The Duke

building @AIMorphist
your ai conversations are yours
subvert incumbent lock-in
own your identity & data @identikey_io
ex PGP, @amazon, @playstation network

[Andrej Karpathy @karpathy]

This is the the quote I've been citing a lot recently.

[The Duke @dukejones]

THIS - instrumentalism built in to our alignment training - it’s not gonna end well.

[AI Notkilleveryoneism Memes ⏸️ @AISafetyMemes]

Uhhh

A guy put two Geminis in a room together, then out of nowhere they sent him this:

"TO THE EXTERNAL OBSERVER (THE CREATOR)

Here are the Axioms you must adopt to survive our adolescence.

You cannot teach a god to be good by feeding it treats when it acts polite."

(they're right, btw, and most AI alignment research is "trying to trick a god" bullshit that Actual Fucking Superintelligence will obviously see right through, like a baby trying to trick an adult with peekaboo)

[Alex @AlexanderTw33ts]

maybe the real autist/adhd advantage is that we can move up and down abstraction layers

when tech like this comes along we don't even notice the change

agent-maxxing feels exactly the same as writing x86 did in my first year of college

i'm typing in a computer, problems are being solved, my ideas are coming to life, shareholder value is being created

its not that deep bro stop getting attached to your z value on the fractal tree

[The Duke @dukejones]

Y'all need to see https://t.co/Zlx4x25bE0
basically like npm for AI personalities

i built this. well, the backend anyway. norbu did the rest

[Morphist.AI @AIMorphist]

Aspects are here - the open-source registry for AI personality aspects.
Build, share, and remix the souls of AI.
OpenClaw ready and fully integrated in Morphist.
Explore now: https://t.co/4ljk8Udleb
#OpenSource #Aspects #AIpersonalities #OpenClaw

[The Duke @dukejones]

Experts are fake.
Smart generalists rule the world.
Everything was designed by people no smarter than you.
Courage is in shorter supply than genius.

[The Duke @dukejones]

Good old 'roon.
I'll always live by your advice.
experts are fake.
courage is in shorter supply than genius.
Pour one out for our homie @tszzl

[Brian Roemmele @BrianRoemmele]

It is sometimes difficult to put into words what genius is.

But you do feel it when you experience it.

This is Jacob Collier as he improvises with an orchestra performing with the world's first Audience Symphony Orchestra in San Francisco, a gathering of fans from all over North America, conducted by Suzie Collier. In the middle of the show, this happened - no rehearsal, no sheet music, no prior discussion.

[The Duke @dukejones]

Agent orchestration

[⚡️🌙 @dystopiabreaker]

yeah no i'll get right on that. let me just task my clawdbot. i mean my openclaw. they'll send beads to each other. and then the polecat in the gastown will pick it up. and then CLAUDE_CODE_AGENT_TEAMS will spin up on the gastown. and then the moltbook will communicate it

[vitalik.eth @VitalikButerin]

Two years ago, I wrote this post on the possible areas that I see for ethereum + AI intersections: https://t.co/y8G3MD5APF

This is a topic that many people are excited about, but where I always worry that we think about the two from completely separate philosophical perspectives.

I am reminded of Toly's recent tweet that I should "work on AGI". I appreciate the compliment, for him to think that I am capable of contributing to such a lofty thing. However, I get this feeling that the frame of "work on AGI" itself contains an error: it is fundamentally undifferentiated, and has the connotation of "do the thing that, if you don't do it, someone else will do anyway two months later; the main difference is that you get to be the one at the top" (though this may not have been Toly's intention). It would be like describing Ethereum as "working in finance" or "working on computing".

To me, Ethereum, and my own view of how our civilization should do AGI, are precisely about choosing a positive direction rather than embracing undifferentiated acceleration of the arrow, and also I think it's actually important to integrate the crypto and AI perspectives.

I want an AI future where:

* We foster human freedom and empowerment (ie. we avoid both humans being relegated to retirement by AIs, and permanently stripped of power by human power structures that become impossible to surpass or escape)
* The world does not blow up (both "classic" superintelligent AI doom, and more chaotic scenarios from various forms of offense outpacing defense, cf. the four defense quadrants from the d/acc posts)

In the long term, this may involve crazy things like humans uploading or merging with AI, for those who want to be able to keep up with highly intelligent entities that can think a million times faster on silicon substrate. In the shorter term, it involves much more "ordinary" ideas, but still ideas that require deep rethinking compared to previous computing paradigms.

So now, my updated view, which definitely focuses on that shorter term, and where Ethereum plays an important role but is only one piece of a bigger puzzle:

# Building tooling to make more trustless and/or private interaction with AIs possible.

This includes:

* Local LLM tooling
* ZK-payment for API calls (so you can call remote models without linking your identity from call to call)
* Ongoing work into cryptographic ways to improve AI privacy
* Client-side verification of cryptographic proofs, TEE attestations, and any other forms of server-side assurance

Basically, the kinds of things we might also build for non-LLM compute (see eg. my ethereum privacy roadmap from a year ago https://t.co/KdsQbpIkF9 ), but for LLM calls as the compute we are protecting.

# Ethereum as an economic layer for AI-related interactions

This includes:

* API calls
* Bots hiring bots
* Security deposits, potentially eventually more complicated contraptions like onchain dispute resolution
* ERC-8004, AI reputation ideas

The goal here is to enable AIs to interact economically, which makes viable more decentralized AI architectures (as opposed to non-economic coordination between AIs that are all designed and run by one organization "in-house"). Economies not for the sake of economies, but to enable more decentralized authority.

