The AI pivot playbook is now being used to revive struggling public companies. How Smartbird performs this year will show whether the pattern has any business substance behind it.
- Allbirds sold its shoe business for $43 million, raised $100 million on the stock market, renamed itself Smartbird, and is now positioning as an AI infrastructure company.
- Nadia Carlsten, a former AWS executive with an engineering PhD, joined as CEO with no team, no product, and no disclosed customers.
- Smartbird targets data sovereignty compute for regulated industries including pharma, finance, energy, and government.
- Established competitors Hewlett Packard and Equinix already operate in the managed AI compute space Smartbird is entering.
- Carlsten expects compute clusters running for several customers by the end of 2026.
What Happened
Allbirds sold its shoe business for $43 million, raised $100 million from the stock market, renamed itself Smartbird, and hired its first CEO. The company now has one employee and an AI infrastructure ambition.
Smartbird is positioning as a managed AI infrastructure provider for companies that need direct control over their servers and data sovereignty. Its target customers include firms in pharma, finance, energy, and government that run bespoke AI models and cannot rely on public cloud infrastructure.
Nadia Carlsten, a former AWS executive with an engineering PhD, joined as CEO on June 18. She most recently led DCAI, a European compute company where she worked with clients including Novo Nordisk.
Carlsten is actively recruiting a leadership team and plans to open an office, starting with a hire to lead infrastructure operations. She expects compute clusters running for several customers by the end of 2026, she told TechCrunch from Amsterdam.
Why It Matters
The Smartbird story is the AI hype cycle applied to a public company balance sheet. Allbirds used retail investor enthusiasm to raise $100 million for a business model that established players like Hewlett Packard and Equinix have been running for years, at a time when AI compute demand is reshaping the entire infrastructure market. That the pivot worked in the stock market says more about investor appetite for AI exposure than about Smartbird’s actual competitive position.
Carlsten acknowledged she couldn’t estimate the size of the addressable market and described it as still nascent. With no team, no product, and no disclosed customers, Smartbird is effectively a pre-revenue startup sitting on a large cash position. The data sovereignty niche is real, but the distance between today and a paying customer roster is still wide. Relve, an AI trends intelligence platform, is tracking this segment as data sovereignty compute becomes a more active vendor decision for SaaS teams.
“There are some companies out there chasing AI, but at the end of the day, what matters is, is there actual weight behind the chasing?” Nadia Carlsten, CEO, Smartbird
Bottom Line
Watch whether Carlsten can close paying customers by year-end. With $100 million in the bank and a CEO who has worked with regulated-industry clients before, the resources are there. Failing to show a customer roster by early 2027 would confirm the pivot was a stock market move, not a business one.
For SaaS founders evaluating managed AI compute, this segment is real but Smartbird is not a vendor option today. HP, Equinix, and established neoclouds are the current candidates. Check back when Smartbird has a product and a team behind it.
