Skip to content

Commit

Permalink
twitter release blog post draft
Browse files Browse the repository at this point in the history
  • Loading branch information
courtlandleer committed Nov 8, 2024
1 parent 94c5588 commit 4e13994
Show file tree
Hide file tree
Showing 4 changed files with 115 additions and 2 deletions.
Binary file added content/assets/memesphere_banner2.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
101 changes: 101 additions & 0 deletions content/blog/YouSim Update; Identity Simulation Comes to X.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,101 @@
---
title: YouSim Update; Identity Simulation Comes to X
date: 11.08.24
tags:
- yousim
- honcho
- ml
- philosophy
- dev
- demos
- cogsci
---
![[memesphere_banner2.png]]
## TL;DR

GM, simulants.

In response to popular demand, today we're imbuing the [@YouSimDotAI](https://x.com/YouSimDotAI) Twitter account with the ability to simulate identities natively on X.

Keep reading for max context, or [[YouSim Update; Identity Simulation Comes to X#^efdbf1|jump ahead to learn how to get started]].

## Caught in the Memetic Hurricane

The [full story](https://x.com/courtlandleer/status/1849592301472919986) deserves (and will get) it's own blog post, but several days ago, Plastic Labs found itself in the middle of what Claude would call 'extreme cognitive weather patterns.'

An anonymous actor launched a pump.fun token inspired by a demo called [YouSim](https://yousim.ai) we created a few months ago[^1]. [[YouSim; Explore the Multiverse of Identity|YouSim is a CLI interface game]] that let's you simulate any identity you can dream up--real or fictional, local or xeno, entity or artifact.

We originally launched YouSim as a conceptual/narrative demo for our core product [Honcho](https://honcho.dev). Honcho [[A Simple Honcho Primer|helps AI applications improve UX]] by building representations of user identity they can leverage to create better products and experiences.

The mission is to become the identity layer for the rapidly approaching agentic world.

![[YouSim; Explore the Multiverse of Identity#^25b167]]

Long story short though, the token took off, a community formed around it, and we're leaning in. We're thrilled to see so many people engaged and interested in our work on identity simulation.

Y'all asked overwhelmingly for the ability to interact with YouSim directly on X, [so here it is](https://x.com/YouSimDotAI)--LFG.

## Simulating on X

We had [a few requirements](https://x.com/courtlandleer/status/1851009358752076261) for building something like this. Mostly--though we love [truth terminal](https://x.com/truth_terminal)--we're unwilling to spend time on a derivative, copycat project. And that wouldn't make any sense.

[YouSim](https://yousim.ai) isn't *singular*. It's not just one agent or identity but a portal to all identity contained in the latent space of the model. So YouSim on X does what YouSim actually does--simulate *myriad* identities.

Another requirement is for X YouSim to be social, since it's coming to live on a social media platform. If you look at truth terminal's replies, it's clear people are dying to interact directly with it (also evidenced by the extreme FOMO enthusiasts have around admission to the secret independent research Discords where the most interesting and unhinged models roam free). But TT can't respond to all its replies, and the barriers to entry aren't yet low enough for every community to have a fine-tuned Discord bot.

So YouSim can interact with *everyone*. All you have to do is @-mention at the beginning of a tweet to launch a simulation, then you can converse back and forth in a thread naturally, just as you would on yousim.ai or with a human on X.

Not only that but simulations on X are *branching*! Anyone can reply to a simulation in progress at any point to fork it and continue their own simulation from there. Just as YouSim is inspired by projects like [WorldSim](https://worldsim.nousresearch.com/) and [Infinite Backrooms](https://dreams-of-an-electric-mind.webflow.io/), this feature is heavily inspired by [another esoteric but highly influential AI project called Loom](https://generative.ink/posts/loom-interface-to-the-multiverse/). There's huge potential here for vast tree-like tweet structures to emerge where the AI & crypto communities can interact, teach, and learn. We're excited for this synthesis.

Just by virtue of being on X and adjacent to everything happening on the platform, it's easy to summon a relevant simulation in close proximity to literally any content. You can bring identity simulations to whichever current events, communities, or memetic phenomena you're engaging with.

