Why We Need AI Regulation—Now

The call came just after dinner. A woman picked up, and on the other end was her daughter’s voice—crying, terrified, saying she had been kidnapped. But her daughter hadn’t gone anywhere. She was safe in her room. The voice on the phone was a fake—an AI-generated deepfake so realistic, even her mother couldn’t tell the difference.

When I read that story, I felt a chill down my spine. That’s not science fiction. That’s today’s reality.

We’ve crossed into a new era where artificial intelligence can mimic human voices, generate fake photos that fool millions, and write convincing lies in seconds. And right now, we have almost no rules stopping any of it.

That should scare you.

AI has exploded into every part of our lives. You might’ve used it today—asking your phone for directions, chatting with a virtual assistant, seeing AI-generated art or ads online. I use it too. It’s fast, helpful, and—let’s admit it—impressive. But beneath the cool tech lies something we haven’t figured out: how to make sure it doesn’t destroy more than it creates.

Because right now, AI can do real harm. It can be used to scam elderly people out of their savings by faking a loved one’s voice. It can flood the internet with deepfake videos that spark panic or smear someone’s reputation. It can automate hiring, policing, and lending decisions in ways that quietly reinforce racism, sexism, and class bias.

And if that weren’t enough, it’s coming for our jobs too. Goldman Sachs estimates AI could automate the equivalent of 300 million jobs. Think about that. Millions of people could find their roles replaced by algorithms trained on their own work.

So why aren’t we doing more to regulate it?

Some countries are trying. The European Union passed a bold new law that bans certain dangerous uses of AI outright—like real-time facial recognition in public spaces—and forces companies to prove their systems are safe before they’re deployed. That’s smart. It treats AI like what it is: powerful technology that needs oversight, just like planes, medicine, or nuclear power.

In the United States, the government has finally woken up. Last year, President Biden signed an executive order requiring the developers of the most powerful AI systems to share safety test results with the government. It also called for new standards to watermark deepfakes and protect people’s privacy. It’s a good start—but we need actual legislation, and we need it fast.

Because while lawmakers debate, tech companies are racing ahead. Some AI models are now so powerful they can create content so realistic, even experts have a hard time telling real from fake. I’ve seen photos of Pope Francis in a puffer jacket, Elon Musk holding hands with robots, and war scenes that never happened—images shared thousands of times before anyone realized they were fake. Imagine that level of deception during an election, or in the middle of a war.

And these systems aren’t just visual. They write too. AI can now write articles, product reviews, even legal documents. But it doesn’t always get the facts right. I’ve seen AI confidently cite fake studies, invent quotes, and make up laws. In the wrong hands—or even in the hands of someone careless—that kind of power can mislead millions.

Worse, AI doesn’t just reflect the world. It can reinforce its ugliest parts. We’ve seen algorithms in hiring that penalize women and people of colour. Facial recognition software that misidentifies Black and brown people at far higher rates. Predictive policing tools that double down on over-policing in marginalized neighborhoods. These aren’t bugs. They’re baked in. And without transparency, we don’t even know how many people have been unfairly denied a job, a loan, or justice because of biased AI.

You’d think, with all that risk, regulation would already be in place. But it’s not.

Right now, most AI tools are being released without safety checks, ethics reviews, or clear accountability. There are no universal standards saying what’s fair, what’s safe, or who’s liable when things go wrong. If an AI tool ruins your credit score or sends you a fake medical alert, good luck finding someone responsible.

I’ve heard people say we should wait and see. That regulation might stifle innovation. But history tells a different story. When we waited to regulate social media, we got disinformation, mental health crises, and the erosion of public trust. When we failed to control early internet data collection, we ended up with a surveillance economy we can barely escape. Waiting only makes the damage harder to fix.

This time, we have a chance to get ahead of the problem.

That means setting real rules—now. We need transparency: people should know when they’re dealing with AI and when content is artificially generated. We need accountability: if an AI system causes harm, someone needs to answer for it. We need fairness: no system should be allowed to discriminate or operate in secrecy. And we need to protect privacy like never before.

We also need a strong global framework. AI doesn’t respect borders. A deepfake made in one country can go viral in another in seconds. We can’t afford a fractured system where rules differ wildly from place to place. We need coordination—between governments, industries, and civil society—so that the benefits of AI are shared, and the harms are prevented.

To be clear: I’m not anti-AI. I’m anti-chaos. I believe AI can help us cure diseases, tackle climate change, and unlock human creativity in ways we’ve never seen. But only if we shape it to serve people—not profit alone.

Right now, that window is open. The choices we make today will decide whether AI becomes a force for progress or a weapon for harm. We can’t sit back and hope tech companies do the right thing. We have to demand laws that reflect our values, and we have to do it while there’s still time.

Because once AI rewrites the rules, we may not get to rewrite them back.