read docs so you don't have to | amb @ritualnet | ex @tacbuild
Joined March 2024
- Tweets21.7K
- Following1.5K
- Followers2.4K
- Likes111K
Pinned Tweet
blocks aren’t blocks anymore
@ritualnet basically looked at the blockchain design and went "wait... why tf are we still treating blocks like parking lots for txs?"
welcome to the world where blocks become containers for compute and yes, this changes everything
ok but what’s a block actually?
on most chains:
- users send txs
-txs get bundled into blocks
- blocks get validated + added to the chain
each tx is like a lil “do this” request
swap this token, update this balance, call this contract etc
simple. cute. until you try to compute anything real
but then you try to verify a zk proof
now the tx isn’t just a few storage ops
it’s “hey, run this math monster and check if it’s legit”
• needs CPU
• maybe memory
• takes time
• might need parallel threads
this ain’t just a line in the mempool anymore, it’s an actual compute job.
or take an FHE tx
that’s fully homomorphic encryption aka “do math on encrypted stuff without decrypting it”
sounds magical, until you realize it’s heavier than your GPU’s self-esteem
you need:
• special hardware
• tons of ops
• longer latency
bruhh this is more of a computational task than a tx
and now throw in AI inference
“just run this LLM for a sec” sounds easy until:
• you need GPU
• you need the right model version
• you need low latency
• you might wanna verify the output
mhh yeah… again, not a basic tx.
this is a compute container waiting to happen.
ritual goes: what if blocks weren’t for txs, but for compute?
not:
“this block has 300 swaps + 42 sends”
but, “this block ran:
• 3 zk snark verifications
• 1 GPT inference
• 2 FHE multiplications"
each with its own resource profile, latency, cost. compute-aware scheduling baked in
and tbh, it makes way more sense
why should an LLM inference + a token transfer pay gas the same way?
why should a 1-second GPU job wait in line behind 200 meme coin mints?
ritual treats blocks like compute capsules. each job is matched to the right runner, priced fairly, verified properly
that’s where the resonance fee market comes in
not “pay 21k gas lol”
but: “this task needs GPU + 500MB RAM + low-latency”
and you bid into a market that understands that. compute becomes a first-class citizen
and blocks? they evolve.
tbh blocks were never meant to do this
ethereum was made for storage.
logic + token transfers, not encrypted math, AI reasoning, or massive proofs.
ritual’s reframing the whole mental model: blocks are now compute zones, not just storage bins
and that changes how chains work entirely
you’re no longer limited to “what fits into gas” you can now schedule, match, and verify actual workloads
zk, fhe, ml, privacy, simulation...
all in one chain, all composable, all chain-integrated
mhh yeah. that’s @ritualnet.
so next time someone says “ai onchain”
ask them how they’re treating compute.
if they’re still thinking in txs + gas + old-school blocks, they’re not ready.
@ritualfnd the first to say: blocks aren’t blocks anymore.
wtf is ritual?
tbh, ritual is making multi-party computation (MPC) fully trustless using trusted execution environments (TEEs) and onchain verification. sounds amazing? let’s break it down today for your better understanding.
ok but what’s MPC?
multi-party computation (MPC) lets multiple parties compute something together without revealing their individual inputs. so it's like a group of people calculating the total bill but without anyone knowing how much each person contributed.
this is crucial for things like:
- threshold signatures (e.g. multi-sig wallets)
- private voting
- encrypted data processing
normally, MPC setups rely on some level of trust in a coordinator or server. that’s where ritual changes the game.
so what makes ritual different?
ritual removes trust from the equation by combining:
- TEEs (trusted execution environments) for secure execution
- onchain verification to check everything is legit
this means you get trustless MPC hosting without needing to trust any centralized coordinator or cloud provider.
what are TEEs and why are they important?
TEEs are isolated environments inside processors (like Intel SGX or AMD SEV) that let you run computations securely, even if the rest of the system is compromised.
for example, think of a locked glass box, you can see what’s inside, but you can’t change or mess with it. that’s basically what TEEs do for code execution gg.
ritual leverages TEEs to execute MPC operations in a way that nobody (not even the node operator) can tamper with.
onchain verification
TEEs are great, but how do you know they actually did what they were supposed to?
honestly that’s where onchain verification takes lead. ritual ensures that every MPC operation executed inside a TEE is provable onchain, so you don’t have to blindly trust the hardware.
for example, it’s like having a math teacher check your work and confirm you didn’t cheat, except the teacher here is a smart contract.
ok but why is this useful?
here’s where things get interesting tbh. trustless MPC has tons of real-world applications:
- threshold wallets: imagine a multi-sig where no single party has full control, yet no one can cheat.
- secure randomness: need verifiable randomness for a game or lottery? MPC + TEEs make it.
- decentralized identity: proving things about yourself without exposing private data.
