One Developer, One Lobster Logo, and a $1 Trillion Question

Six months ago, Peter Steinberger was an under-the-radar Austrian developer tinkering with a lobster-themed AI coding project. This week, the CEO of the world's most valuable company stood on stage and called his creation "the most popular open-source project in the history of humanity."

That's not a typo. Jensen Huang of Nvidia dedicated a major portion of his GTC keynote to OpenClaw — an agentic AI platform that barely existed half a year ago. On the sidelines of the event in Santa Clara, he told CNBC's Jim Cramer: "This is definitely the next ChatGPT." He added that OpenClaw "exceeded what Linux did in 30 years" in mere weeks.

The hyperbole is worth pausing on. Because underneath it sits a genuinely uncomfortable question for anyone with money in AI stocks: if an anonymous developer can build the next big thing in a matter of months, what exactly are investors paying $1 trillion for?

What OpenClaw Actually Does — And Why It Matters

At its core, OpenClaw is a platform for building and managing AI agents — software that acts on your behalf rather than just answering your questions. Think of it as the difference between a search engine and an assistant who actually does the searching, negotiating, and buying for you.

Huang used a telling example in his keynote: an OpenClaw agent that scouts eBay for deals and places bids automatically. But the platform's real reach is broader. Developers and hobbyists are using it to deploy AI agents across WhatsApp, Telegram, Slack, Discord, and Signal — running everything from their personal computers at home, not from expensive cloud servers.

That last detail is critical. When a developer runs OpenClaw on a Mac Mini they bought for a few hundred dollars, they're not paying OpenAI or Anthropic per API call. They're using open-weight Chinese AI models that are, by most practical measures, good enough — and dramatically cheaper. The economics of AI are shifting in real time.

The Crack in the Foundation Model Thesis

OpenAI and Anthropic have raised money at a combined private valuation exceeding $1 trillion. The implicit bet behind those numbers is that their models are so superior that developers and enterprises will keep paying premium prices to access them. OpenClaw's rise is stress-testing that assumption.

"It solidified the open-source community and proved that fully autonomous AI can be run at home without relying on the Magnificent 7 or Big AI," said David Hendrickson, CEO of consulting firm GenerAIte Solutions. "I suspect this was the black swan moment most big AI companies feared."

Forrester analyst Charlie Dai framed it structurally: "As foundation models rapidly commoditize, attention is moving toward agent frameworks that emphasize autonomy, usability, locality, and control." The model is the engine. The agent framework is the car. When engines become interchangeable, car design starts to matter more.

Sponsored

Advertise with Us

[email protected]

Sponsored

Nvidia read this shift clearly. At GTC, the chipmaker announced NemoClaw — a free security layer built specifically to make OpenClaw safe enough for enterprise adoption. Nvidia doesn't sell software; it sells the chips that run all of this. The more AI agents proliferate, the more compute gets consumed. OpenClaw's success, whatever form it takes, is good for Nvidia's hardware business.

OpenAI Hired the Creator. But Doesn't Own the Code.

Sam Altman moved fast. Last month, on a Sunday, he posted on X that Steinberger was joining OpenAI and that OpenClaw would "live in a foundation as an open-source project that OpenAI will continue to support." Altman called him "a genius with a lot of amazing ideas" who would "help drive the next generation of personal agents."

It's a savvy move — and an admission. By bringing Steinberger inside, OpenAI gains his talent and, arguably, some narrative control. But the open-source nature of OpenClaw means the technology itself belongs to no one. Any developer can fork it, modify it, build on it. Anthropic is already debuting similar features, including a new channels tool. Gavriel Cohen, an Israeli developer, built his own hardened variant called NanoClaw after finding OpenClaw's security too loose for business use — and then released that to the open-source community too.

Cohen's story is a microcosm of the whole phenomenon. He imagined using OpenClaw to run AI agents for his marketing agency across WhatsApp and Slack. Then he discovered the software couldn't distinguish his personal WhatsApp groups from his work chats. The nightmare scenario: a colleague asks if he's free for an afternoon meeting, and the agent replies that he has to take his daughter to ballet — because it's reading his personal messages. "I can't rely on this, and I don't feel safe with it," Cohen told CNBC. "It's not responsible to connect my customer data to it."

After spending days fixing it with help from Anthropic's Claude Code, Cohen released NanoClaw publicly. His wife started using an AI agent named Andy to track baby stroller prices on WhatsApp. Cohen's estimate of its commercial value: "a SaaS product that you would maybe spend $10 a month on." He's since shut down his marketing firm, launched a startup called NanoCo, and partnered with container company Docker.

Not Everyone Is Convinced the Giants Are Losing

Jerry Chen of Greylock — an Anthropic investor — pushes back on the commoditization narrative. "The buzz around OpenClaw stems from making AI more tangible to a broader audience," he said, "but it doesn't take away from the importance of the underlying foundation models, which remain more powerful than open-weight alternatives." His framing: OpenClaw expands the market; it doesn't necessarily cannibalize the premium tier.

The open question Chen raises is the right one: "Does OpenClaw become the de facto standard — the Linux of the market, as Jensen puts it — or just the first of many open and closed-source agentic operating systems?"

Linux is instructive here. It democratized server computing, devastated proprietary Unix vendors, and yet Microsoft, Oracle, and others adapted and survived. The analogy isn't perfectly comforting for OpenAI and Anthropic, though — those Unix vendors had hardware and enterprise relationships to fall back on. AI labs have models and brand. If the models commoditize, the brand has to carry more weight than it ever has.

The Skeptic Who Changed His Mind (A Little)

Jay Goldberg of Seaport Research Partners is the lone analyst among roughly 70 tracked by FactSet with a sell rating on Nvidia. He initiated that call in April after the stock had already surged, and the shares have climbed more than 60% since. He's been wrong on the price, but his underlying question — "what's the point of all this AI, there are no consumer use cases" — has been consistent.

After buying a Mac Mini and spending time with OpenClaw, Goldberg says he finally gets it. As a parent of three, he gets about 10 emails a week he dreads reading — school notices, picture day reminders, early pickup requests. He can imagine an agent handling all of it. "It's not just the functionality of the thing itself, but it's the pieces of our lives that we give it access to," he said.

He's not upgrading his Nvidia rating. But he called Huang's Linux analogy "nailed it" — and admitted he's watching TikTok videos about OpenClaw trying to understand it better before he trusts it with his life. His current assessment of the software: "janky, incredibly insecure, and my Mac Mini is kind of half working." His assessment of the potential: "very easy to see how this can become really powerful and really useful."

That gap — between what it is now and what it could become — is exactly where the investment question lives.