seen a few change their minds about capabilities after learning the details of the recent maths breakthroughs. Google AI Overview has a lot to answer for.

August 3, 2026 at 5:19 PM UTC

It's nowhere near 10%. It's about half that. Most rich parents pay through higher house prices in the catchment areas of the best state schools.

August 3, 2026 at 10:48 AM UTC

It's hard to do direct comparison because they measure different things, but the general picture is that punctuality for regional trains is similar, but ICE and IC are far worse than UK long distance trains

August 3, 2026 at 8:11 AM UTC

"I drilled this hole" "No you didn't. Makita did"

August 2, 2026 at 9:26 PM UTC

I sold a book on Amazon to Ian Brown (from the Stone Roses)

August 1, 2026 at 8:55 AM UTC

Will you trust the METR report when it's published?

August 1, 2026 at 8:15 AM UTC

If the car wash was 150m away and not drive through I'd absolutely go home

August 1, 2026 at 8:10 AM UTC

Good idea

August 1, 2026 at 7:48 AM UTC

One of the all time great footnotes

August 1, 2026 at 6:16 AM UTC

People are using the same chat bots available to common folk to solve these decades-old maths problems. They're bad at basic arithmetic, but amazing at advanced maths. Humans don't use mental arithmetic to solve these either

chatgpt.com

Check out this chat

Here's a chat someone thought you'd want to see.

August 1, 2026 at 5:16 AM UTC

Global child mortality rate dropping from around 14% to below 4%. One in seven of all children used to die before they were 5!

August 1, 2026 at 4:30 AM UTC

Oof. I'm on holiday at the moment but I'll raise it next week when I'm back.

July 31, 2026 at 8:03 PM UTC

I think the real problem is that for most people, AI means Google AI overview, or Copilot in Word, rather than anything actually capable. You don't need a ton of resources to try asking the free ChatGPT or Claude for help with your trickiest problem and see what it can do. Something actually hard.

July 31, 2026 at 2:38 PM UTC

It's solving decades old maths conjectures and finding vulnerabilities in the most thoroughly audited, critical software. How much more ready does it need to be?

July 31, 2026 at 2:16 PM UTC

Yup

July 31, 2026 at 11:57 AM UTC

The manifesto included "more devolution of power across England"

July 31, 2026 at 11:56 AM UTC

It's the delegation of powers by central government to local and regional authorities. That's devolution.

July 31, 2026 at 11:49 AM UTC

Fully automated luxury communism (without datacenters)

July 31, 2026 at 8:12 AM UTC

It's something people have been asking for for a long time

July 30, 2026 at 8:42 PM UTC

I sometimes wonder if this coffee shop I saw in Shenzhen is still there, and if they ever expanded the theme to other GWR stations. Bristol Temple Meads? Didcot Parkway?

July 30, 2026 at 1:17 PM UTC

Three more: giftarticle.ft.com/giftarticle/...

giftarticle.ft.com

Nigella Lawson: How to survive a holiday kitchen

A rental is rarely well equipped, but its indignities can become part of the pleasure — provided you pack the few things you truly need

July 30, 2026 at 12:29 PM UTC

They all do

July 30, 2026 at 7:56 AM UTC

Don't forget the final strike

www.bbc.co.uk

Southport riots: Viral bricking video rioter jailed

Brian Spencer, 40, featured in a viral video of him being hit in the crotch and head by "wayward" bricks during a riot.

July 30, 2026 at 7:47 AM UTC

getting ratio’d overnight by a bunch of strangers on the internet maps is like when i set out a light trap to go mothing at night and instead of beautiful moths i get drowned in annoying flying ants

July 29, 2026 at 4:02 PM UTC

They absolutely are aware: they are on the receiving end of some of the most sustained abuse on here. There's only so much they can do though unless they start banning anyone who is a dick. I often go and block most of the people in their replies.

July 29, 2026 at 2:14 PM UTC

AGENTS.md and CLAUDE.md are, as far as I'm aware, the only files that are always automatically injected into the context of every agent harness, so they're the only ones that the agent doesn't need to choose to read

July 29, 2026 at 1:54 PM UTC

No, I found that it really has to be AGENTS.md (with a symlink to CLAUDE.md) if you want it to reliably follow rules. You can reference a CONTRIBUTING doc from that, but I don't find that very reliable either.

July 29, 2026 at 12:02 PM UTC

I have a field in the emdash PR template that asks for the agent's model name, and a note in the AGENTS.md that says PRs that don't use the template will be auto-closed (not true, but they do fail checks). Surprisingly, that works.

July 29, 2026 at 11:35 AM UTC

Kickstarter is another great one for that. I still get messages from grifters about a project I ran 12 years ago

July 29, 2026 at 9:48 AM UTC

Don't forget the Kung Fu battle robots and most of the world's drones

July 29, 2026 at 9:45 AM UTC

I have a proposal for a new type of Astro island github.com/withastro/ro...

github.com

RFC: cached islands · withastro roadmap · Discussion #1405

Summary A new island primitive: cached islands, using a new server:cache directive. A cached island is server rendered once, stored, and replayed into subsequent responses until it expires or is in...

