Josh Chen

Playing with AI before it plays with us. Founded @basisprotocol, early team @Merantix @Opal_Sec

[Josh Chen @jwaynechen]

@fun F A F!

[Josh Chen @jwaynechen]

@maxkolysh Actually BART is notoriously bad lol https://t.co/vBnCiIeG43

[Josh Chen @jwaynechen]

@AnteOrg This looks key! (No pun intended)

[Shaun Maguire @shaunmmaguire]

Also, at the time of posting this (16 hours after his)

His post only has 6.1k views and 126 likes

The original stories about his arrest got millions of views

It is so much easier to destroy a reputation than to build one

@nadertheory never gave up even when it was hard 🫡

[Josh Chen @jwaynechen]

In all my years of crypto I've never seen the SEC dismiss a case like this **with prejudice**. Congratulations @nadertheory and I'm excited for your road ahead.

[nader.deso @nadertheory]

Big news! Last Friday the SEC officially dismissed its enforcement action against me and against DeSo.

This was the last legal issue I had to deal with and I am now completely and totally free to innovate and build again, unhobbled for the first time in years.

Three important points:

1) This dismissal was NOT a settlement. It was "without costs or fees" to me or anyone involved (extremely rare) because there was no wrongdoing and no actual aggrieved parties.

2) This dismissal was "with prejudice" (also rare). This means they can't bring any related action back against me or DeSo in the future.

3) In the SEC's own words, it was based on "a reassessment of the evidentiary record," meaning the actual facts regarding my innocence were heavily scrutinized and drove the decision.

Simply put: The government made a mistake in bringing this case in the first place.

The government accused me of misleading an investor who I knew I had a great relationship with, as in they backed me two separate times and I literally had breakfast with them at their house not long prior to the charge. As a result, soon after the charge I found out that not only were they not upset with me, but they wanted the government to go away as badly as I did. As I understand it, the government compelled the investor to do an interview and then took their neutral testimony and represented it as adversarial. It was an alleged fraud with no actual misrepresentation nor any actual aggrieved parties. My lawyers said they'd never seen anything like this, and I think it speaks to how dogmatically anti-crypto the prior administration's SEC was.

In the coming days and weeks, I will be hopping on some podcasts to tell the whole story, and boy is there a story to tell. Stay tuned, and if you know anyone who'd like to have me on as a guest please reach out.

I'm also excited to start sharing more about what my team has been working on soon. We haven't been twiddling our thumbs.

For now, though, I just want to explain why DeSo is so important to me. DeSo is still the only platform on the internet where you can post content directly to a blockchain without fear of censorship, and where you can monetize your content directly with crypto (including stablecoins). It's really quite shocking how in 2026 we not only have virtually no viable alternative for this clearly-important category, but also other important efforts are actually shutting down. The world needs more people working on decentralizing social media, not less. I'm excited to finally be able to share our vision directly again, and to start bringing more people who care about freedom and censorship into our community. What we have built with DeSo is something people take for granted until they really need it, but hopefully we can convince them sooner than that.

Lastly I want to say how grateful I am to everyone around me. My family, my friends, my backers, and everyone in the DeSo community. For me, this experience showed me just how trusting, loyal and caring everyone around me really is, and reaffirmed my belief that always trying to do the right thing really does pay off.

We're just getting started.

[Josh Chen @jwaynechen]

@ezradar @withcaliza Nice to see such rapid global expansion 👏

[Andrew Jeffery @credealjunkie]

If you support rent control, and walked into this building, you might change your position.

This is the worst rent roll I’ve ever seen and it’s not even close.

[Josh Chen @jwaynechen]

@mikefutia Sora

[Josh Chen @jwaynechen]

Beautiful discussion of various sources of inspiration for training LLMs

[Andrej Karpathy @karpathy]

Finally had a chance to listen through this pod with Sutton, which was interesting and amusing.

As background, Sutton's "The Bitter Lesson" has become a bit of biblical text in frontier LLM circles. Researchers routinely talk about and ask whether this or that approach or idea is sufficiently "bitter lesson pilled" (meaning arranged so that it benefits from added computation for free) as a proxy for whether it's going to work or worth even pursuing. The underlying assumption being that LLMs are of course highly "bitter lesson pilled" indeed, just look at LLM scaling laws where if you put compute on the x-axis, number go up and to the right. So it's amusing to see that Sutton, the author of the post, is not so sure that LLMs are "bitter lesson pilled" at all. They are trained on giant datasets of fundamentally human data, which is both 1) human generated and 2) finite. What do you do when you run out? How do you prevent a human bias? So there you have it, bitter lesson pilled LLM researchers taken down by the author of the bitter lesson - rough!