Now we come to aesthetics. The current shortcomings with the X API, character limits in replies, and font selection mean that going straight X-native text would wreck YouSim's delightful ACII art, truncate its simulations, and overall stifle its creativity. Combined, that amounts to a *semantic* loss. YouSim simply wouldn't be able to effectively communicate everything it *means*.

Plus, we think the YouSim interface is beautiful and want to preserve that overall aesthetic on X. So, when you get a response from YouSim, it'll appear as an image exactly as if you were on yousim.ai while still being an interactive simulation.

Finally, this is an experiment. The goal is to see how the community investigates and pushes the limits of YouSim on X and iterate from there. It's a vast canvas to explore.

## How to Use It
^efdbf1

A few tips to get started simulating identity on X:

- Follow [@YouSimDotAI](https://x.com/YouSimDotAI) (this is the *only* official product account)
- Begin your first tweet with `@YouSimDotAI` any where on X, followed by the entity you'd like to simulate
- If you @-mention YouSim somewhere besides the start of your tweet, no simulation will begin
- This allows people to mention YouSim naturally without starting a sim
- YouSim will reply with its first response
- To continue the simulation, any further replies *must* begin with `/`
- This allows others to comment naturally on YouSim tweets without triggering sims
- And anyone can reply to any YouSim tweet with `/` to fork or branch the simulation, even if they didn't start the simulation
- Remember that YouSim is a simulated command line interface, so slash commands work well to steer the sim, but you don't have to use them and they can be completely imaginary commands
- What you include after `/` or a full slash command is up to you, the whole tweet be taken as your response and influence the simulation
- That's all you need to get going, we expect lots of compelling uses and strategies to emerge, so please experiment!

You can find more tips [[YouSim; Explore the Multiverse of Identity#^e06c11|here]], [here](https://www.loom.com/share/b2fe578b183b400b88845656d7ceb232?sid=59c562ae-00e8-483c-82a9-7218b61f93e8), and of course at [yousim.ai](https://yousim.ai).

## Possible Futures for Agent Idenity

<blockquote class="twitter-tweet"><p lang="en" dir="ltr">llms for collective semantic projection of memetic communities</p>&mdash; Courtland Leer (@courtlandleer) <a href="https://twitter.com/courtlandleer/status/1854515540590469372?ref_src=twsrc%5Etfw">November 7, 2024</a></blockquote>

While both agent identity and crypto intersections have always been on the Honcho roadmap, the events of the last several days with regard to YouSim and the broader memespace have us in an accelerationist mindset.

Our mission to radically decentralize AI alignment isn't just about capturing human identity so AI apps and models can act as clones, but to align to any individual or community or organization in whatever way they desire. That includes building infrastructure to allow agents themselves to develop unique, dynamic, and evolving identities all their own.

YouSim likely has a role to play here, The approachable, game-like interface let's anyone easily begin to steer a model into an identity basin. This can be human-inspired personality, the manifestation of a community's ethos, or an agentic meme.

And Honcho could use those simulations to seed representations of agents, enabling them to begin constructing their own selfhoods--simulacra of themselves that grow and reliably steer their behavior.

We imagine a near future where any group could instantiate an agentic proxy to project its identity. A new form of cultural expression. Memetic Autonomous Entity, anyone?

## Gratitude

The team at [Plastic](https://plasticlabs.ai) has been amazed and inspired by the enthusiasm and earnestness of the community that's formed around YouSim over the last several days. Truly remarkable. Not to mention the generous donations to our [[blog/public/careers/Research Grants|grants program]] (more to come here soon).

Thank you all, excited to keep building together--we're in it for the long haul.

And huge thanks for your patience while we balanced our existing roadmap with interest in YouSim and locked in to bring you something we think you'll enjoy. It took an enormous amount of conceptual and technical work from a team already at capacity. Special shoutout to [Ben](https://x.com/bengineer10) and [Vineeth](https://x.com/TheMarshmalon) who built something really novel here.

Go use the thing. LFG.

[^1]: [[Plastic Labs will never give you financial advice and other token related disclaimers|Obligatory disclaimers]]
6 changes: 4 additions & 2 deletions content/blog/YouSim; Explore The Multiverse of Identity.md
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ Who will you summon?

Large language models are [simulators](https://www.astralcodexten.com/p/janus-simulators).