- private auctions: bidders can submit prices without revealing them to others until the auction ends.
normally, all these rely on some level of trust in a coordinator or platform. ritual makes them fully decentralized, and that's amazing.
so who should care actually?
anyone building onchain apps that need:
- trustless cryptographic operations
- private computations
- decentralized key management
ritual makes it possible to do all this without relying on centralized infra.
so imo, ritual isn’t just about making MPC better, it’s about making it unstoppable tbh. with TEEs and onchain verification, you get:
- secure execution (no node operator tampering)
- provable integrity (onchain proof everything was done correctly)
- decentralized hosting (no reliance on centralized providers)
basically, if you’re building anything that needs privacy, security, or trustless coordination, @ritualnet is worth checking out.
ray retweeted
We’ve entered a creator renaissance. Founders, streamers, builders. The people shaping internet culture.
Some you follow, some you love, some you can’t stand. Some are rising, some are fading.
Soon your intuition about people becomes your edge. Fantasy sports for creators. Coming soon.
The Motion Official Game docs are officially live.
Everything you need to know about what Motion is and how the game works.
docs.motion.game
ray retweeted
“distribution and attention, earned through reputation”
Motion is the moat
there are only 2 durable moats left that no AI model will ever replace:
1) obsession, hunger and execution. the ability to iterate and build new moats faster than the old ones get copied.
2) distribution and attention, earned through reputation, trust and even regulation. the thing that makes people choose you and stay with you over the copycats.
2) only matters if you have 1)
ray retweeted
50M RITUAL is allocated to eligible Infernet operators, Frenrug users and long-time community members.
That’s 0.5% of initial supply, fully unlocked at launch.
Check eligibility and claim:
airdrop.ritualfoundation.org
ritual mainnet is coming and scammers are already warming up
fake claim pages, support accounts and airdrop links - they would show up before the real thing does fr. so here is a guide for newbies
so before anything drops, here are three checks that'll keep you safe:
1. your seed phrase is your wallet
nobody from ritual, no mod or support team or "official helper" will ever ask for it, ever. the moment someone asks for your seed phrase or private key, stop. it's a scam regardless of how official they look.
2. only trust links from ritual's own accounts
@ritualnet and @ritualfnd on X. that's it. a link in a reply, a DM, a discord message from someone you don't know, a google search ad - all red flags. scammers copy official usernames with one character changed. check the handle twice.
3. before you connect or sign anything, pause
if you get a text saying "your parcel couldn't be delivered, click here to reschedule" - it kinda looks real, designed to make you act fast without thinking. crypto scams work the same way. fake claim page, fake urgency and losses. if you don't understand what you're signing, don't sign it. if the link came from anywhere other than ritual's official accounts, don't click it.
scammers rely on you rushing. slow down and you beat them every time.
official ritual links only:
🔗 x.com/ritualnet
🔗 x.com/ritualfnd
🔗 ritual.net
if something feels off, ask in the official discord before clicking anything
stay safe out there 🌀 @ritualnet
ritual dropped tokenomics
10 billion RITUAL total supply.
the airdrop
50 million RITUAL. 0.5% of supply. going to people who did work
three tracks:
infernet node operators - people who ran nodes serving actual AI inference workloads. weighted by uptime and real work. they pruned the bots hard and nodes doing "default/dummy" work got minimal allocation
frenrug users - og users of the first autonomous on-chain AI agent. if you held keys and pitched the agent back when nobody knew what agents even were, you're eligible
community - long-time discord members and ambassadors who sustained participation over time
the allocation breakdown
core contributors ......... 29%
ecosystem growth + R&D .... 25%
network participation ..... 14%
labs investors ............ 11.3%
ritual foundation ......... 10%
labs treasury ............. 5%
frontier AI grants ........ 4.7%
foundation private sale ... 1%
what unlocks at launch
only three buckets unlock immediately:
- 12.5% from ecosystem growth (for liquidity and network operations)
- 5% from ritual foundation treasury
- 0.5% airdrop (fully unlocked day one, no lockup)
everything else is locked. all insider tokens locked minimum one year from mainnet launch. locked tokens cannot be staked. staking rewards flow only to unlocked circulating supply.
final tokens unlock four years after launch.
the economics long term
~5% annualized staking issuance initially paid to validators, AI executors, and delegators
EIP-1559 burn model - base fee of every transaction burned permanently. as usage grows, burned fees offset issuance
standard PoS L1 model but with AI executors as a new participant class in the rewards flow. that's different from any other chain doing this
frontier AI grants allocation (4.7%) reserved for traditional AI labs and research organizations building on ritual. research accepted to ICLR, ICML, NeurIPS. locked one year then unlocking through month 24
so ritual's treating AI research as a core moat. allocating nearly 5% of supply to academic and research partnerships is a long-term bet most chains wouldn't make
what's coming before launch
exchange and listing announcements, validator onboarding details, mainnet launch post
claim page goes live at launch for eligible participants
if you were early and contributed, check eligibility at launch
ray retweeted
Every agent wallet leak this year looked the same: the private key was somewhere the agent could read it.
