Before the Internet, no one needed antivirus software. That’s a stupidly true statement. The Internet created enormous possibilities. But yeah, it also created the possibility for people to transmit viruses that could infect hundreds of millions of computers around the world. Now we all need antivirus software to keep our computers from getting infected and shutting down on us.

I kept thinking about this when I was reading Anthropic’s registration statement for its initial public offering. The statement promises great things that the company can do for the economy and humanity, such as hugely increasing productivity and developing cures for previously incurable diseases. But it also lists some of the potential harms that its AI can cause.

Among these harms is the fact that it will radically increase the ability of evil doers to hack the websites of companies and governments all around the world. This can totally transform the business of ransomware in the same way that the printing press revolutionized the publishing industry.

Unless Anthropic CEO Dario Amodei and his gang have some preventive plans they have not disclosed (I mean the type of defense, not the specifics), it’s hard to see why we would not have hundreds or thousands of ambitious young men (I’m assuming most will be men) using the latest Anthropic AI to attack all sorts of websites. This could be the quickest way to become the next billionaire in the AI age. After all, there is no shortage of people in Silicon Valley with the moral compass of an Elon Musk or Mark Zuckerberg.

At the very least, this risk will likely force businesses and governments to spend enormous sums on new security measures to defend their systems. Maybe this is where Anthropic and the rest of the industry see the big demand for the AI that they expect to earn them trillions. Defense against malicious AI attacks should also create plenty of jobs, for those concerned that AI will take all the jobs.

It also doesn’t help that the government agencies that are charged with preventing these sorts of assaults, most notably the FBI, have dumped many of their most experienced people. And those who remain seem more involved in trying to compile dirt on Donald Trump’s political enemies than preventing and uncovering real crimes.

And even in the case where a crime is successfully detected and prosecuted, Donald Trump has made it clear that the pardon window is always open. As long as the AI attackers are smart enough to stash a few tens of millions of their “earnings” away in a safe account, they will be able to buy their way to freedom, along with a declaration from the president about how they were treated “very unfairly.”

So, when you hear people being skeptical about what AI can do, tell them they don’t know what they are talking about. The potential is unlimited.

Dean Baker is a senior economist at the Center for Economic and Policy Research and the author of the 2016 book Rigged: How Globalization and the Rules of the Modern Economy Were Structured to Make the Rich Richer. Please consider subscribing to his Substack, from which this is reprinted with permission.