At a critical moment when the world is being propelled into a new round of technological revolution driven by AI, an internet dominated by machine roles is quietly forming. In the future, billions of AI Agents will perform tasks, complete transactions, and make decisions in the digital world, becoming new 'digital work entities.' To ensure that these autonomously interacting machines are secure, compliant, and controllable, a unified identity, payment, and governance infrastructure is almost an indispensable underlying capability.
Kite AI was born in such a wave of the times.
This startup targets the underlying trust framework of the 'machine internet' and plans to build the world's first AI-driven payment blockchain network, securing strategic investments from top institutions such as Paypal Ventures, Coinbase Ventures, and General Catalyst.
In the latest episode of Founder’s Talk in (The Round Trip), Kite AI co-founder and CEO Chi Zhang shared her entrepreneurial path, her judgment on the future of the agent internet, and how Kite AI aims to seize the opportunity in this imminent explosive ecosystem.
The era of the machine internet is coming.
Chi Zhang has been deeply involved in the AI and data fields for many years, from a PhD at Berkeley, to the automated machine learning platform DotData, and then to product management of data engineering at Databricks, she has always been at the forefront where data and intelligent systems intersect.
She stated that when the number of AI Agents is about to surpass that of humans, a huge structural shift has already occurred:
"The future internet is not designed for humans, but for machine roles."
To enable these Agents to truly operate independently and reliably, they must possess three core capabilities:
Identity: Machines need to be able to prove 'who I am'.
Payment: Machines must be able to complete payments and settlements instantly and securely.
Governance and restrictions: All tasks must be carried out under clear, programmable rules to avoid Agent behavior going out of control.
Kite AI's positioning is to build a programmable trust infrastructure that can simultaneously support the above three.
Why now? Why Agent?
Despite the rapid development of AI models and computing power in recent years, Chi believes the real 'tipping point' for the industry comes from the explosion of Agent technology that started in the second half of last year.
From Manus's general Agent, to OpenAI's ChatGPT Operator, to various systems that can complete complex tasks without a lot of instructions—Agent capabilities are rapidly enhancing.
This means:
Machines are beginning to make automated decisions.
There will be trading demands between machines.
Machines will call external API data in large numbers.
Machines will replace humans to complete a large number of workflows.
However, the existing internet architecture is designed for human vision, experience, and processes.
For machines, it does not need a login interface, nor does it need to input credit card CVV, let alone a beautiful front end.
Therefore, it is necessary to reconstruct a **machine-native** infrastructure.
Kite AI's position: not a protocol, but a 'settlement and verification layer'.
In the current Agent ecosystem, open standards such as x402, A2A, and AP2 are emerging, with participants like Visa, Google, and Coinbase driving forward. These standards are similar to ERC-20 and ERC-721, and are the 'language' of Agent interaction.
And Kite AI's role is:
"We are not a language; we are the underlying blockchain that executes all languages."
In other words:
How these standards are settled, how they are verified, and how they truly occur on-chain or off-chain will be carried by the foundation of Kite AI.
This makes Kite the most foundational yet crucial part of the entire machine payment ecosystem.
High frequency, real-time, machine speed: pressures that traditional blockchains cannot bear.
Payments between machines are not at human speed, but at 'machine speed'.
High frequency
Low latency
Real-time settlement
Massive concurrency
Strong demand for privacy and trust
Whether it is Ethereum, Solana, or any existing public chain, it is difficult to meet the throughput and cost requirements of this level.
Therefore, during the system design phase, the Kite team was very clear:
They need to build a chain that is completely different in structure.
In the process of seeking solutions, they saw Brevis's zero-knowledge proof technology:
Can support large-scale identity verification.
Provides scalable verification mechanisms.
In high-frequency scenarios, it can provide capabilities similar to state channels.
This has made Kite's identity system 'ZK Passport' possible, and also gave the machine Agent's payment path the capability of being both efficient and privacy-preserving.
Why them? The 'scarcity cognition' brought by deep combinatorial perspectives.
Chi mentioned that, despite many relevant technologies already being available, there are still very few people who can truly piece together 'Blockchain + Identity + Payment + AI Agent' into a reasonable structure.
This is because this issue not only requires:
Understanding of AI models, data, and Agents
Experience in payment and risk control
Deep mastery of blockchain architecture
It also requires a rare 'cross-domain perspective' to design a system that can truly scale.
"At first, there were a thousand architectural methods, but only five paths may remain that can withstand huge traffic in the future."
Kite's important work is to be on those few 'scalable paths' from the earliest stage.
The power behind the funds: the real value of strategic investors.
Kite has just completed a $33 million financing, and the importance of this financing lies not only in the funds themselves.
Paypal, Coinbase, SBI, Japanese institutions, and investors such as GC have brought:
Channel resources
Compliance and payment experience
Enterprise cooperation network
Global market coverage
These are key to accelerating Kite AI's entry into real scenarios such as e-commerce agency, enterprise procurement agency, and cross-border payments.
For example, in the e-commerce field:
In the future, AI Agents will help users automatically purchase daily necessities, plane tickets, and hotels, or assist businesses in automatically procuring raw materials and parts—this itself is a market at a billion-dollar level, and traditional payment systems are not suitable for such automated, high-frequency, cross-border machine transactions.
Conclusion: The infrastructure of the machine internet is taking shape.
The era of AI Agents has just begun, and for these Agents to become true 'digital labor', they must have a trustworthy, instant, and verifiable underlying system.
Kite AI is attempting to become this foundation.
In the context of an era where identity, payment, and governance elements are concentrated in a programmable trust layer, Kite's direction is not only clear but also highly forward-looking—it is not building a single product, but laying the foundation for the future machine internet.
