Crypto Firms Demand Access to AI Tools – But Who Will Monitor the Monitors?

There they are, the big players in the crypto world—Coinbase, Kraken, BitGo, and others—gently but firmly tapping the shoulders of AI developers like OpenAI or Meta. “Hey, we’d really need access to your super-AIs. No shady purposes, we promise—just to protect ourselves from scammers and hackers.” It almost sounds like a cry for help—or a risky game of poker.
In an open letter, they’ve banded together to demand exactly that: access to the same cutting-edge AI models that cybercriminals or state-backed hackers could misuse. Their argument? Current security measures for these “frontier models” block legitimate applications—like defending blockchain networks from attacks—while malicious actors keep moving freely. “The good guys shouldn’t be at a disadvantage just because the bad guys can also wield these tools,” they argue.
The Big Question: Who Actually Monitors the Monitors?
The irony? A crypto industry that constantly battles hackers, fraud, and market manipulation now wants access to these powerful AI tools. “Give us the tools so we can defend ourselves better!”—it’s almost the classic “Stop, thief!” scenario. Critics shake their heads. “Are we really supposed to hand loaded guns to those who are always in the middle of shootouts?”
Cybercriminals are already using AI to crack smart contracts or craft flawless phishing emails. Meanwhile, companies like Chainalysis deploy AI to track money laundering. The letter’s signatories claim they could use AI to identify legitimate security flaws in blockchain systems and fend off attacks in real time. But who decides who gets access? And by what rules?
A Race Against Their Own Dark Side
This debate comes at a particularly inopportune time. While governments are still debating vague security guidelines,



AI development is racing ahead unchecked. The crypto sector is among the first to be both victims and beneficiaries of this tech—and often feels like the tech world’s stepchild.
Coinbase framed it in a metaphor many can relate to: “Current security models are like a locked gate behind which the enemy moves freely.” At the same time, the companies emphasize they’re not calling for total deregulation—just a “balanced” solution that enables innovation without ignoring public safety.
The Other Side: Why More Control Might Be the Only Way Forward
Not everyone is convinced. Security expert Bruce Schneier compares AI systems to nuclear power: “You can’t just unleash them without risking catastrophic consequences.” The examples he cites are chilling—deepfake phone scams to manipulate DeFi protocols—showing just how dangerous unchecked AI can be.
So, is it wise to give the crypto industry, itself plagued by fraud and manipulation, access to such powerful tools? Or would that be like handing a child a lighter and saying, “Just be careful!”
What’s Next—and Who Makes the Rules
The AI labs remain silent for now, but the debate is in full swing—and it will extend far beyond the crypto world. Should there be a “two-tier AI” system—one for the good, one for the bad? And who even defines who belongs in the first group?
One thing is clear: This discussion will shape not just the tech world, but our collective future. In an era where cyberattacks are growing more sophisticated, the answer could determine whether progress becomes more perilous—or whether we learn to wield these powerful tools responsibly. Crypto firms, at least, aren’t sitting idly by while criminals get better tools. The question is: Who will ultimately control those who control the controllers?