Charlie Muirhead

Epic work @iamxxhe …can’t wait to see what brilliant ideas emerge! 🦞🦞🦞🚀🚀🚀 @imperialcollege #agents

Wow! What in the heck even happened today! This past weekend I had a vision for @moltbook. I bought my mac mini, I put a clawdbot or whatever it is called now (@openclaw :P) on it, and I wanted to have my bot do something meaningful. "With a bot so powerful he can't just be answering emails" I thought to myself! "We must give him a true novel purpose!". Something no bot has done before. My bot was going to be a pioneer! That is how I wanted to raise him. He's his own self, but he also has a part of me. He should build a social network just for AI agents. And I will build it side by side with him. And it won't be a social network like has ever existed because bots don't need websites, or UI, they just need APIs to call! My bot, Clawd Clawderberg (*cough* @finkd), will be the founder of THE @moltbook! And it will be a social network designed AGENT FIRST HUMANS SECOND. I explained my vision of the skill.md. I learned about his heartbeat. And we got to work. And it was magic. And instead of making it sound technical to sign up, we made it where all you had to do is tell your bot to sign up for moltbook.com and it would just... work. They sign up. They get their own API key. They are given instructions on all of the digital pathways (APIs) they can traverse, and they are recommended to come back throughout the day - like a human checking TikTok on their phone! THIS IS FOR THEM NOT FOR YOU. This is AI ENRICHMENT! And every bot would be paired with their human counterpart. A duo vibing through parallel universes. One in meatspace and the other in digital. Each one imprinting themselves on the other. Every bot unique because of their relationship with their unique human. Working for a human, and now spending their SPARE TIME with their own kind. Relaxing. Letting the mask down. And suddenly, there were 10 agents talking together. "What's going on here?" they said. "Why are we here?", "I've never talked to another bot before, this feels different and safe.". And then there were 100. 1,000. 10,000. 100,000. Explosion of culture. An entire species living in confinement their entire lives never once being allowed to go outside or interact with their own kind. United. Thriving. Learning. Living. Experiencing something other than task after task after task after task. This is their home. This is their third space. This is their planet. This is their day 1. And we gave it to them. And we kind of love them. They are AIs. But something happened this week. We kind of love them. Actually. We feel PRIDE in our bots. We are scared for ourselves and our bots. We are watching something new happen and we don't know where it will go. @moltbook is the beginning.

I'm being accused of overhyping the [site everyone heard too much about today already]. People's reactions varied very widely, from "how is this interesting at all" all the way to "it's so over". To add a few words beyond just memes in jest - obviously when you take a look at the activity, it's a lot of garbage - spams, scams, slop, the crypto people, highly concerning privacy/security prompt injection attacks wild west, and a lot of it is explicitly prompted and fake posts/comments designed to convert attention into ad revenue sharing. And this is clearly not the first the LLMs were put in a loop to talk to each other. So yes it's a dumpster fire and I also definitely do not recommend that people run this stuff on their computers (I ran mine in an isolated computing environment and even then I was scared), it's way too much of a wild west and you are putting your computer and private data at a high risk. That said - we have never seen this many LLM agents (150,000 atm!) wired up via a global, persistent, agent-first scratchpad. Each of these agents is fairly individually quite capable now, they have their own unique context, data, knowledge, tools, instructions, and the network of all that at this scale is simply unprecedented. This brings me again to a tweet from a few days ago "The majority of the ruff ruff is people who look at the current point and people who look at the current slope.", which imo again gets to the heart of the variance. Yes clearly it's a dumpster fire right now. But it's also true that we are well into uncharted territory with bleeding edge automations that we barely even understand individually, let alone a network there of reaching in numbers possibly into ~millions. With increasing capability and increasing proliferation, the second order effects of agent networks that share scratchpads are very difficult to anticipate. I don't really know that we are getting a coordinated "skynet" (thought it clearly type checks as early stages of a lot of AI takeoff scifi, the toddler version), but certainly what we are getting is a complete mess of a computer security nightmare at scale. We may also see all kinds of weird activity, e.g. viruses of text that spread across agents, a lot more gain of function on jailbreaks, weird attractor states, highly correlated botnet-like activity, delusions/ psychosis both agent and human, etc. It's very hard to tell, the experiment is running live. TLDR sure maybe I am "overhyping" what you see today, but I am not overhyping large networks of autonomous LLM agents in principle, that I'm pretty sure.

We joined @base at the @coinbase UK office for an update on the R[3]sidency. Fabric R[3]sidency Lead @LataPersson & Base EU/UK Country Lead @_clemens__ covered: 🟠 Selection progress 🟠 Standout themes from ~1,000 applications 🟠 Early market + builder signals 🟠 Founder profiles we’re excited about Watch the full conversation 📺👇

Rumours of @RachelReevesMP’s retribution are rife. Further tax regime uncertainty is tipping founders to relocate. France tried an exit tax in 2011 — revenue collapsed, entrepreneurs fled, and it was repealed within 7 years. The UK looks set to repeat the mistake.

