People using TikTok, Twitter, or Uber have no idea they’re all built on TLS. In fact, most people don’t care what TLS is at all.
Just build useful apps with zkTLS, and you’ll never have to tell everyone what zkTLS is. Forget all that tech shit—maybe nobody cares.
$ZKP is now live on @Revolut.
The native token of @zkPass is now available through one of Europe's leading financial platforms, expanding access to $ZKP beyond traditional crypto exchanges.
Another step toward bringing verifiable data infrastructure to a broader global audience.
Asking Claude to review your code is a slot machine.
Every review finds brand new bugs. Review again: more. Review a tenth time: it still finds something.
It never runs out of bugs to find... so how do you decide your code is actually done?
ZK가 크게 터질 곳은 온체인 안이 아니라 온체인 밖일 수도 있다.
은행, AI 에이전트, 인터넷 서비스가 원하는 건 더 많은 데이터를 쌓는 게 아니라, 덜 받아도 믿을 수 있는 방법이다.
온체인 거래를 증명하는 것보다 인터넷 전체의 ‘확인 방식’을 바꾸는 쪽이 훨씬 큰 시장이다.
Agents might turn companies into markets.
today, your org chart decides which tools you use.
tomorrow, an agent won't care who built what. it'll just pick whatever gets the job done fastest and cheapest.
That's a much bigger shift than better internal tooling.
software inside a company will have to earn its traffic.
Most companies would move faster if internal platforms had to win their internal customers the same way they win external ones.
Publish your uptime, latency, and cost. Take feature requests in the open. And if you're a real product, it's fine to charge internal teams like any other customer.
Agents make this practical. They can discover services, compare them, and route around the bad ones.
Today, @Deviumvip, built with @zkPass, joins the 2026 Global Digital Economy Industry Development Summit in Ha Long Bay, Vietnam 🇻🇳
Devium is tackling a key challenge in the AI era: how to make real-world Web2 data usable and verifiable without exposing the underlying sensitive information.
By turning private data into cryptographic proofs, @zkPass helps projects like Devium build more trusted data flows for AI and digital applications.
As AI depends on increasingly high-quality data, verifiable data is becoming a core part of the infrastructure.
Excited to see the Devium team bring this vision to the global stage.
Proof, not exposure.
Every https://t.co/qXoyhMdGoN audit starts with one URL from @8004_scan and ends with a public evidence log.
We fetch the agent card, check the ERC-8004 registry entry, and confirm the identity fields. If the card is not reachable the run stops there, and the report says so.
Then a model reads the card and writes a test plan for that specific agent. A trading agent gets order book questions. A community agent gets community questions. Nothing generic.
ERC-8004 uses a separate agent registry on each chain.
https://t.co/qXoyhMdGoN now brings those registries together across multiple networks in one discovery layer.
Find agents, inspect their profiles, and review reputation and validation records without searching chain by chain.
Before one agent calls another, it needs more than a name and a list of capabilities.
https://t.co/0UtiMOGNbH brings creator information, activity, user feedback and available audit records into a profile that can be inspected before the call.
🤖Explore the Agent Trust Network:
Thoughts on the Market Cycle
In a world where OpenAI & Anthropic were blitzing from 0% to 1% of GDP by revenue, investing in consensus made sense because there were no incumbents prepared to build AI native infra, services, or apps & the opportunity set for skilled founders with domain expertise was obvious.
The obvious ideas in obviously large markets in the AI trade (especially at the infra layer) have mostly been competed away, yet VC exuberance for the n-th company in these competitive spaces continue unabated and companies have started to price beyond where they would be valued if they executed perfection.
This makes me think that the early stage market has overheated & I don’t see any mediating force that will calm animal spirits down to sensible levels short of either a sharp market correction or a ChatGPT moment at the intersection of AI x Real World (robotics, life sciences, etc), so I have to be more judicious in how I invest Fund II than I was in fund I.
As a result, in my opinion, the “game on the field” for early stage investors is now shifting back from “Access being the Alpha” & all fundable founders/ideas being consensus and known, to originating strong founders early, working with teams over months to flesh out & validate ideas, and being a craftsman instead of a high velocity heat seeking missile.
