Flapping Airplanes Raises One Billion for Leaner AI Models

A year-old AI research lab just convinced investors to hand it a billion dollars to build leaner AI models. Flapping Airplanes announced on Wednesday that it had closed a Series A round co-led by Index Ventures and Lux Capital, reportedly valuing the company at six billion dollars despite having no commercial product on the market yet. The bet is one of the largest early-stage AI investments announced this year, and it signals that Silicon Valley's money is shifting from simply building bigger data centers toward making the software inside them radically more efficient.

The round arrives just nine months after the lab launched with a one-hundred-eighty-million-dollar seed round that valued it at one and a half billion dollars. In other words, its valuation roughly quadrupled in under a year. Earlier backers Sequoia and GV returned for the new round, and they were joined by Kleiner Perkins, Valor, Atreides and the chipmaker Nvidia, among others, according to TechCrunch's reporting on the announcement.

What exactly is a billion dollars buying? The lab's central argument is that AI progress has become too dependent on ever-larger datasets and computing clusters. Instead of assuming the next leap requires more data centers, its researchers are reworking the algorithms, the training methods and the low-level software that runs on graphics chips. If that sounds abstract, the pitch gets concrete fast: the founders describe their work as a megakernel framework, a single program that can run an entire training job, serve live answers to users and even play a video game at the same time.

The founders behind the bet

The team is as much of the story as the technology. Flapping Airplanes was founded in late 2025 by Aidan Smith and brothers Asher and Ben Spector. Smith is a former founder and Thiel fellow who worked at the brain-implant company Neuralink while still in college, according to TechCrunch. Asher Spector holds a doctorate in statistics from Stanford, while Ben Spector was a computer science doctoral student at Stanford who founded Prod, a student-run startup accelerator, during his time at MIT.

That youth is part of the pitch. Sequoia partner David Cahn, an investor in the company, has described the founders as building the "young person's AGI lab," a line that captures how the round is being sold: not as a bet on a product, but as a bet on a generation of researchers who grew up inside the AI boom and think the current playbook is wasteful. Index Ventures confirmed the financing, and one of its partners is joining the company's board, which suggests the lead investors want a close view of where the research goes next.

The megakernel idea is the lab's signature technical claim. A kernel is a program that runs on a graphics chip, and a megakernel bundles many tasks into one program. The founders say their framework can perform entire training runs, serve inference and play video games, and even do all three simultaneously in a single kernel. Lead investor Index says this removes wasted time between steps so each chip does more useful work. If it holds up at scale, the same hardware could train bigger models for less money, which would ripple through every company renting that hardware.

Why leaner AI models matter to everyone else

For readers outside the lab, the stakes are practical. Training today's leading models costs staggering sums and consumes enormous amounts of electricity, and those costs eventually show up in subscription prices, slower free tiers and the environmental footprint of every chatbot query. Models that learn from less data could make powerful AI cheaper to run, faster to improve and less dependent on whoever owns the biggest cluster of chips. That is the promise, at least.

The counterpoint is that there is no product yet and no public benchmark proving the approach works at the scale the valuation implies. A six-billion-dollar price tag for a research lab with no revenue is a bet on a technical breakthrough that has not been demonstrated outside the company's own claims. The AI funding market has rewarded that kind of bet before, and it has also burned investors when ambitious research did not translate into working systems.

Context helps here. The same week brought news of a large legal-AI raise, with Legora pulling in five hundred and fifty million dollars at a valuation north of eight billion, according to industry coverage. Investors are financing companies they believe can control valuable parts of the AI economy even when the path to profitability is uncertain. Flapping Airplanes sits at the infrastructure end of that spectrum: if its efficiency research works, it becomes a toll road for everyone training models; if it does not, it is an expensive science project.

There is also a generational angle worth noting. The founders are in their twenties, funded by people betting that the next big AI ideas will come from researchers who never had to unlearn the old ones. Whether or not the megakernel delivers, the round is a reminder of where the industry's confidence currently sits: less data, less waste, more clever software. For anyone watching AI from the sidelines, that shift matters more than any single funding number. Readers curious about how younger users are already leaning on AI can see our story on Gen Z trusting AI holiday planning more than Google, and this week's other big tech money story is the SpaceX-Tesla merger talks.

Sources: TechCrunch, TechStartups