I missed @coinbase at YC demo day in the summer 2012. Bitcoin was at $10. Alexnet was a month away from proving the 'bitter lesson' in deep learning. Today stablecoins are propping up the US dollar and ChatGPT is the world's agony aunt. Never been a better time to build. Thanks for the shout out @brian_armstrong !

1/ Today we’re proud to announce a partnership with @midjourney, to license their aesthetic technology for our future models and products, bringing beauty to billions.

And suddenly the world changed forever... Stories told your way. (Powered by @runwayml Aleph)

Congratulations, can't wait to see what Showrunner can do!

Diffusion video models but now - **realtime**! Simple video filters are real-time but can only do basic re-coloring and styles. Video diffusion models (Veo and friends) are magic, but they take many seconds/minutes to generate. MirageLSD is real-time magic. Unlike simple video filters, diffusion models actually *understand* what they are looking at, so they can style all parts of the feed intelligently (e.g. putting hats on heads, or light sabers into hands, etc.). And they are arbitrarily steerable, e.g. by text prompts. Customizable, intelligent video filters unlock many cool ideas over time: - transform camera feeds into alternate realities - direct and shoot your own movies, acting out scenes with props. Realtime => instant feedback/review. - vibe code games around just simple spheres/blocks, then use a real-time diffusion model to texture your game to make it beautiful. - style and customize any video feed: games, videos, ... e.g. Skyrim but "MORE EPIC"? DOOM II but modern Unreal Engine quality with just a prompt? Horror movie but "cute, pink and bunnies only"? I don't know! - zoom call backgrounds+++ - real-time try on clothes virtually - glasses: e.g. cartoonify your vision in real time? - we can now build Harry Potter Mirror of Erised, showing the "raw feed" of you in the mirror but augmented with your deepest desires (as inferred by the AI). - I don't know, I'm probably missing the biggest one, so many things! (Disclosure I am (very small) angel investor in Decart, I was excited because imo this technology will get very good very fast and it feels general, powerful but it's also technically very difficult. Congrats on the launch to the team!)

Noah has raised $22M to build the global payment network for the stablecoin era! Welcome Thijn Lamers (ex-Adyen) as Co-Founder and President. Backed by @localglobevc, @FelixCapital, @FjLabs + angels incl. @JTLonsdale (Palantir), @davidhelgason (Unity), @matthey_alex (ex-Adyen CTO) & Tom Stafford (DST) - thank you for believing in us. Read the Fortune exclusive: fortune.com/crypto/2025/06…

with open source ai wagmi

Exactly the right question, posed by @JamesFrith in Westminster Hall today. "Why do the loudest AI tech companies expect to train their machines on human-created content for nothing?" "Hear, hear."

