Brett Adcock says Figure AI deployments are imminent, staking humanoid race on intelligence
The Figure AI CEO says multiple customer deployments will be announced within 90 days as Figure 03 shifts from demos to production work.
By Ryan Merket · Published
Why it matters
Founders and operators care less about humanoid hype and more about unit economics. If Figure AI can show robots doing paid work with limited human babysitting, it validates a software-first edge and signals real demand. If not, the narrative shifts back to capital scale and hardware logistics.
Brett Adcock (@brettadcock) says Figure AI will announce multiple customer deployments in the next 90 days and argues that intelligence, not manufacturing scale, will decide the humanoid race. The update adds that Figure 03 is moving from live demos to production environments doing real economic work.
Founder first
Adcock founded Figure AI in 2022 and serves as CEO. Before robots, he co-founded the recruiting marketplace Vettery and later co-founded electric air-taxi maker Archer Aviation, experiences that put him at the center of two different build-at-scale companies. He has since launched projects including Cover (2023) and Hark (2025), per his Wikipedia entry. The through line: ship hard tech into commercial settings.
From demos to production
The Aligned News post says Figure 03 is transitioning from stage demos to production settings focused on real economic work. No customer names, industries, or locations were included in that post, and there are no disclosed numbers on how many robots, how long deployments run, or whether these are pilots versus long-term contracts. But the near-term promise is clear: move beyond slick videos and show robots producing measurable economic value for real operators.
That is a different kind of milestone for any robotics company. Running in production means tolerances, uptime, and safety have to hold under messy conditions. It also means product and go-to-market need to mature together: integration, support, and change management on the customer floor become as important as grasp planning and motion control.
Intelligence over manufacturing scale
Adcock’s framing, as relayed by Aligned News, is that the winner in humanoids will be decided by robot intelligence more than by sheer hardware volume. In practice, that places the bet on software: perception, planning, dexterity, and learning systems that can generalize across tasks and environments with minimal reprogramming. It is a familiar split in the field: scale matters when you have something worth scaling, but getting to consistent autonomy on unstructured work remains the hard part.
The claim also sets expectations for what Figure AI will showcase over the next 90 days. If Figure AI is right, the upcoming deployments will be evaluated less on how many robots are shipped and more on what those robots can reliably do, how quickly they adapt, and whether they deliver unit economics that make sense without armies of human babysitters.
What to watch
- Customer clarity: Which industries and workflows Figure 03 tackles first, and whether those are pilots or revenue contracts.
- Capability breadth: Are tasks narrowly scripted or does the system handle variation with minimal human intervention?
- Operational evidence: Safety, uptime, and productivity data that show robots doing repeatable work in production.
Figure AI has been explicit about wanting to turn humanoids into general-purpose labor that fits into existing facilities. If Adcock and team can turn the next 90 days into proof that Figure 03 is more than a demo reel, the intelligence-over-scale thesis gets its first real trial where it counts: on the factory or warehouse floor.