Allbirds Founder's AI Biz: A Vision, No Team

The announcement in April that Allbirds, the direct-to-consumer footwear brand known for its distinctive Silicon Valley-style sneakers, was pivoting to artificial intelligence initially struck many as a move straight out of a satirical tech comedy. It represented a dramatic shift from its core identity.

This strategic pivot mirrored a familiar pattern seen in "meme stock" phenomena, where a struggling publicly traded company embraces a popular trend to attract retail investors and subsequently boost its stock valuation.

Remarkably, the strategy proved successful. The company divested its shoe manufacturing operations for $43 million, secured an additional $100 million in capital from the stock market, and has since rebranded itself as Smartbird.

Now, the formidable task of establishing Smartbird as a leader in its new domain falls to Nadia Carlsten. A distinguished former AWS executive with a Ph.D. in engineering, Carlsten previously headed the European compute firm DCAI before assuming her role as Smartbird’s CEO yesterday.

Speaking to TechCrunch from Amsterdam, Carlsten outlined her immediate priorities: “We’re going to be recruiting a brand new team for the AI business, and we’re going to be getting an office. The shoe business has officially closed as of yesterday, so that’s all done…The first task that I’m tackling right now is rounding up the leadership team, looking for somebody to lead infrastructure operations, for example.”

Essentially, Smartbird can be viewed as a startup, led by a singular founder, but commencing operations with a substantial seed funding round. The exact trajectory beyond these initial steps remains to be fully defined.

Smartbird's ambition is to become a vital AI infrastructure provider, capitalizing on the seemingly insatiable demand for computational power required to train and deploy deep learning models. However, its approach diverges from that of "neoclouds," which focus on arbitraging chip prices against GPU time or inference tokens. Carlsten intends to target more precisely managed deployments. Smartbird’s ideal clientele comprises organizations that require direct oversight of the servers hosting their models—often due to specific political or business model requirements—and prioritize data sovereignty over the scalable nature of public cloud offerings.

Carlsten acknowledged the nascent stage of this particular market, making it challenging to estimate its current size, especially as many companies are still in the pilot phase for AI tools. Drawing on her experience at DCAI, where she collaborated with European entities like Novo Nordisk that emphasize data sovereignty or utilize bespoke models, she noted, “we certainly have anybody that’s within the pharmaceutical industry, energy industry, financial, the public sector.”

From Carlsten’s perspective, Smartbird will not be in direct competition with hyperscalers or general neocloud providers. Instead, its primary rivals will be internal company projects. Nevertheless, established players already offer similar services, such as Hewlett Packard with its single-tenant managed AI compute service, and data center behemoth Equinix.

While Smartbird presents a viable business model, its growth potential compared to the expansive, hyper-growth cloud services market is yet to be determined. Carlsten anticipates having compute clusters deployed for several customers by the close of the year. In contrast, other startups, like the inference cloud General Compute, have announced significantly larger ambitions, including a reported $300 billion chip order upon emerging from stealth last month.

Carlsten maintains that Smartbird does not require massive chip commitments to realize its vision. Her potential customers typically need hundreds to thousands of chips, emphasizing that it’s “not about large scales and huge numbers of GPUs, they’re more about agility of these clusters, and more about having control of the infrastructure stack.”

Furthermore, Smartbird is unlikely to compete on price. Traditional cloud services rigorously optimize chip utilization around the clock to offer the most cost-effective compute. However, Carlsten believes that companies with highly specialized workflows will achieve greater efficiency by operating their own dedicated servers.

The surging demand for AI infrastructure is a dominant market force, propelling stock prices for chipmakers, cloud providers, and energy companies, and even making seemingly futuristic concepts like orbital data centers appear plausible to investors. Despite this fervor, Carlsten firmly asserts that Allbirds' transformation was the result of meticulous planning.

“It wasn’t, ‘Let’s just do AI, because it’s AI, and it’s hot,’” stated Carlsten, who commands an annual salary of $700,000 and received approximately $9 million in stock for taking on the role. “It was really about, do we have a chance to build a business over time that is going to find this niche in the market and be able to grow over time?”

A notable consequence of Allbirds’ pivot was the relinquishment of its public benefit corporation (PBC) status. This designation was originally intended to solidify the shoe company’s sustainability pledges. PBC charters are commonly adopted by companies to underscore non-financial commitments; for example, OpenAI operates as a PBC focused on AI safety. This change in Smartbird's status, however, suggests that PBC charters may not be as immutable as often perceived.

Carlsten affirmed that Smartbird’s board has made a long-term commitment to the execution of her AI strategy.

“There are some companies out there chasing AI,” she conveyed to TechCrunch, “but at the end of the day, what matters is, is there actual weight behind the chasing?”