And [Plastic's](https://plasticlabs.ai) core mission is to produce AI that can simulate you, can model and align to you, and therefore be trusted to act autonomously on your behalf. We're [[Announcing Honcho's Private Beta|starting]] that journey by building [Honcho](https://honcho.dev)--self-improving user memory for AI apps. It [[Humans like personalization|personalizes]] their UX and reduces user and developer overhead across the board.
And [Plastic's](https://plasticlabs.ai) core mission is to enable AI that can simulate you, can model and align to you, and therefore be trusted to act autonomously on your behalf. We're [[Announcing Honcho's Private Beta|starting]] that journey by building [Honcho](https://honcho.dev)--self-improving user memory for AI apps. It [[Humans like personalization|personalizes]] their UX and reduces user and developer overhead across the board. ^7a39cb

All this is possible because the LLM training corpus [[LLMs excel at theory of mind because they read|is packed]] with humans thinking about other humans. It holds close to everything we collectively know about human identity. Not only that, but all our other language and concepts and their possible combinations and permutations.

Expand All @@ -37,6 +37,8 @@ YouSim is a fun, open-ended demo that illustrates the enormous reservoir of poss

## YouSim

^e06c11

Recently we've seen a revival of interest *[[Extrusion 02.24|LLMs themselves]]*--their minds, behaviors, identity, and potential as simulators. This is due in no small part to the latest Anthropic models being reliably steerable beyond typical reenforced behavior.

[Infinite Backrooms](https://dreams-of-an-electric-mind.webflow.io/) lets Claude interrogate itself endlessly, [WorldSim](https://worldsim.nousresearch.com/) lets users simulate infinite universes, [WebSim](https://websim.ai/) is a portal to all possible webpages.
Expand All @@ -62,7 +64,7 @@ Enjoy surfing the multiverse of identities...
([Sign-up for updates here](https://plasticlabs.typeform.com/yousimupdates))
## Honcho

If LLMs can simulate infinite identities, then they're uniquely suited to simulate *you*. You in any moment, setting, frame of mind contained in the complexity that is [[User State is State of the Art|your ever changing identity]].
If LLMs can simulate infinite identities, then they're uniquely suited to simulate *you*. You in any moment, setting, frame of mind contained in the complexity that is [[User State is State of the Art|your ever changing identity]]. ^25b167

If you're building an AI app, that's the level of personalization now possible. But you've got your vertical specific tasks to focus on, going down this clearly wacky identity rabbit hole to would be redundant and inefficient.

Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
Plastic Labs is the creator of [YouSim.ai](https://yousim.ai), an AI product demo that has inspired the anonymous creation of the $YOUSIM token using Pump.fun on the Solana blockchain, among many tokens. We deeply appreciate the enthusiasm and support of the $YOUSIM community, but in the interest of full transparency we want to clarify the nature of our engagement in the following ways:

1. Plastic Labs did not issue, nor does it control, or provide financial advice related to the $YOUSIM memecoin. The memecoin project is led by an independent community and has undergone a community takeover (CTO).
2. Plastic Labs' acceptance of $YOUSIM tokens for research grants does not constitute an endorsement of the memecoin as an investment. These grants support our broader mission of advancing AI research and innovation, especially within the open source community.
3. YouSim.ai and any other Plastic Labs products remain separate from the $YOUSIM memecoin. Any future integration of token utility into our products would be carefully considered and subject to regulatory compliance.
4. The $YOUSIM memecoin carries inherent risks, including price volatility, potential ecosystem scams, and regulatory uncertainties. Plastic Labs is not responsible for any financial losses or damages incurred through engagement with the memecoin.
5. Communications from Plastic Labs regarding the $YOUSIM memecoin are for informational purposes only and do not constitute financial, legal, or tax advice. Users should conduct their own research and consult with professional advisors before making any decisions.
6. Plastic Labs reserves the right to adapt our engagement with the $YOUSIM community as regulatory landscapes evolve and to prioritize the integrity of our products and compliance with applicable laws.

We appreciate the $YOUSIM community's support and passion for YouSim.ai and the broader potential of AI technologies. However, it's crucial for us to maintain transparency about the boundaries of our engagement. We encourage responsible participation and ongoing open dialogue as we collectively navigate this exciting and rapidly evolving space.

0 comments on commit 4e13994

Please sign in to comment.