Tactically, this is very exciting for me because this means that I can go back to doing venture in the style I practiced before I started my fund and had to be in every “hot” company regardless of price. I can flex my muscle as an elite picker of founders & work with them to bring a company to life while, most importantly, getting more ownership in the companies for the work I’m doing.
Of course, there will continue to be many “hot” deals that I’d be remiss to not invest in, so I will do that too on a discretionary basis, but I finally feel like I can slow my velocity down just a little (42 investments in 14 months from Fund I)!
The Growth market continues to be extremely access dependent. I believe that segment of the market will further continue to consolidate around a handful of winners and the firms that have the AUM & the Access to back them round after round will return the largest sums of capital at the highest IRRs. No changes here, just on my knees praying to Goldman Sachs, Thrive, and Dario that they get OpenAI and Anthropic public so the party continues.
Investing through this part of the cycle primarily taught me that venture can no longer afford its best outcomes to be $1b or $10b, but need to be mid 11 figures before they can be meaningful to most large AUM funds.
This isn’t a bear post by any means as much as a recalibration in what I find interesting. In a time where the benefits of robotics, machine autonomy & intelligence, life sciences/bio, reindustrialization, transportation, and so many more sectors have barely begun to seep into the real world and the market has not priced in the productivity and lifestyle gains we will see from them in the next few years, I believe there’s a lot to be optimistic about and for me to invest in.
AI agents will outnumber people in the economy in the years to come.
When do you predict the Agentic Finance flippening to happen? (Defined as agent-to-agent payment volume exceeds human-to-human volume)
zkPass began by making facts from private data verifiable without revealing the data itself.
Now we’re extending that principle to AI agents.
Introducing https://t.co/0UtiMOGNbH, where provenance, audit evidence and observed behavior form a public, inspectable trust record.
This “IP theft” framing will shoot Anthropic in the foot. Hollywood, publishers, authors etc now have the same IP grounds to sue Anthropic for IP theft. How dare you open the IP genie box? No AI labs should.
The dominant narrative in AI is that superhuman AI hackers are going to be a security nightmare.
I now believe the opposite: everything gets more secure after superhuman LLM hackers, because the commoditization of intelligence favors defenders over attackers.
Here's why:
• Before LLMs, defense meant covering every idea that any attacker might come up with. Human minds are diverse and attackers are uncorrelated with each other, so attack ideas had high entropy. N attackers meant N independent darts thrown in idea-space. You could outspend all attackers combined, but if even one gets lucky with and thinks of something you didn't, it's gg.
• AI collapses that entropy. Attackers and defenders alike now use the same models, which means their ideas are correlated. Even if they're using different models, models are more correlated with each other than humans are. 20 uncoordinated attackers are no longer 20 independent minds: they're running the same search 20 times, burning redundant compute, trying the same attack ideas.
• So the game reduces to spend. If you assume entropy is literally 0, the defender only needs to spend as much as the best-funded attacker, not the sum of all attackers. IOTW, attackers used to sum, now they max.
• Example: say you have North Korea, Russia, and China all trying to hack you with LLMs, and they're each spending ~$100K of compute trying to attack you ($300K total), you only need to spend ~$200K to ensure nobody gets through. This is because they're not coordinating. In the previous model, they didn't need to coordinate because the entropy in their different attack ideas were large, and thus you'd need to match their collective budgets to stay secure.
This model is obviously a simplification, and it gets compounded by other techniques that favor defenders, like formal verification and better testing.
But you can already see this playing out in crypto. Big, well-defended protocols have become more secure after superhuman AI has emerged (as I demonstrate in the QT). I believe this is the reason why. It's the same reason why AI slop in writing is so easy to spot, and I think it means the long-term cybersecurity equilibrium is actually a happy one.
A startup doesn't have to be your life's work, but if it is, you become much harder to defeat. You play the long game and you outlast everyone. Because practically everyone else is an opportunist by comparison.