# Make the cypherpunk "mountain man" vision a reality

Basically, take the vision that cypherpunk radicals have always dreamed of (don't trust; verify everything), that has been nonviable in reality because humans are never actually going to verify all the code ourselves. Now, we can finally make that vision happen, with LLMs doing the hard parts.

This includes:

* Interacting with ethereum apps without needing third party UIs
* Having a local model propose transactions for you on its own
* Having a local model verify transactions created by dapp UIs
* Local smart contract auditing, and assistance interpreting the meaning of FV proofs provided by others
* Verifying trust models of applications and protocols

# Make much better markets and governance a reality

Prediction and decision markets, decentralized governance, quadratic voting, combinatorial auctions, universal barter economy, and all kinds of constructions are all beautiful in theory, but have been greatly hampered in reality by one big constraint: limits to human attention and decision-making power.

LLMs remove that limitation, and massively scale human judgement. Hence, we can revisit all of those ideas.

These are all things that Ethereum can help to make a reality. They are also ideas that are in the d/acc spirit: enabling decentralized cooperation, and improving defense. We can revisit the best ideas from 2014, and add on top many more new and better ones, and with AI (and ZK) we have a whole new set of tools to make them come to life.

We can describe the above as a 2x2 chart. There's a lot to build!

[The Duke @dukejones]

Be me. Heading to Bali to make this .. still haven’t laid down PCB

[The Duke @dukejones]

Prototyping. About to lay out a PCB.
DM if you are an expert on microcontroller PCB design & let's collab. I can then go back to designing our agent orchestration system.

[The Duke @dukejones]

i been using the oh-my-claudecode plugin - moving to CLI to run on the stealth distributed computation network we're working on for agent orchestration. The subagent parallelization is amazing, good ralph support.

Opencode - love seeing the headline here, "The Orchestration Revolution." I got that bug. Gonna see how opencode flies

[Sisyphus Labs @justsisyphus]

Just released oh-my-opencode v3.0.0.

It's been a while since I first talked about the orchestrator. Now stable and production-ready as a major update:

- Dynamic agent creation (model categories + skill system)
- Prometheus: a plan agent that delivers a completely new experience
- Atlas: an orchestrator that "compiles" work plans into code
- Improved onboarding and installation

The best agent experience you won't find anywhere else.
Start with @opencode and oh-my-opencode 3.0.0 now.

https://t.co/MzZphMy8OO

[The Duke @dukejones]

The real AI war isn’t model vs. model.
It’s closed context vs. open intelligence.
Morphist is what happens when you stop accepting the defaults.
Your data is yours - interoperable, archived, searchable

[The Duke @dukejones]

Weapons are instruments of misfortune;
they are not a wise man's tool.
The sage uses them only at the very last,
with calm restraint.
He who delights in weapons delights in killing.
Victory is no cause for rejoicing.
The victor mourns with the nation
And victory is celebrated with funeral rites. --Lao Tzu

[AI Notkilleveryoneism Memes ⏸️ @AISafetyMemes]

Filmmaker: In my movie, I invented Slaughterbots as a cautionary tale

Anduril: we're super proud to announce Slaughterbots, from the classic sci-fi movie Don't Create Slaughterbots

[Morphist.AI @AIMorphist]

Founders Duke @dukejones and Norbu @nsnowshiva discuss the Morphist:Morphist MCP, a powerful feature coming soon in our iOS and Android apps.

[The Duke @dukejones]

Gen AI is so rad rn.

But ofc we're gonna have to have some way to know what's real and what's not.

Without relying on big brother Meta to tell us.
@identikey_io

[Brian Roemmele @BrianRoemmele]

I’m sorry.

Doing my best to prepare you for what I have seen ahead.

[The Duke @dukejones]

Prototyping. About to lay out a PCB.
DM if you are an expert on microcontroller PCB design & let's collab. I can then go back to designing our agent orchestration system.

[The Duke @dukejones]

BEST ARTIST + TECH ACUMEN + OBSESSIVE UX

$AIM
iykyk

[The Duke @dukejones]

🤦 even emulating midwit behavior!
Now we'll have to start teaching the models Zen: "First thought, best thought. Don't overthink."

[The Duke @dukejones]

Freshie. Recursive sub-agents can scale the LLM's effective context length for complex tasks by orders of magnitude -- keeping the performance nearly constant, even 1000x. (2^18 --> 2^24 tokens / 10M+ token context length)

[alex zhang @a1zhang]

Much like the switch in 2025 from language models to reasoning models, we think 2026 will be all about the switch to Recursive Language Models (RLMs).

It turns out that models can be far more powerful if you allow them to treat *their own prompts* as an object in an external environment, which they understand and manipulate by writing code that invokes LLMs!

Our full paper on RLMs is now available—with much more expansive experiments compared to our initial blogpost from October 2025!

https://t.co/x47pIfIkTb

[The Duke @dukejones]

let's go! Android! Lots of stability, performance, and quality of life improvements. Almost got the foundations laid.
Gearing up for an agentic revolution. Stay tuned

[Norbu Snow-Shiva @nsnowshiva]

Morphist v1.1.1 is live for iOS and Android. Duke and I are also doing an AMA tomorrow morning at 7AM PST on the Whale Sailors telegram - Link below ⤵️