Valve's Steam Antitrust Fight Has Nothing to Do With Robots, But Everything to Do With Platform Power

Valve is getting sued for allegedly abusing its market dominance with Steam, and I know what you're thinking: Bob, what does PC gaming have to do with warehouse automation?

Bear with me. I've got a point.

The lawsuit, filed in both the US and UK, claims Valve's Steam platform (which controls something like 75% of PC game distribution) strong-arms developers into unfair terms. The allegation is familiar territory if you've followed the Apple and Google app store battles. Basically, if you want access to the customers, you play by the platform's rules. Bloomberg has a detailed breakdown, and it's worth reading even if you don't care about games.

Here's why I care: I spent 12 years at Kuka watching platform dynamics reshape industrial automation. And I'll be honest, the parallels are uncomfortable.

When I started in the industry, you bought a robot arm, you bought the controller, you bought the software. Usually all from the same vendor, sure, but you owned the stack. You could hire a third-party integrator. You could swap out components. The robot was a tool, not a subscription.

That's changing. Fast.

Look at what's happening with cloud-connected robotics platforms. I called my old colleague Werner at Siemens last month (he's still there, somehow), and he was telling me about customer negotiations where the conversation isn't about hardware specs anymore. It's about API access, data rights, software licensing tiers. The robot is almost incidental.

The Steam lawsuit matters for robotics because it's testing a legal theory about platform power that could apply anywhere a dominant player controls the ecosystem. Valve's defence, from what I gather, is that Steam provides genuine value and developers can sell elsewhere if they want. Which, sure. But when 75% of your customers are on one platform, "sell elsewhere" isn't really a choice.

Sound familiar? Replace "game developers" with "automation integrators" and "Steam" with (well, I'll let you fill in the names). We're not there yet in robotics. But we're heading that direction.

I remember when we were developing the KR AGILUS at Kuka, back around 2012. The whole pitch was speed and precision for small-part handling. Electronics assembly, that sort of thing. We controlled the hardware, the controller software, the programming environment. It was a vertically integrated product.

Now I look at what companies are building and it's... different. The robot arm is almost a commodity. The value is in the orchestration layer, the fleet management, the AI that decides what the robot should do. And whoever controls that layer controls the customer relationship.

I don't know how the Valve case will shake out. Antitrust law moves slowly, and gaming is different from industrial automation in about a hundred ways. The revenue models are different, the customer relationships are different, the switching costs are different.

But the underlying dynamic, a platform becoming so dominant that it can dictate terms to everyone who depends on it, that's universal. And it's something the robotics industry should be watching.

I talked to a younger engineer at a warehouse automation company last year (I won't name names) who told me their biggest competitive moat wasn't their robots. It was their customer data. "Once they're on our platform for two years," he said, "they can't leave. We know their operations better than they do."

He meant it as a good thing. I found it, in a way, troubling.

The Valve lawsuit claims the company uses its market position to enforce a 30% revenue cut and prevent developers from offering better deals elsewhere. Whether that's illegal is for courts to decide. Whether it's good for innovation is a different question.

In robotics, we're still early enough that competition is real. You can buy arms from Fanuc, Universal Robots, ABB, a dozen Chinese manufacturers. You can run ROS on most of them. Integration is painful but possible.

I just wonder how long that lasts. The economics of software platforms favour consolidation. The company with the best data trains the best AI, which attracts more customers, which generates more data. We've seen this movie before.

Anyway. The Steam lawsuit probably won't set precedent that directly affects robotics. Antitrust cases are narrow like that. But the questions it's raising about platform power, market access, and fair dealing? Those are coming to our industry eventually.

Might be worth paying attention.