Anch

Building for trust isn’t about fancy UI or “trust symbols.” It’s about giving users real visibility. You can now see your palm activity in the @VeryAI app, showing exactly when and where you’ve verified.

This paper is a clear articulation of why agent trust can't just be a model problem. the fee only vs fund involving split is the right framing most people lump all agent risk together and miss that the real exposure is pre-verification: the agent needs your money before it can prove it won't lose it. What i keep coming back to is the identity dependency underneath. the underwriter's entire pricing model - the sigmoid collateral curve, the premium loading, the false positive/negative tradeoffs. All of it hangs on knowing who the agent is and trusting that identity over time. A spoofed agent with a clean history gets cheap collateral and a low premium. That's where KYA stops being a nice to have and becomes the load bearing wall for everything in this paper. I have spent years building identity verification systems where we learned this the hard way: the financial risk model is only as good as the identity layer underneath it. liveness detection, credential delegation chains, attestation of who actually controls the entity, those are the failure modes that don't show up in simulations because the simulation assumes identity is already solved (section 7 assigns a fixed failure probability per agent, but in practice you don't know if you're even talking to the same agent twice). Genuinely curious how @t54ai is thinking about closing that loop between KYA and ARS. curious how t54 thinks about composing KYA with external agent verification layers vs building the full identity stack in house.

😀Excited to share our new paper: Quantifying Trust: Financial Risk Management for Trustworthy AI Agents. arxiv.org/pdf/2604.03976 featured in @FortuneMagazine: fortune.com/2026/04/08/age… Together with @Chi_Wang_, @tianyi_peng , @virtuals_io, and @t54ai, we propose the Agentic Risk Standard (ARS), a settlement-layer protocol that uses escrow, underwriting, and collateral to provide enforceable guarantees for transactions involving AI agents I’ve been working on agent infra since 2024, and most current works on agent infra are about Docker environments or efficiency. But is that all we need for infra for AI agents? Agents are already participating in real economic activity: providing services, making purchases, moving money. That means agent infra is not just technical, it’s also financial. Right now, using agents is basically “risk-on by default”: If things go well → everyone benefits If things go wrong → the user eats the loss That asymmetry is the real problem. 👀👀So I started thinking: what would finance for AI agents look like? We propose 👉Agentic Risk Standard (ARS)👈: 1⃣ Quantify agent risk 2⃣ Introduce escrow, premiums, collateral, underwriting, claims 3⃣ Shift from “try it and see” → “clear accountability and compensation” ⛑️Trustworthy AI shouldn’t stop at “models are reliable enough” But rather, it should answer: Who pays when things fail? How much? Under what rules? ARS moves us from model-level reliability to product-level accountability, from implicit trust to enforceable guarantees. ARS is not AI safety (which tries to make models behave) and not fintech with an AI skin (which keeps the human as the backstop). It's a risk layer: escrow holds payment, underwriting prices risk, collateral keeps agents accountable when things break. 🤗 The Agentic Risk Standard is an open-source initiative. We invite researchers, developers, and risk managers to contribute to the protocol and expand the agentic economy. paper: arxiv.org/pdf/2604.03976 github: github.com/t54-labs/Agent… website: t54.ai/ars

@ethtwit @PhoenixTrade Nyc only and not Sf?

I'm hiring a Marketing Associate to own social, content, and community for @PhoenixTrade end-to-end. If you speak crypto, are design-strong, and are allergic to being managed reach to me or apply below. You'll be in person in NYC and working with one of the strongest teams in the industry.

49M fake accounts. 7 people. 1,200 SIM boxes. now imagine you don't even need the SIM boxes. just agents with wallets, running 24/7, no human verification at the source. @VeryAI just shipped clawkey: palm scan → verify the human first, then agent gets a verifiable human backed identity. KYC verified the customer. KYA verifies the human behind the agent.

"For certain things like Fed decisions, unemployment, and GDP, Kalshi is really the only game in town" -Johns Hopkins professor and Federal Reserve researcher Jonathan H. Wright

@altitude applied

We're hiring a PMM. Stablecoin-native fintech needs a defining story. Come write it.

@AlsieLC @Visa the interesting part is visa’s distribution reach meeting dune’s data infrastructure. Such a great learning opportunity too.

Something I didn't expect when working with @Visa on this report: how deeply their crypto team understands onchain data. They're not just issuing cards. They built a stablecoin advisory team. They launched stablecoin settlement pilots. They have a Tokenized Asset Platform for banks. This partnership wasn't surface-level.

VeryAI is now live on @solanamobile! You can discover and install the VeryAI native app directly from the Solana dApp Store on your device. Native distribution for a new kind of internet: → verify real humans → prevent deepfakes → prove who and what is real Proof of Reality, now on Seeker.

@0xpratik Great opportunity and team to work with.

we're hiring! podcast ops @ eigen you’ll be a good fit if you: > get anxious when an episode isn't scheduled > know what slugline means without googling it > think show notes are actually important > have strong opinions about RSS feeds > find broken publishing workflows personally offensive > shipped something this week (even if it was just an episode) > would rather own the chaos than delegate it this is not a strategy role. this is for someone who loves execution and knows how to keep a show moving 3 month contract, with option to extend sf bay area preferred 🌁

As autonomous agents scale, the internet is facing a new fundamental question: Who’s actually behind the agent? Clawkey.ai is how we bring KYA (Know Your Agent) to life ↓

@pareen then what is?

Twitter is NOT the best place to keep track of a startup

Introducing Atlas 1. Willow's new frontier speech-to-text model. It outperforms ElevenLabs, Deepgram, OpenAI, and more by a wide margin. Built on the first scalable, human-powered transcription infrastructure ever built for real-time dictation.

moved to the US when I was 16. it made me who I am today. grateful and proud. 🫶

@itszach Inspiring. Choosing to believe in yourself and taking a leap is hard.

You can change a lot in 12 months. In March of last year, within 2 weeks I… > Left a job I loved after 3 years > Started a company > Ended a 5-year relationship Then… > Turned 25 > Raised $10M > Traveled to every continent > Built strong relationships > Put together a kickass team > Launched our product You can start over at any moment. Might as well start now.

@rareblurs The brutal truth 🫣

“unfortunately after reviewing your application..”

@pareen No way!! Coffee, culture, and vibes. But Ahmedabad is also 🤌🏼

@anchm03 mumbai is oldies

i want to hire 4 teenage coders and content creators in ahmedabad for summers will pay 2x of highest summer stipend you can get work from a hacker house with other cool builders share/RT/tag people you know or connect me to schools/colleges i can hire from

@itszach 🙌🏼🙌🏼

up and to the right