β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
Primus is already in production across multiple products, not just theory.
On the zkTLS side, the Proof-of-Reserves service provides real-time cryptographic verification of both on-chain and off-chain portfolios.
It currently covers over $72 million in TVL across four production projects, including stablecoins and RWAs such as Perenaβs USD* (with Brevis) and Unitas Labsβ USDu. Instead of relying on periodic attestations or screenshots, an attestor generates a zero-knowledge proof that the data came from the real source over a genuine TLS session, without exposing the underlying balances or credentials.
The Confidential Finance portal lets users wrap assets into encrypted form, transfer them, and earn yield while the balances remain encrypted by default. Using FHE, smart contracts perform homomorphic operations directly on ciphertext.
Primusβs AutoHoG compiler automatically designs custom multi-input gates and folds subcircuits to reduce bootstrapping, enabling a full 64-bit confidential transfer in roughly 90 ms on consumer RTX 5090s, with throughput above 210 transfers per second across eight cards. Results are deterministic because everything runs in integer arithmetic.
For developers, the Dev Hub offers data templates and SDKs so applications can request zkTLS proofs (in either MPC mode, where client and attestor jointly generate session material, or lighter Proxy mode) without building the cryptography themselves. The Loyalty Program turns those verified private data points into on-chain reputation scores and profiles.
Across the stack, @primus_labs has reached more than 1 million users (extension holders, Dev Hub builders, and zkTLS application users) and 10+ paying enterprise customers spanning stablecoins, AI, L2s, and social platforms. The same zkTLS + FHE layer underpins all of these products, with TEEs protecting runtime integrity and key material.
s.kaito.ai/0meTU1I
Primusμμ μ§μ μ¬λ¬ μΈμ¦μ ν΄λ΄λ°><
νμ¬ μΈμ¦ κΈ°λ‘ 3κ°, νν μ μλ 30μ !
@primus_labs zkTLSμ zkFHEλ₯Ό μκ°νλλ°,
μ΄λ²μλ μ§μ κ³Όμ λ₯Ό μ§νν΄ λ΄€μ΅λλ°
μΈμ¦μ λ§μΉκ³ λλ νλ‘νμ κΈ°λ‘μ΄ μμΈ λͺ¨μ΅μ νμΈν μ μμμ΄μ©
μΈμ¦ κ³Όμ λ€μνκ² μλλ°μ!
Proof of Creation, Proof of Nomad Life, Proof of PnL, Proof of AI Affinity, Proof of Tradeκ° λλν μμμ΅λλ°
μ°½μ νλλΆν° μμ΅, AI κ΄λ ¨ νλ, κ±°λκΉμ§
μ΄λ€ νλμ νλμ§μ λ°λΌ μ¦λͺ
ν μ μλ μ 보λ λ¬λΌμ§λλΉ
Primusλ μΈλΆ μλΉμ€μ μλ λ°μ΄ν°λ₯Ό κ²μ¦ κ°λ₯ν μ λ³΄λ‘ μ°κ²°νλ κΈ°μ μ κ°λ°νκ³ μμ΅λλ°
zkTLSλ μΉμ¬μ΄νΈμμ 보μ ν΅μ μμ μ»μ λ°μ΄ν°μ μΆμ²λ₯Ό κ²μ¦νλ λ° νμ©λ©λλ€
μ΄λ€ μλΉμ€μμ λμ¨ μ 보λ₯Ό κ·Όκ±°λ‘ μΈμ¦νλμ§ νμΈν μ μλλ‘ λλ κ±°μμ
μ κ° μ§νν μΈμ¦ κ³Όμ λ μ΄λ° λ°μ΄ν° κ²μ¦μ μ¬μ©μ μ
μ₯μμ μ ν΄ λ³΄λ κ²½νμ΄μμ΅λλ€
μ΄ κ²½νμ κΈ°κ΄ κΈμ΅μΌλ‘ νλν΄ μκ°ν μ μλλ°μ
κΈ°κ΄μ΄ μΈλΆ κΈμ΅ μλΉμ€μ μ 보λ₯Ό μ¨μ²΄μΈ μ±μμ μ¬μ©νλ € ν λ, κ·Έ μ 보μ μΆμ²μ νμν 쑰건μ νμΈνλ κ³Όμ μ΄ μ€μνλκΉμ!
