Paradigm breaks with the past and raises $1.5 billion to bet on AI and robotics

News hit the venture capital market like a bombshell: Paradigm, one of the world’s largest cryptocurrency-focused funds, announced on February 28, 2026, that it is raising a new fund of up to $1.5 billion. The difference? The focus is no longer solely on cryptocurrencies. Artificial intelligence, robotics, and cutting-edge technologies now share the stage. This expansion is no coincidence. It’s a response to an arithmetic problem that has been developing for years.

The numbers dilemma: when money grows but projects dwindle

Start with simple math. Paradigm manages $12.7 billion in assets. It sounds impressive. But look at its fund trajectory: in 2021, it raised $2.5 billion — a record for a crypto fund. Three years later, in 2024, it managed only $850 million. The decline doesn’t reflect a lack of fundraising capacity. It reveals something quite different: the cryptocurrency market doesn’t offer enough opportunities to absorb such high volumes, maintaining expected returns.

Sector numbers in 2025 tell this story. Total global VC investment in cryptocurrencies reached $49.8 billion — a record volume. But here’s the twist: the number of VC transactions dropped about 60%, from roughly 2,900 to just 1,200. Money is increasing while opportunities decrease. Funds are now concentrating capital in a few mega-rounds instead of distributing among hundreds of early-stage startups.

For small and medium funds, this might be manageable. For Paradigm, it’s a structural problem. With a $12.7 billion portfolio to allocate, it becomes increasingly challenging to find large enough early-stage projects capable of absorbing this volume while maintaining realistic return expectations.

The shadow of FTX: when research wasn’t enough

To understand the new $1.5 billion fund, we need to go back to November 2022. That’s when FTX — the empire of founder Sam Bankman-Fried — collapsed in a matter of days, causing havoc across numerous institutions. Paradigm had recorded an accounting investment of $278 million. It all turned to zero.

For an elite institution reputed for being “research-driven” and technologically sophisticated, this was not just a bad credit. It was a public judgment error that required explanations — to LPs (limited partners), the market, and itself.

What happened afterward seemed strange at the time. In 2023, observers noticed a silent change on Paradigm’s official website: all mentions of “crypto” and “Web3” were removed, replaced by a more neutral phrase: “investment in technology.” No formal announcement of this rebranding was made. The community discovered and questioned everything.

The alarm among LPs was real: Was Paradigm leaving the crypto market? Matt Huang, co-founder, had to appear publicly. He stated that Paradigm “has never been as excited about crypto as now,” but added a crucial detail: “The development in AI is so remarkable that it cannot be ignored. Framing AI and crypto as zero-sum competition is a popular narrative but incorrect. Both are very interesting, and there will be significant overlap.”

At that moment, it was clearly a strategic defense. But, viewed today, it reveals that internally a forced question was already being answered: what to bet on in the next decade?

Matt Huang: two years positioning at the intersection

If you only look at today’s official announcement, Paradigm’s transformation seems sudden. But Matt Huang’s actions over the past two years tell a completely different story.

In 2024, Paradigm invested $50 million in Nous Research, an AI infrastructure focused on research and development of large open-source language models. It wasn’t exploratory pilot funding. Fifty million signaled a structural commitment.

A few months later, Paradigm and OpenAI jointly launched EVMbench — a testing tool to evaluate the ability of different AI models to detect and fix vulnerabilities in smart contracts. Here, the core cryptocurrency infrastructure met AI capability assessment, putting both topics on the same table.

Simultaneously, Matt Huang was building another company: Tempo. A stablecoin payment infrastructure that aligns perfectly with his role as a board member of Stripe. In 2025, Stripe formed a strategic partnership with Paradigm and launched its own stablecoin payment product.

Connect these dots: Matt Huang isn’t planning to “enter AI.” He has been living at the intersection of AI and crypto for at least two years. He didn’t bet on AI or crypto in isolation, but at the very moment these two waves will collide.

Why AI×Crypto and not just a transition to pure AI?

Here’s the common mistake in the market narrative. Paradigm entering AI doesn’t mean competing directly with a16z or Sequoia for the same projects. If it were just that, it would have no advantage: the overall AI space is already saturated with traditional VC giants with deeper funds and stronger resources.

Paradigm’s real logic is different: it’s not competing for the overall AI pie but for the still-not-fully-perceived intersection.

AI agents are one of the hottest concepts today. These autonomous intelligent agents capable of executing tasks independently have already begun replacing humans: in searches, programming, data analysis, administrative processes. But they face a fundamental problem that no one has fully solved yet: money.

When an AI agent needs to make a payment, receive funds, or transfer money between services, what does it use? PayPal? Conventional bank accounts? These systems were designed for humans, requiring manual authentication and authorization, incompatible with autonomous machine execution.

But stablecoins can. Smart contracts can. Programmable money can. That’s the insight Matt Huang had: these two lines — autonomous AI and crypto payment infrastructure — will inevitably converge. When that happens, those positioned on both sides will capture exponential returns.

The story LPs need to hear now

There’s a pragmatic dimension that cannot be ignored. Paradigm’s LPs — institutions and individuals who entrusted their capital — saw the $2.5 billion raise in 2021 shrink to $850 million in 2024. The vast difference needs a convincing narrative.

Saying “we will continue investing in early crypto projects” is already a familiar story to justify a $1.5 billion goal in 2026. But “we will leverage the technological advantages of the crypto ecosystem to enter the field of advanced technologies in the hottest era of AI and robotics” — that’s a narrative that works.

In 2025, 61% of all global VC flow went into AI, totaling approximately $258.7 billion. This is the largest current reservoir in venture capital. Paradigm’s new fund represents a smart move: drawing water from this overflowing lake rather than remaining stuck in a shrinking one.

For LPs, it’s a bigger story and a more credible growth logic that justifies a new capital commitment.

The eloquent silence of Matt Huang

Return to 2023. When Matt Huang was forced to clarify the website rebranding, he said a phrase that at the time sounded like a tactical defense: “AI and crypto are not a zero-sum competition.”

Today, that phrase reads as an early announcement. Paradigm took three years to recover from the ashes of FTX. Instead of taking the easier path — scaling down, focusing on crypto, waiting for the next bull cycle — it chose a more challenging path with much greater expansion potential: betting on the collision of AI and crypto, building a presence in both universes simultaneously.

Matt Huang has not yet publicly responded to today’s Wall Street Journal report. But his Tempo is still under construction, Nous Research continues operating, and EVMbench has already been launched.

He no longer needs words. These actions already communicate everything.