A Silicon Valley AI lab just turned a $180 M seed into a $1 B Series A, reshaping how GPUs train models.
Flapping Airplanes, the year-old startup pioneering data-efficient AI, announced a $1 billion Series A led by Index and Lux, catapulting its valuation to $6 billion. The round brings back early backers Sequoia, GV, Kleiner Perkins and adds Nvidia, Valor and Atreides to the cap table.
Funding Round Details
- Total raise: $1 billion Series A
- Lead investors: Index, Lux
- Existing backers returning: Sequoia, GV, Kleiner Perkins
- New strategic investors: Nvidia, Valor, Atreides, A*
The valuation jump from $1.5 billion at seed to $6 billion in nine months signals strong confidence in the company’s technology roadmap.
Founders and Their Pedigree
Flapping Airplanes was founded by Aidan Smith—former Thiel Fellow and Neuralink alumnus—and Stanford brothers Asher and Ben Spector.
- Aidan Smith: ex-Neuralink, Thiel Fellow, serial entrepreneur
- Asher Spector: Statistics PhD, co-author of the megakernel concept
- Ben Spector: Computer science PhD, creator of the secretive student accelerator Prod
"We’re building the young person’s AGI lab," said Sequoia partner David Cahn, highlighting the founders’ ambition to democratize advanced AI.
The Megakernel Promise
The startup’s flagship offering is a "megakernel" that can simultaneously handle training, inference and even real-time gaming within a single GPU kernel. By overlapping memory-bound inference with compute-bound training, the system keeps the chip busy, promising:
- Up to 10× improvement in data efficiency
- Lower training costs, making large models economically viable
- Potential to run sophisticated AI without ingesting massive datasets
Implications for the AI Landscape
If the megakernel lives up to its claims, it could:
- Reduce the carbon footprint of AI research
- Lower barriers for startups lacking massive compute budgets
- Accelerate the deployment of AI in edge devices and consumer products
What’s Next for Flapping Airplanes
The fresh capital will fund aggressive hiring, expand GPU-level engineering, and accelerate product rollouts. Industry watchers expect a beta demo later this year, with pilot programs slated for autonomous robotics and interactive entertainment.