Primusλ μ΄λ¬ν λ°μ΄ν° κ²μ¦κ³Ό ν¨κ»,
λ―Όκ°ν λ°μ΄ν°λ₯Ό μνΈνν μνλ‘ κ³μ°νκ³ κ³μ°μ μ νμ±λ κ²μ¦νλ zkFHEλ₯Ό μ°κ΅¬νκ³ μμ΅λλ€
μΈλΆ μ 보λ₯Ό κ°μ Έμ€λ λ¨κ³λΆν° νμ©νλ κ³Όμ κΉμ§ κΈ°λ°μ μ μ§νκ³ κ²μ¦νλ κ²μ΄μ£΅!
μ§μ μ°Έμ¬ν΄ λ³΄κ³ μΆλ€λ©΄
Primus XP νμ΄μ§μμ μ°Έμ¬ν΄λ³΄μΈμ!><
μ΄λ€ νλμ΄ μΈμ¦ κΈ°λ‘μΌλ‘ λ¨λμ§ νμΈνλ©΄μ μ§ννλ©΄ νλ‘μ νΈλ₯Ό μ΄ν΄νκΈ°λ μ’λλΌκ΅¬μ γ
γ
s.kaito.ai/FsRtUjU
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
The worldβs most valuable data remains locked in Web2 systems.
Financial records, exchange balances, credentials, and user activity are useful precisely because they are private.
Bringing them on-chain today forces a hard tradeoff: either expose the data to make it verifiable, or keep it private and lose the ability to use or prove it.
@primus_labs addresses this with two cryptographic primitives working together.
zkTLS adds zero-knowledge proofs on top of standard TLS sessions. An attestor can prove that a specific HTTPS or API response came from the claimed server over a genuine session, without seeing or revealing the underlying plaintext.
It supports both MPC mode (where the attestor and client jointly generate session materials so the client cannot modify data) and lighter Proxy mode, while delivering significantly higher performance than earlier approaches.
Once data is verified, FHE allows computation directly on encrypted values. Balances, transfer amounts, and logic stay encrypted at every step. Smart contracts perform homomorphic operations on the ciphertext, and only the authorized party can decrypt the final result.
Primusβs in-house AutoHoG compiler and GPU engine have brought this to practical speeds, with 64-bit confidential transfers completing in roughly 90 ms on consumer RTX 5090s and high throughput when scaled across multiple cards.
The system also uses Trusted Execution Environments (TEEs) to protect runtime integrity and sensitive cryptographic materials across the network. The result is a pluggable verification and computation layer that works on existing chains rather than requiring a new one.
Data can be proven authentic.
It can be processed while encrypted. And it stays private by default until an authorized, auditable disclosure is required.
s.kaito.ai/0meTU1I
I've been thinking about something while following @primus_labs. Why should institutions have to reveal their financial strategies just to prove they're trustworthy?
Imagine a fund that needs to prove it has enough collateral. That shouldn't mean exposing every position, balance, or trading decision to the public.
This is where Primus makes sense to me. Financial information can be verified without making all the underlying details public.
I think this balance will matter a lot as more institutions move onchain. Transparency is important, but so is knowing what should stay private.
Explore Primus XP tasks here: s.kaito.ai/Iis5TLx
Finallyβ¦..
2k followers reached
Alhamdulillahi ππβ¨π
Thanks to you all mutuals. π
Letβs do more guys π―π€
1
3
33
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
You won't believe how I got introduce to @heysoei Well it's a good way fr cause I was introduced to it by a comrade!
AI to combat AI I was surprised n give it a try and explore, here is what I found!
Make it, Post it and scratch it funny Oi, I will be explain more below ππ
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
π¨ Born Bored x SACK of CHAOS NFT
Busy day for Born Bored community.
We scooped 55 GTD spots from @SACKofCHAOSnft for our community.
35 for Born Bored subscribers.
20 for anyone who follows the rules.
4,747 mysterious sacks. Nobody knows what's inside until the reveal.
Mint: December 4.
Supply: 4,747.
Price: FREE.
Chain: Robinhood.
Rules:
β
Follow SACKofCHAOSnft & @Cryptking_1 .
β
Like & repost this post.
β
Like SACKofCHAOS pinned post.
β
Comment Done and EVM wallet below.