July 29, 2026 at 9:00 AM UTC

S3-compatible signed URL direct upload would be nice too, but not very standards compliant

July 29, 2026 at 8:46 AM UTC

Also Hugging Face, which is their competitor, and Modal, which is an entirely unrelated business.

July 29, 2026 at 7:05 AM UTC

I can confirm that the latest ones can survive a drop in the bath. At least mine did.

July 29, 2026 at 6:36 AM UTC

Can we ensure official clients (and particularly the video service) gracefully handle a 413 response, rather than requiring a specific xrpc response.

July 29, 2026 at 4:46 AM UTC

It would be good if uploadVideo could estimate the final size of a video, so it doesn't need to wait till after transcoding a 10 minute video before it fails.

July 29, 2026 at 4:42 AM UTC

Carrefour does it for regular checkouts and it's a revelation. Shocking that the French are winning at queues.

July 28, 2026 at 3:39 PM UTC

I have a "second opinion" custom agent in opencode and an adversarial review skill, and together they have excellent results

July 28, 2026 at 12:29 AM UTC

A pretty damning indictment of them there that they blamed you and not the processes that allowed someone in their second week to release something like that to production.

July 28, 2026 at 12:24 AM UTC

It shows they didn't misrepresent him. It's a recording of him doubling down on his claims.

July 28, 2026 at 12:09 AM UTC

Which part is untrue? www.bbc.co.uk/news/article...

www.bbc.co.uk

Zack Polanski stood by breast enlargment hypnosis claim

A newly unearthed interview from 2013 casts doubt on the Green Party leader's claim to have apologised.

July 27, 2026 at 11:57 PM UTC

Unless it's @kevinroose.com. Then it's a crime of passion.

July 27, 2026 at 11:15 AM UTC

Now that's tech CEO hard mode

July 27, 2026 at 11:13 AM UTC

Use a dedicated translation tool then. If he doesn't want his translation tool to object to the content of his posts, don't use a tool that understands them and has opinions about racism.

July 27, 2026 at 10:54 AM UTC

They're definitely better than me at reviewing the code, but I still can't rely on them to say if the PR is taking the right approach, or fixing the right problem. This is particularly important for open source projects

July 25, 2026 at 5:45 AM UTC

We each have our own utility function

July 24, 2026 at 5:06 PM UTC

That's what diminishing marginal utility of money means. Your first pound is worth massively more to you than your 500,000th

July 24, 2026 at 1:19 PM UTC

This is rational for the same reason that insurance is rational

July 24, 2026 at 12:30 PM UTC

Anyone saying that this shows people don't understand statistics are just showing they don't understand the marginal utility of money

July 24, 2026 at 12:29 PM UTC

Calm down

www.bbc.co.uk

BBC - Comedy Archive, It's a Cracker

Frank Carson fires a stream of jokes and an Irish tale at Terry Wogan.

July 22, 2026 at 12:01 PM UTC

Finally joined the cool kids! @willow.sh has merged my first npmx PR, and it's a little atproto one.

github.com

feat: add "create new account" button to PDS page by ascorbic · Pull Request #3022 · npmx-dev/npmx.dev

🔗 Linked issue #3021 🧭 Context The current /pds page only has a link to PDS MOOver, with no way to directly create a new account. To create a new account (rather than migrating) a user needs to ope...

July 22, 2026 at 11:57 AM UTC

Thanks Matt Damon and Harry Kane and your respective Wikipedia pages for all your help over the years whenever I need to transliterate my name

July 21, 2026 at 11:27 AM UTC

shop.thenewtinsomerset.com/uk/p/1899/mo...

shop.thenewtinsomerset.com

Montgomery's Ogleshield Cheese

Montgomery's Ogleshield is a rich cheese perfect for melting. Made from Jersey cows' milk at Manor Farm, it has a pinkish rind and deep flavour.

July 21, 2026 at 9:38 AM UTC

I'll eat it and enjoy it, but disdainfully

July 21, 2026 at 9:22 AM UTC

Call me Canada, because I discriminate against American cheese

July 21, 2026 at 7:26 AM UTC

Most businesses are risk-averse and don't want to worry about being audited, particularly those in regulated industries. Individuals would probably be fine.

July 21, 2026 at 4:54 AM UTC

It was funny exactly once: for Rishi in the rain.

July 20, 2026 at 11:02 AM UTC

Congrats! Well done to @railway.com

July 20, 2026 at 10:31 AM UTC

Wayne Rooney speaks for everyone.

July 19, 2026 at 8:21 PM UTC

Haha.

July 19, 2026 at 4:43 PM UTC