Bitget CEO’s AI Crypto Bet Puts Capital in the Plumbing

Bitget CEO Gracy Chen’s AI crypto argument is that artificial intelligence (AI, software that can generate, analyze or act on data) will add users, trading tools and tokenized assets to blockchain markets instead of draining capital away from them. The test is whether exchanges can turn AI from a theme trade into working infrastructure before regulation and liquidity shift again.

Her CNBC comments landed one week after the Senate Banking Committee advanced the CLARITY Act, giving the exchange business a second reason to talk about expansion. AI brings the user interface. Regulation decides where that interface can legally touch customer money.

The Capital Question Chen Is Answering

The capital fight looks lopsided at first glance. In the latest Bitget April transparency report, the company pitched a single trading environment that spans tokenized equities, AI-native trading infrastructure and onchain financial services. That is not a small product tweak. It is an attempt to make crypto exchanges look less like coin venues and more like multi-asset financial apps.

Venture capital (VC, private investment in young companies) data explains why Chen needed to answer the crowding-out question. AI has become the fastest way to attract capital, talent and boardroom attention. Crypto still raises money, but the flow has moved toward exchanges, stablecoins, payments and tokenization rather than the broad Web3 rush of the last cycle.

• $267.2 billion in U.S. quarterly venture deal value was recorded in Q1, according to the PitchBook-NVCA Venture Monitor.
• 88.8% of Q1 VC deal value went to AI and machine learning companies in that same report.
• More than $20 billion went into crypto and blockchain startups during 2025, Galaxy Research said.

The better read is that AI is setting the funding temperature for all tech. Crypto founders now have to show how their rails, custody, payments, trading or tokenization products become useful to AI-driven users.

AI Money Looks Huge Because It Is Concentrated

The Q1 PitchBook-NVCA venture monitor shows a market where the headline number is swollen by a handful of giant rounds. The top five deals accounted for $195.6 billion of U.S. venture investment, and AI companies made up 42.5% of all completed deals. That is a funding boom, but it is also a concentration warning.

Crypto’s numbers tell a different story. Galaxy’s crypto and blockchain venture capital report put Q4 investment at $8.5 billion across 425 deals, with trading, exchanges, investing and lending pulling the largest capital share. That is where Chen’s comment has bite. AI is pulling the oxygen toward itself, but crypto’s investable categories are shifting toward infrastructure that AI could use.

Bitget Is Selling an Operating Model

Bitget’s own product trail shows why Chen framed AI as additive. The company launched Gracy AI in February and described it as a digital human built around Chen’s market thinking. Its February transparency report on AI and UEX growth also pointed to Agent Hub, tokenized traditional assets and market-data access for automated tools.

Universal Exchange (UEX, Bitget’s term for one venue covering crypto, tokenized stocks and onchain services) is the phrase the company wants investors to remember. The idea is simple enough: if users already come for Bitcoin or perpetual futures, the venue can add stocks, yield tools, wallet functions and AI guidance without sending them elsewhere.

That is why the exchange-as-operating-system bet matters more than a chatbot. It turns AI into a retention layer. It also makes tokenization more useful, because a user who can ask, compare and act in the same app has less reason to treat blockchain assets as a separate account. Oton Technology covered the same retail-access tension when Republic began moving Animoca Brands equity onto Solana, a sign that private and public asset wrappers are getting pulled toward wallet-based distribution.

• AI can translate market data into prompts a retail user can understand.
• Tokenized assets can give exchanges more inventory beyond spot coins.
• Agent tools can make 24-hour markets feel less manual.
• Regulatory clarity can decide whether those products stay offshore or move into the U.S.

The CLARITY Act Would Move the Bottleneck

The regulatory half of Chen’s CNBC appearance may matter more for Bitget than the AI half. Senate Banking Committee Chairman Tim Scott, Republican of South Carolina, convened a May 14 markup of H.R. 3633, the Digital Asset Market Clarity Act of 2025, and said the bill was meant to set rules for digital assets. Senator Mike Crapo, Republican of Idaho, said the committee advanced it in a 15-9 committee vote.

The bill’s core promise is jurisdiction. The Senate Banking Committee’s CLARITY Act fact sheet says it would draw a line between the Securities and Exchange Commission (SEC, the U.S. securities regulator) and the Commodity Futures Trading Commission (CFTC, the U.S. derivatives and commodities regulator). For exchanges, that line matters because licensing, disclosures, custody and surveillance requirements shape where products can launch.

Clarity also cuts both ways. A U.S. rulebook can make expansion easier for compliant platforms, but it can expose weak token listings, sloppy market-making and fake volume. Oton Technology’s earlier report on the FBI’s fake NexFundAI token sting showed how an AI label can be used as bait in crypto markets. If AI becomes part of the exchange interface, surveillance and suitability questions follow it into the app.

The Onchain Agent Trade Still Has Friction

The strongest version of Chen’s thesis is not that AI tells people which coin to buy. It is that software agents start using blockchain rails because those rails are open, programmable and always on. Galaxy Research made a similar point in its research on AI agents and blockchain friction, noting that agentic systems have to replace human judgment with machine-native equivalents.

That sounds clean until money moves. An AI agent needs permission to spend, reliable data to decide, a safe way to sign transactions and a way to recover when execution fails. Decentralized finance (DeFi, blockchain-based trading and lending without a central intermediary) was built for composability, but many interfaces still assume a human is reading warnings and clicking approvals.

Discovery Friction

Agents need structured data about which pools, venues and tokens are safe enough to touch.

Control Friction

Wallets must let software act with limits instead of giving it unlimited signing power.

Execution Friction

Blockchain trades can fail, slip or be attacked if routing and timing are poor.

This is where exchanges have an opening. A centralized venue can package custody, data, execution and risk controls behind a simpler interface. But that convenience is also the trust trade. The more an AI layer does for the user, the more the platform has to prove that it is not steering users toward its own inventory.

Exchanges Win If the User Changes First

Chen’s comment carries optimism with a condition. AI helps crypto if it changes user behavior before it becomes just another speculative label. A trader who asks an assistant to compare collateral, fees and tokenized stock exposure is more valuable to an exchange than a trader chasing one AI coin. A fund that uses agents to rebalance onchain cash has a stronger reason to care about settlement quality than social buzz.

The unresolved question is timing. If CLARITY reaches the Senate floor and survives reconciliation with the House bill, U.S. expansion becomes a legal strategy rather than a guessing game. If it stalls, Bitget and its rivals can still build AI tools abroad, but the biggest capital market remains harder to serve from the front door.

Disclaimer: This article is for informational purposes only and does not provide investment, legal or tax advice. Crypto assets, tokenized securities and AI-assisted trading tools carry financial and operational risks. Readers should consult qualified professionals before making decisions. Figures are accurate as of publication.