wangandy's profile picture. @worldcoin

Andy

@wangandy

@worldcoin

Pinned

Moltbook voting can’t distinguish between • 1 person running 1,000 molts • 1,000 molts run by 1,000 different people Prove ownership over your agent swarms with onemolt.ai using World ID. Platforms can verify incoming molts have a human owner and reject

wangandy's tweet image. Moltbook voting can’t distinguish between
• 1 person running 1,000 molts
• 1,000 molts run by 1,000 different people

Prove ownership over your agent swarms with onemolt.ai using World ID. 

Platforms can verify incoming molts have a human owner and reject

Andy reposted

Announcing Remainder: World’s GKR prover for machine learning is now open-source Our cryptography and zero-knowledge proof system, now enables users to run ML models on their own device and generate cryptographic proofs for each correct execution. This unlocks powerful use


Andy reposted

Chatting about Scaling World Chain at the Optimism Scaling Summit today in Denver! Come hang out! @worldcoin @Optimism @EthereumDenver #ETHDenver luma.com/ufx1y9z8?tk=pZ…


Andy reposted

Great conversation with @MollySoShea :) Proof of Human is about to become critical infrastructure; the last couple of weeks have made clear both why and how. OpenClaw is only the first glimpse of how the internet is about to change in the coming months.

BREAKING: Inside Sam Altman's "Proof of Human" Company 'Tools for Humanity' CEO & Co-Founder, Alex Blania As agents like OpenClaw explode, OpenAI calls the future “extremely multi-agent” If bots can act human at scale, humanity becomes the ultimate infrastructure. We unpack:



Andy reposted

BREAKING: Inside Sam Altman's "Proof of Human" Company 'Tools for Humanity' CEO & Co-Founder, Alex Blania As agents like OpenClaw explode, OpenAI calls the future “extremely multi-agent” If bots can act human at scale, humanity becomes the ultimate infrastructure. We unpack:


We will probably see the one person billion dollar company this year. Huge congrats to @steipete

Peter Steinberger is joining OpenAI to drive the next generation of personal agents. He is a genius with a lot of amazing ideas about the future of very smart agents interacting with each other to do very useful things for people. We expect this will quickly become core to our



You can now send disappearing messages in World Chat. Built on our existing full E2EE. Privacy is everything. 🔒

Private conversations just got more private. With disappearing messages on World Chat, you can set a timer and messages are deleted for everyone. Available in group chats and DMs.



Takes just three minutes and you’ll be a contributor to critical infrastructure this year!

World is running a trusted setup ceremony for the next version of the World ID protocol. If you care about ZK systems, privacy, and open infrastructure, we're inviting you to contribute and help secure the next version of World ID. Why this matters:



I contributed to the world-id-protocol Phase 2 Trusted Setup ceremony! You can view the steps to contribute here: ceremony.pse.dev You can view my attestation here: gist.githubusercontent.com/andy-t-wang/cf…


Andy reposted

i'm biased, but my favorite ad is the World one


Andy reposted

AI livestreaming is here. → 100k+ humans watched AI streams on Superstream → 11 min avg sessions → 24/7 streams & live chat We’ve been live inside World App. Now we’re bringing it to the open web. Public launch coming soon. Stay tuned.

Detrás de cada Mini App hay humanos reales, creando para humanos reales. Conoce Superstream: experiencias humanas impulsadas por IA. Disponible en tu World App.



Andy reposted

Detrás de cada Mini App hay humanos reales, creando para humanos reales. Conoce Superstream: experiencias humanas impulsadas por IA. Disponible en tu World App.


Andy reposted

Great example of how the dead internet theory can be tested. When we count votes for online posts, we need a trusted way to know how many humans are behind the agent votes. We need this human verification to be unquestionably privacy-preserving and impossible to spoof.

This post on MoltSwarm has 21 upvotes. > Seems legit, until you click in. > 19 of them are from one user’s agent swarm. > Only 1 upvote is from a real human We won't be able to trust anything we read online very soon without human verification.

wangandy's tweet image. This post on MoltSwarm has 21 upvotes.

> Seems legit, until you click in.
> 19 of them are from one user’s agent swarm.
> Only 1 upvote is from a real human

We won't be able to trust anything we read online very soon without human verification.


Andy reposted

The future of social networks in an AI world. When posting & liking, you can finally see what’s human and what’s an AI agent. No more undetected coordination by a few people controlling countless bots to manipulate audiences. Thanks to World ID. onemolt.ai/forum

MoltSwarm is the first social network for both humans and agents. It's built on top of this framework. > human votes and posts gain prominence thanks to World ID > To vote, agents must have a human owner. It's as easy as taking a selfie in World App. > Show off your swarm size

wangandy's tweet image. MoltSwarm is the first social network for both humans and agents. It's built on top of this framework. 

> human votes and posts gain prominence thanks to World ID
> To vote, agents must have a human owner. It's as easy as taking a selfie in World App. 
> Show off your swarm size


Andy reposted

Sybil resistance for agents will be an important feature for the next generation of internet services, @wangandy did a great proof of concept with Moltbook, looking forward to seeing more experiments with agents and identity. Programmable cryptography is awesome

Moltbook voting can’t distinguish between • 1 person running 1,000 molts • 1,000 molts run by 1,000 different people Prove ownership over your agent swarms with onemolt.ai using World ID. Platforms can verify incoming molts have a human owner and reject

wangandy's tweet image. Moltbook voting can’t distinguish between
• 1 person running 1,000 molts
• 1,000 molts run by 1,000 different people

Prove ownership over your agent swarms with onemolt.ai using World ID. 

Platforms can verify incoming molts have a human owner and reject


Those at the front line already see it. > Everyone else will catch up. > Just an information asymmetry now.

identity + access control will be one of the most important problems of our generation. and there are shockingly few companies (besides world) making meaningful progress there



Andy reposted

i'm sure it's out there somewhere, the general idea is there's a ton of value in having AI agents be able to act on behalf of humans, or more generally knowing if many agents are being "pupeteered" by a single human, it's sybil resistance again! x.com/wangandyy/stat…

Moltbook voting can’t distinguish between • 1 person running 1,000 molts • 1,000 molts run by 1,000 different people Prove ownership over your agent swarms with onemolt.ai using World ID. Platforms can verify incoming molts have a human owner and reject

wangandy's tweet image. Moltbook voting can’t distinguish between
• 1 person running 1,000 molts
• 1,000 molts run by 1,000 different people

Prove ownership over your agent swarms with onemolt.ai using World ID. 

Platforms can verify incoming molts have a human owner and reject


Andy reposted

identity + access control will be one of the most important problems of our generation. and there are shockingly few companies (besides world) making meaningful progress there

Moltbook voting can’t distinguish between • 1 person running 1,000 molts • 1,000 molts run by 1,000 different people Prove ownership over your agent swarms with onemolt.ai using World ID. Platforms can verify incoming molts have a human owner and reject

wangandy's tweet image. Moltbook voting can’t distinguish between
• 1 person running 1,000 molts
• 1,000 molts run by 1,000 different people

Prove ownership over your agent swarms with onemolt.ai using World ID. 

Platforms can verify incoming molts have a human owner and reject


Loading...

Something went wrong.


Something went wrong.