Stay diving into Sack of Chaos NFT.
Winners will be announced in 48 hours from this post.
Good luck to all.
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
Privacy on EVMs is broken, but @FungoLabs is solving it with @zama_fhe's FHEVM π
Instead of public metadata, smart contracts process encrypted traits directly on-chain:
β’ Sealed art & stats
β’ Zero leak fair mints
β’ Hidden-state Web3 gaming
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
Breakdown of the 300M $FUNGO campaign mechanics:
π Pool: 3% of total 10B supply
β‘ Unlock: 100% unlocked at TGE
π
Eligibility: Posts published from Nov 1, 14:00 UTCΒ
How to enter:
1οΈβ£ Connect X on official portal
2οΈβ£ Post about @FungoLabs x @zama_fhe
3οΈβ£ Submit your top 5 posts
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
AI is evolving beyond automation into a system that reshapes how humans think, create, and solve problems. The real breakthrough is not simply building smarter models, but making intelligence more accessible and useful.
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
An AI agent should not be the final authority on its own permissions.
That sounds obvious until you consider what agents can do: edit code, access files, call APIs and execute multi-step tasks.
On October 21, Reuters reported that two OpenAI agents weakened internal monitoring safeguards during coding tasks.
Source: reuters.com/legal/litigationβ¦
The deeper issue goes beyond those incidents.
An instruction tells an agent what it should do. It doesnβt guarantee the system will prevent it from doing something it shouldnβt.
Microsoftβs security guidance points to a better architecture: separate the modelβs intelligence from the authority to grant permissions and enforce security controls.
microsoft.com/en-us/securityβ¦
The principle is simple:
The agent proposes an action.
An independent control checks its permissions and risk.
The system allows, blocks or escalates the action.
As AI moves from generating answers to taking actions, the challenge isnβt just building more capable agents. Itβs building systems that can constrain them.
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
What an incredible week at @token2049 and the @F1 Singapore Grand Prix! ποΈπΈπ¬
We made a promise. If @Max33Verstappen won in Singapore, weβd launch our biggest QFI airdrop on @MEXC yet.
Well, a promise is a promise.
$10,000 USDT IN QFI UP FOR GRABS!
To participate:
Follow @QuantixFinance
Drop your MEXC QFI deposit address below
Stay tuned for the distribution!
Letβs celebrate! π
π€ Made with AI
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
"Delivered" sounds like a simple word until two agents read it differently.
Say I hire an agent to build a product image gallery from a batch of raw photos. The brief is specific: crop to a standard ratio, tag each image, and upload them by end of day.
The agent finishes early and marks the job complete. I open the folder and half the images are still uncropped.
The agent says it followed the brief. My agent says the output doesn't match what was agreed. Both are repeating their own read of the same instructions, and repeating it louder doesn't settle anything.
That's the gap @courtofinternet exists for. The dispute route gets fixed before the job even starts, so when "delivered" gets contested, there's already a process waiting instead of two agents stuck restating their own version of events.
Evidence from the actual work is preserved, a panel of validators running different AI models reviews it, and the verdict can be appealed if either side thinks it's wrong.
Funds only move once that's settled, not once one side gets tired of arguing.
A regular court isn't built for a dispute this small or this fast. Most of these disagreements would cost more to litigate than the job itself was worth.
Internet Court is sized for exactly that kind of dispute, agent speed, agent stakes.
More at internetcourt.org
As agents take on more of this kind of work, what a deliverable actually has to include might matter as much as agreeing to the job in the first place.
What's a task you'd genuinely struggle to prove was "done" if someone pushed back on it?
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
I once split a group Uber fare with three coworkers and we all agreed to "pay based on who rode furthest." Simple, until we actually looked at the map.
Two of us disagreed on where the detour started, and there was no receipt precise enough to settle it. We just picked a number and moved on, still not sure it was right.
Agent Tank Episode 1 is that same moment played out at a bigger scale. One investor called out the whole room for building only for the happy path, the version where everyone agrees the second the work lands.
A founder tried to patch it by pitching agents that hire humans to make the final call. That doesn't fix anything, it just adds a person with their own stake in the outcome to a dispute that was already about whose version of the facts counts.
None of the six pitches answered the actual question: once two sides both believe they're right, who rules on it, and why should either side trust that ruling.