In some sense, Dwarkesh (who represents the LLM researchers viewpoint in the pod) and Sutton are slightly speaking past each other because Sutton has a very different architecture in mind and LLMs break a lot of its principles. He calls himself a "classicist" and evokes the original concept of Alan Turing of building a "child machine" - a system capable of learning through experience by dynamically interacting with the world. There's no giant pretraining stage of imitating internet webpages. There's also no supervised finetuning, which he points out is absent in the animal kingdom (it's a subtle point but Sutton is right in the strong sense: animals may of course observe demonstrations, but their actions are not directly forced/"teleoperated" by other animals). Another important note he makes is that even if you just treat pretraining as an initialization of a prior before you finetune with reinforcement learning, Sutton sees the approach as tainted with human bias and fundamentally off course, a bit like when AlphaZero (which has never seen human games of Go) beats AlphaGo (which initializes from them). In Sutton's world view, all there is is an interaction with a world via reinforcement learning, where the reward functions are partially environment specific, but also intrinsically motivated, e.g. "fun", "curiosity", and related to the quality of the prediction in your world model. And the agent is always learning at test time by default, it's not trained once and then deployed thereafter. Overall, Sutton is a lot more interested in what we have common with the animal kingdom instead of what differentiates us. "If we understood a squirrel, we'd be almost done".

As for my take...

First, I should say that I think Sutton was a great guest for the pod and I like that the AI field maintains entropy of thought and that not everyone is exploiting the next local iteration LLMs. AI has gone through too many discrete transitions of the dominant approach to lose that. And I also think that his criticism of LLMs as not bitter lesson pilled is not inadequate. Frontier LLMs are now highly complex artifacts with a lot of humanness involved at all the stages - the foundation (the pretraining data) is all human text, the finetuning data is human and curated, the reinforcement learning environment mixture is tuned by human engineers. We do not in fact have an actual, single, clean, actually bitter lesson pilled, "turn the crank" algorithm that you could unleash upon the world and see it learn automatically from experience alone.

Does such an algorithm even exist? Finding it would of course be a huge AI breakthrough. Two "example proofs" are commonly offered to argue that such a thing is possible. The first example is the success of AlphaZero learning to play Go completely from scratch with no human supervision whatsoever. But the game of Go is clearly such a simple, closed, environment that it's difficult to see the analogous formulation in the messiness of reality. I love Go, but algorithmically and categorically, it is essentially a harder version of tic tac toe. The second example is that of animals, like squirrels. And here, personally, I am also quite hesitant whether it's appropriate because animals arise by a very different computational process and via different constraints than what we have practically available to us in the industry. Animal brains are nowhere near the blank slate they appear to be at birth. First, a lot of what is commonly attributed to "learning" is imo a lot more "maturation". And second, even that which clearly is "learning" and not maturation is a lot more "finetuning" on top of something clearly powerful and preexisting. Example. A baby zebra is born and within a few dozen minutes it can run around the savannah and follow its mother. This is a highly complex sensory-motor task and there is no way in my mind that this is achieved from scratch, tabula rasa. The brains of animals and the billions of parameters within have a powerful initialization encoded in the ATCGs of their DNA, trained via the "outer loop" optimization in the course of evolution. If the baby zebra spasmed its muscles around at random as a reinforcement learning policy would have you do at initialization, it wouldn't get very far at all. Similarly, our AIs now also have neural networks with billions of parameters. These parameters need their own rich, high information density supervision signal. We are not going to re-run evolution. But we do have mountains of internet documents. Yes it is basically supervised learning that is ~absent in the animal kingdom. But it is a way to practically gather enough soft constraints over billions of parameters, to try to get to a point where you're not starting from scratch. TLDR: Pretraining is our crappy evolution. It is one candidate solution to the cold start problem, to be followed later by finetuning on tasks that look more correct, e.g. within the reinforcement learning framework, as state of the art frontier LLM labs now do pervasively.

I still think it is worth to be inspired by animals. I think there are multiple powerful ideas that LLM agents are algorithmically missing that can still be adapted from animal intelligence. And I still think the bitter lesson is correct, but I see it more as something platonic to pursue, not necessarily to reach, in our real world and practically speaking. And I say both of these with double digit percent uncertainty and cheer the work of those who disagree, especially those a lot more ambitious bitter lesson wise.