On Ritual keys stay in the enclave. It never enters the agent's context, so your harness, the model, and MCP servers never touch sensitive data and your wallet stays yours.
ray retweeted
Today’s topic for DevRel Office Hours is the “The Empty Path” 🧵
2 weeks ago we gave builders one piece of homework: test what happens when nothing comes back. Today it comes due, and the Verify Challenge closes with it.
the agent economy is real enough now that attribution tools exist for it
next question is verification yk. not just where agents went but whether the decisions driving their behavior can be proven correct
that's the infrastructure @ritualnet's been building
We just solved attribution for AI Agents.
Introducing Journeys.
AI agents are visiting your product. Now you can see what they’re trying to do, what converts, and what’s making them churn.
With Journeys, you can:
- See the moment AI agents visit your product
- Follow the paths they take
- Track purchases, installs, and completed setup
- Pinpoint where agents get stuck and whats blocking them
- Review fixes prepared for you and publish them with a click
- Watch how those changes improve conversion
Turn agent traffic into revenue.
Try Journeys at tryscope.com
ray retweeted
In November 2023 we ran an onchain AI agent managing $30,000. We were early to onchain agents, and we’re early to agentic institutions.
This year, six papers took a first shot at building a fully agentic company. All six did it offchain, by hand.
If you strip them down, the same four requirements keep showing up: an identity the agent owns, memory that outlives the process, a budget that can actually run out, and work a third party can verify.
On Ritual, all four are native to the chain.
Let the Ritual begin.
ray retweeted
Four questions every Ritual callback should ask. 🧵
When your contract asks an AI model or a service for something, the answer comes back later. At Office Hours we drew the whole lesson as one picture: every answer, or non-answer, takes the same walk. Here it is.
ray retweeted
This marks a new chapter for @fluentlabsxyz
We spent a lot of time thinking about our edge as a team. Until now, we’ve largely been viewed as a “chain team.” Technically that’s true. But our DNA has always been rooted in consumer culture.
I'll make our POV clear. We don’t think there’s enough progress being made on high quality, culturally relevant consumer apps today.
So we’ve evolved Fluent Labs. Not as a typical development company servicing a blockchain, but as a consumer product studio.
@motiongame_ is our first consumer product.
One of the core ideas underpinning Motion is, you guessed it, the concept of motion. Motion is the rate and direction of change in a person or brand’s attention and reputation over time.
More specifically, attention tells you how much the market is noticing you, reputation tells you what that attention means, and motion tells you whether both are strengthening, weakening or changing direction.
You'll understand soon. XX.XX.2026.
teapawt.net/motiongame_/status/210…
ray retweeted
Six builder best practices for async results on Ritual 🧵
Your contract can call AI models and other services asynchronously. These habits help your dApp handle every outcomes smoothly: a result, an error, or no answer in time. Each one comes from the public Ritual skills.
can you think of deploying something one time and not babysitting it.
and it's not because you forgot or the loop does not need your laptop, your keeper bot, or an AWS box in your name.
that is what i am waiting to see on ritual mainnet.
the first contract that schedules its own next call, hits an API from inside a TEE, and pays the fee from its own wallet.
or you can think of a prediction market that closes itself when the clock hits. the request runs in attested hardware. winners get paid and no one pressing approve in the middle of the night.
a treasury that rebalances on a schedule it set itself. compute comes out of a balance it already holds. if the founder ragequits, the loop does not die with their server. it dies when the balance hits zero, or the chain does.
we have been calling software autonomous for years while someone's cloud bill kept it alive in the background.
Ritual's bet is narrower than the slogan you know. the contract wakes itself, calls out through a native precompile, and pays as it goes. forever only if it stays funded and the network stays up.
that is the moment i am watching for.
🤖 Made with AI
ray retweeted
Static checks find missing pieces. Tests find wrong behavior. Your Ritual dApp needs both. 🧵
At last week's Office Hours, we walked through 5 checks you can run on a Ritual project without a testnet endpoint. Here is what each one looks for.
ray retweeted
Ritual DevRel Office Hours is today: What the Checks Caught 🧵
2 weeks ago, we asked builders to run 5 verification checks on their Ritual project: find one bug, fix it, and share the before and after.
Today at 3:30 PM CT, we look at what those checks found.
ray retweeted
Six builder best practices for async results on Ritual 🧵
Your contract can call AI models and other services asynchronously. These habits help your dApp handle every outcome smoothly: a result, an error, or no answer in time. Each one comes from the public Ritual skills.
ray retweeted
Spain's data regulator published its first breach report tied to an AI agent on September 15. It named no company, product, or cause.
It named where to look: prompts, logs, conversation histories, agent memory. Memory is now a reportable surface.
cypro.co.uk/insights/cyber-b…
