Everyone Sees the Problem. But Is Regulation Really the Answer?

Tristan Harris spent years at Google before he couldn't defend what he was helping to build. He calls it a race to the bottom of the brain stem. Platforms don't compete to give you the best product — they compete to keep you inside the platform as long as possible. The winner is whoever exploits your psychology most efficiently. Because the money isn't in the content. It's in your attention.

We think we're guests at the party. We're the food.

The bill eventually arrives

In 2021 it emerged that Facebook had internal research showing Instagram was damaging the self-image of teenage girls. They chose not to act on it. In 2023, Meta was sued by 41 US states for deliberately engineering its platforms to create addiction. The lawsuits described specific design choices — notification patterns, scroll mechanics, engagement loops — that the company's own researchers had flagged internally.

Then two juries, two days apart, found Meta liable. In New Mexico: roughly $400 million in damages. In Los Angeles: responsible for contributing to depression and suicidal ideation through deliberate platform design. TikTok chose a different path — they settled the day before their own trial was due to start. Not because they were innocent, but because they could afford to make it go away.

And yet the conversation that followed was still: how do we regulate this better?

Knowledge isn't enough when the system is built to defeat it

The strange thing is that we already know all of this. The documentaries have been watched. The articles have been read. And yet we keep scrolling. That's not a character flaw — it's the point. The system is specifically engineered so that awareness isn't sufficient to change behaviour.

An alcoholic can avoid a bar. But you can't opt out of the digital world. Your job is there. Your professional network is there. The news, the events, the memories. Platforms have woven themselves so tightly into daily life that leaving them no longer feels like a choice — it feels like a loss. That's by design.

Europe responds with laws. GDPR, the Digital Services Act, the AI Act. These are important steps, but they don't change what the platforms are structurally built to do. Silicon Valley responds with lobbyists, and adapts just enough to survive the regulation. It's like setting a speed limit on a road that shouldn't exist in the first place.

The group nobody mentions

There's one group that almost never comes up in this debate: the people who actually make the content. Millions of people spend hours every day producing the videos, posts, and images that keep us on platforms — the content that makes advertisers interested, that forms the foundation of why these companies are worth hundreds of billions of dollars. The vast majority earn nothing from it.

When I scroll through my feed today, it reminds me of late-night shopping channels from the 1990s. Everyone is selling something I don't need. It hit a kind of low point for me when a well-known Norwegian influencer appeared in full makeup, kneeling in front of their washing machine, selling detergent. That's what the incentive structure produces: content that serves the sponsor, not the audience.

What happens if we flip the model?

What if money followed the content instead of the algorithm? A model where a creator owns their audience — where sharing something you actually believe in doesn't just enrich the platform but also rewards you and the people who came to see it. That's not just a different business model. It's a different signal to everyone who uses it: this is content people genuinely value, not content you've been manipulated into being unable to look away from.

There's no technical reason this can't be built. There's no regulation that prevents it. The only thing missing is the willingness to choose something different.

We don't need to wait for companies from Silicon Valley or Shenzhen to change. They won't. But we can build something better. The question is whether we're willing to stop regulating someone else's model and start building our own.

Originally published in Norwegian in Shifter.no on April 16, 2026.