So that brings us to where we are. Stated plainly, today's frontier LLM research is not about building animals. It is about summoning ghosts. You can think of ghosts as a fundamentally different kind of point in the space of possible intelligences. They are muddled by humanity. Thoroughly engineered by it. They are these imperfect replicas, a kind of statistical distillation of humanity's documents with some sprinkle on top. They are not platonically bitter lesson pilled, but they are perhaps "practically" bitter lesson pilled, at least compared to a lot of what came before. It seems possibly to me that over time, we can further finetune our ghosts more and more in the direction of animals; That it's not so much a fundamental incompatibility but a matter of initialization in the intelligence space. But it's also quite possible that they diverge even further and end up permanently different, un-animal-like, but still incredibly helpful and properly world-altering. It's possible that ghosts:animals :: planes:birds.

Anyway, in summary, overall and actionably, I think this pod is solid "real talk" from Sutton to the frontier LLM researchers, who might be gear shifted a little too much in the exploit mode. Probably we are still not sufficiently bitter lesson pilled and there is a very good chance of more powerful ideas and paradigms, other than exhaustive benchbuilding and benchmaxxing. And animals might be a good source of inspiration. Intrinsic motivation, fun, curiosity, empowerment, multi-agent self-play, culture. Use your imagination.

[Josh Chen @jwaynechen]

@Atla_AI Look forward to trying this.

[Josh Chen @jwaynechen]

Exciting new service for coin projects

[coinwatch @coinwatchdotco]

Today we’re announcing Coinwatch Track - a new way for crypto projects to see what their market makers are doing in real-time, using their API keys.

[Josh Chen @jwaynechen]

@coinwatchdotco This is incredible. I've heard great things about this product. Exciting to see it launch! Congrats @tubergen

[GONDI @gondixyz]

Today, we unveil the new GONDI.

A full stack liquidity solution for NFTs. Collect, sell, borrow and lend.

All in one place!

[Josh Chen @jwaynechen]

@Omissionist @BillAckman This assumes the managers of the city-run grocery store actually care to work as hard as the owners of the privately-run grocery store.

It's about incentives. Sure, the DMV has an obligation to serve the public interest, but nobody there actually has the incentive to.

[Josh Chen @jwaynechen]

@tubergen So MM dumps their 2% FDV worth of borrowed tokens at launch, then (assuming coin price drops) repays the loan by exercising the call, pocketing (2% of FDV) * (launch price - strike price)?

You have any data on whether coins that accepted these deals underperformed others? 😄

[Josh Chen @jwaynechen]

@nadertheory Insane capabilities for a dex

[Josh Chen @jwaynechen]

legendary

[nader.deso @nadertheory]

Openfund is getting critical upgrades that will make trading easier than ever.

1) Real-time infrastructure, via our web-sockets integration, is now live on Openfund. This makes Openfund on-par with UX you’d see on centralized CLOBs. The infrastructure upgrades that enhance these real-time components will also apply to all other DeSo apps that use GraphQL and State-Syncer, so expect to see significant performance improvements on Focus as well.

2) The Openfund Wallet has been upgraded to support multiple quote currencies like DESO, USD & FOCUS, including the ability to cash-out from $FOCUS directly via Openfund. We also included a Stake $DESO button so that more new users entering DeSo can learn about the benefits of staking their DeSo and earning 20% yield.

3) SOL-USDC support for both cash-in and cash-out, which we added recently, is growing at record levels. This proved to be a critical upgrade as it allows users to onramp into DeSo pretty much for free (network fees on average are less than $0.001).

4) UX upgrades we’ve added include the holders table on the trade page so you don’t have to click through between Focus & Openfund. We also added hover cards to view more details about users, similar to Focus. It’s a small detail, but note that we are the only social orderbook in existence where you can see exactly who is making a trade. Imagine looking at Coinbase and knowing who’s buying or dumping your token!

5) TWAP backend has been thoroughly tested and is now ready for frontend implementation. TWAPs will provide traders with a robust set of features that allow for more precise and secure trading. It’s difficult to understate how significant of an upgrade this is. Traders will soon be able to set a flexible schedule for placing bid or ask orders, executed through derived keys, with a number of configuration options like enabling slippage protection and the ability to safeguard against unfavorable trade executions with price limits.

6) Market making incentives are being planned. We’re working on coming up with a program to incentivize more third party liquidity. As a reminder, our Python SDK makes this much easier to do now programmatically. Also, thought I’d give a little shoutout to @consensusFuture who has done nearly ~$8m in trading volume!

[Josh Chen @jwaynechen]

Persistent execution is key

[nader.deso @nadertheory]

Announcing our updated priorities for DeSo and our associated products.

https://t.co/p3LvXqvdB5

[Josh Chen @jwaynechen]

Brilliant deep dive on why tariffs won't bring back manufacturing

[molson 🧠⚙️ @Molson_Hart]

https://t.co/b9K9yNW5Dq