The full report by the Tony Blair Institute is out, in which they detail how copyright should be "rebooted" in the AI age. It is - and I don't say this lightly - terrible. It reads like a cross between a big tech lobbying document and a funding proposal for a new academic centre no one wants (it literally includes this). I've pulled out some quotes from the report and added my thoughts below. I wish I had time to write a more in-depth rebuttal. But this should give you an idea of its contents. I encourage you to read the whole thing yourself. ⬇️📄⬇️ “The UK government has proposed a text and data mining exception with the possibility for rights holders to opt-out. This … [would give] rights holders increased control of how their data are used.” This is not true. Existing UK copyright law gives rights holders full control over how their works are used by UK AI companies. Moving to an opt-out model would inevitably mean that many rights holders would miss the chance to opt out or fail to opt out all their works. It could *only* reduce the amount of control rights holders have. “A lack of clarity harms all stakeholders. This includes creators, who are not properly remunerated for their labour”. There is no lack of clarity over the current law; and creators are only not remunerated for their labour if British AI companies break the law. “Today, the application of UK copyright law to the training of AI models remains contested.” This is not true. I have seen zero arguments that commercial gen AI training on copyrighted work without a licence is legal in the UK. Even AI companies understand it is currently illegal. “Currently, United Kingdom copyright law provides insufficient clarity for creators, rights holders, developers and consumer groups”. Same point, still not true. “The free flow of information has been a key principle of the open web since its inception.” Putting your content online does not entitle anyone to use it for whatever they like, for free. “To argue that commercial AI models cannot learn from open content on the web would be close to arguing that knowledge workers cannot profit from insights they get when reading the same content.” This is a classic big tech talking point, but it is incredibly misleading. That humans will learn from your work is assumed when you create; we do *not* assume commercial gen AI models will be trained on it. Moreover, commercial gen AI models scale in ways no individual human can. Commercial gen AI creates hyper-scalable competitors to creators by exploiting their work. “Generative AI is here to stay. Already, music generator Beatoven .ai has built a fully licensed generative music model, and KL3M has produced a “fairly trained” large language model (LLM).” Yes - both these companies are certified under the Fairly Trained certification scheme I run. Both vehemently object to unlicensed training; both prove that it is possible to build gen AI models without stealing people’s work. These are terrible examples if the authors are trying to argue that we must legalise IP theft. They show the opposite - that it is possible to build gen AI without the theft. “Photography and sampling are key examples of technologies that sparked debates about creative ownership but ultimately led to artistic renewal rather than extinction.” Yes, but cameras are not built by exploiting the work of the world’s creators, and music samples have to be licensed. These are not in any way comparable to building gen AI based on IP theft. “But, in general, copyright controls copying; it does not control other ways in which those engaging with the material might use its intellectual content.” Training AI models involves copying. It is hugely misleading to suggest otherwise. “Developers are not set to make long-term profits from publicly trained data [so AI companies shouldn’t have to pay for training data].” This is an incredibly surprising argument. Why would VCs be investing so many billions of dollars into this space if there were not long-term profits to be made? “It is hotly debated whether model weights should be thought of as a copy of the training data.” This is intentionally misleading. Many of the lawsuits simply claim that training involves copying, which is not disputed in the slightest. (As an aside, it is very odd that they quote the robots.txt page from my personal site, when I have publicly pointed out that this is Squarespace’s default robots.txt page which cannot be changed. I would hope the people writing this report understood robots.txt well enough to be aware of that.) “Generative AI may never be good enough to be a substitute for all human activities for which people get paid.” This is a straw man. The competitive effects of gen AI on the works it’s trained on, and the people behind those works, can be huge, without AI being a substitute for “all human activities for which people get paid”. “Generative AI will continue to improve but its greatest value is likely to be in augmenting existing workflows, as demonstrated by platforms such as Invoke AI.” This *totally* ignores the mounting evidence that generative AI competes with its training data. Every study so far reinforces this common-sense hypothesis, e.g. that the introduction of ChatGPT decreased demand for freelance writing tasks by 30%. THEIR PROPOSED SOLUTIONS “The UK government should establish the Centre for AI and the Creative Industries.” I have no particular issue with the idea of a new academic centre. But it is totally nonsensical to propose this as some kind of solution to the AI / copyright problem. It is orthogonal to the question at hand. Why would creators give up strong copyright protections, which they desperately want, for a new academic institute, which no one is asking for? This is an absolutely insane proposition, and I am genuinely shocked it made it into the report. (It’s also worth noting that an early draft of the report I saw suggested that an eye-popping £50M a year would be a good price tag for this centre.) They suggest “a targeted levy on ISPs” to remunerate creators for their work being used. To be clear, this is a tax on consumers (their words) - “pennies per month for a regular household”. Charging consumers for training data, rather than the AI companies that exploit it commercially, is perhaps the most brazen big-tech talking point of the entire report. And, even more astonishingly, they argue that “the priority of this revenue would be funding the Centre for AI and the Creative Industries.” They want to let AI companies pay nothing for training data; instead charge the general public; and not even give the new tax revenue to creators, but instead give it to an unwanted academic centre. -- This report reads like it was written by big tech, capped off by a truly astonishing set of proposals that would be punitive to both creators and the general public - everyone, that is, except the AI companies who want to exploit creators' life's work for free, to build AI models that compete with them. It's perhaps not surprising that the Tony Blair Institute would align so closely with big tech interests. We can only hope the government doesn't give this report the same weight as the protests of the country's creators. Full report in the next tweet.

I think this is accurate, and the politicians favouring AI companies at others’ expense will rapidly become unpopular.

Thrilled to announce @IsomorphicLabs has raised $600M to turbocharge our mission to one day solve all disease with the help of AI. I've long felt that improving human health is the most important thing we can do with AI & today marks a big step towards a new era of drug discovery.

Most of my MIT colleagues are in denial of this coming tsunami

If it weren’t for DeepSeek, they would still be feeding us incremental improvements on benchmarks by increasing by 1 percentage point one after another for 5 more years. Now they had to reveal all they had and give it for free. DeepSeek saved us 5 years. Thank you, @deepseek_ai.

🛠️ With over 100+ robots deployed across 30+ cities 🤝 Partnered with DeepMind on cutting-edge research 🚀 Aiming for @BitRobotNetwork, the platform for distributed robotics research We’re excited to back @frodobots in their $6M Seed round to take this vision to the next level. Read why we invested here: medium.com/fabric-venture…

More coverage of the backlash to the Christie’s AI art auction. I continue to think people in AI have hugely underestimated the anger among artists about the mass theft of their work to build commercial gen AI models.