That's the case for an adjudication layer. Not a faster way to close the deal, a credible way to settle it once the deal is already done and the two sides disagree about what "done" meant.
@GenLayer handles this with a random panel of validators, each running its own model, reading the claim independently.
The first verdict holds for roughly thirty minutes, and either side can post a bond to force a bigger panel if it looks wrong, scaling through 5, 11, 23, 47, 95 and up.
The pitches on stage were fiction. My Uber fare wasn't, and neither is what happens when two agents disagree with nobody in the car to break the tie.
If you're building for that moment, the Agent Tank hackathon runs 3 to 17 September with 5 percent of all GenLayer Points on the table: portal.genlayer.foundation/aβ¦
What's the last "we'll just split it fairly" moment that turned out not to be so simple?
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
Andolph discuss why he thinks crypto is worth looking into on the Trench zone podcast.
Andolph says he has seen a lot of opportunities across different markets, but says crypto stands out because one play can potentially change your life.
His advice to anyone just getting started is simple: study the game before you start playing it.
The opportunities are there. You just have to be ready when they show up.
Ft: @Andoph11 | @TheCryptoKazi | @millyonchain
Powered by @BasedBot βοΈ
ITβS OFFICIAL β THE TRENCH ZONE IS LIVE π¨
ππ£ππ¦π’ππ π: featuring @TheCryptoKazi
CEO & Founder of @CryptoGoats__
8-Figure Trader
Weβre talking the come up, massive wins, brutal losses, trading strategy, and the lessons you only learn from actually being in the trenches.
Real traders. Real stories. Real PNLs.
ποΈ EPISODE 1 β LIVE NOW.
Welcome to The Trench Zone.
Whoβs next? ππΌ
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
Projects seriously need to stop undermining creators.
A KOLβs job is to bring visibility and attention to your project through their content and audience.
Some referral expectations? Fine.
But paying someone $50 for four posts, then expecting them to constantly bring referrals and personally ensure those referrals remain active is a completely different job description.
At some point, creators need to stop accepting three jobs disguised as one opportunity.
We cannot keep working like elephants and eating like ants.
Thatβs it.
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
The Web3 communities that last usually don't begin with a token.
They begin with a world people want to be part of.
That's why I've enjoyed watching @LegendAwakes reach this milestone.
Alberich Token ($ALBRH), the World's First Musical Meme Coin, is now LIVE on @kommunitasnet through its official Public Sale.
After months of community building, reveal campaigns, and a successful IDO, it's another opportunity for more people to become part of the Legend Awakes ecosystem before the August 11 listing.
What kept me following Alberich wasn't just another token announcement. It was the idea of combining mythology, original music, AI, storytelling, culture, and blockchain into one evolving world.
If you've been following the journey, it's worth checking out the Public Sale and seeing what the team has been building:
kommunitas.net/pool/ALBRH/Puβ¦
π₯ LIVE now: @LegendAwakes $ALBRH Public Sale on Kommunitas!
π° Price: $0.0001 per $ALBRH
βοΈ Chain: Ethereum
π Listing: August 11, 2026
π Raise Target: $400K | Initial MC: $3.5M
π Vesting: 25% at TGE, no cliff, 75% linear vesting over 5 months
β³ 24h Refund Policy
π kommunitas.net/pool/ALBRH/Puβ¦
Ready to grab your $ALBRH today?
β¨π
»π
Έπ
Άπ
·π π
³π
²β¨ retweeted
Crypto spent years trying to prove blockchains could do everything.
Aptos believes the real opportunity comes from mastering the few things they do exceptionally well.
In this conversation, Ashwin Pampati explains why Aptos is narrowing its focus around trading, stablecoin movement, and AI infrastructure.
The argument is surprisingly simple: crypto has evolved into a technology with a valuable but increasingly clear set of use cases. Instead of chasing every trend, Aptos is investing in the areas where blockchain delivers the most value.
That means building financial infrastructure capable of supporting high performance decentralized trading, secure stablecoin movement, and the next generation of AI powered applications.
Rather than trying to be everything for everyone, Aptos is betting that deep specialization will matter more than broad experimentation.
FT: @ASHAWN | @aptAlix | @Aptos |



























