I just saw an interesting news: Paradigm is raising a new $1.5 billion fund, expanding investments beyond crypto into AI and advanced technologies. It seems like a major shift in direction, but actually, it’s not.



Looking back at the numbers, in 2025, VC funding worldwide into crypto totaled over $498 billion. Sounds like a lot, but the problem is that the number of projects decreased by nearly 60%, down to just 1,200 from 2,900. A large amount of that capital went into only a few big projects, not spread out evenly. For small funds, this might not be an issue, but for Paradigm, managing $12.7 billion in assets, it’s a structural problem. They’re not lacking projects, but lacking large enough projects in early stages to invest that amount while still maintaining expected returns.

Looking at their history, in 2021, Paradigm raised $2.5 billion, the largest crypto fund ever at that time. But by 2024, it shrank to $850 million. This contraction isn’t a sign of weakness but an adjustment to changing markets. Still, it indicates that even for Paradigm, focusing solely on crypto, it’s become difficult to find suitable opportunities at their scale.

What changed everything was FTX in November 2022. When FTX collapsed, Paradigm lost $278 million. This publicly exposed the reputation of a “research-driven” firm, forcing them to explain to LPs and the market, and also to themselves. In 2023, the community noticed that Paradigm’s website quietly changed; the words “crypto” and “Web3” disappeared, replaced with more neutral “technology investment.” There was concern about what Paradigm was up to. Matt Huang, co-founder, had to clarify, tweeting that Paradigm “has never been more excited about crypto,” adding that “AI and crypto are not zero-sum competition; they have overlapping interests.” At the time, it seemed like a defensive move, but looking back now, it was a warning of what was to come.

If you look at Matt Huang’s actions over the past two years, you’ll see he’s not just “saying” he’ll invest in AI. He’s been “at the intersection of AI and crypto” for two years. In 2024, Paradigm invested $50 million in Nous Research, an AI infrastructure company focused on open-source large language model research. That’s not a small exploratory move but a serious bet. In February this year, Paradigm partnered with OpenAI to launch EVMbench, a tool to evaluate AI models’ ability to detect and fix smart contract vulnerabilities. Crypto infrastructure now faces AI capability assessments on the same platform.

Matt Huang is also founding Tempo, a staking infrastructure company for digital coins, and he’s a board member of Stripe, which partnered with Paradigm in 2025 to launch a stablecoin payment product. Taken together, Matt Huang isn’t betting solely on AI or crypto but on the convergence of both at some point.

The problem is, when an AI agent needs to pay, receive, or transfer money, can it use PayPal or a bank account? Those systems are designed for humans, requiring identity verification, not suitable for automation. But stablecoins, smart contracts, and programmable money systems can do that. That’s why Matt Huang is building Tempo and investing in Nous Research simultaneously. He believes these paths will eventually meet.

Looking back to 2023, Matt Huang’s statement that “AI and crypto are not zero-sum competition” seemed like a community defense at the time. But in the current context, it’s a warning. Paradigm took three years to move past the FTX incident. They didn’t choose the easy route of downsizing and waiting for the next bull market, but instead took a more difficult path with broader potential: betting on the integration of AI and crypto by establishing positions in both markets simultaneously.

For Paradigm’s LPs, it’s a bigger story. In 2025, 61% of global VC funds, about $258.7 billion, flowed into AI—the largest pool. Paradigm’s $1.5 billion raise aims to draw water from this pool, not stay in the shrinking lake. For LPs, this is a more logical growth story.

This $1.5 billion fund is a milestone on this path. Matt Huang has not officially responded to reports, but Tempo is still building Nous Research, and EVMbench has already launched. These actions speak louder than any statement.
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin