Figure AI Leads Humanoid Commercial Deployments While Tesla Optimus Stays Internal
Industry observers contrast Figure AI paying customer deployments at BMW and logistics sites with Tesla still installing Optimus production lines and targeting internal academy use before external sales.
Humanoid robotics entered a strange phase in 2026: social media sees walking demos daily, yet enterprise buyers still ask a boring question. Who has robots doing paid work outside the lab?
An August 25, 2026 Benzinga interview with RoboStrategy CEO Andrew Kang sharpened that divide. Kang argued Figure AI holds an early U.S. commercial lead while Tesla Optimus remains pre commercial despite enormous attention.
Figure side of the scoreboard
Figure has publicized deployments with BMW Group Plant Spartanburg and Catalyst Brands logistics operations, plus long horizon autonomy demonstrations such as extended package sorting runs powered by its Helix vision language action stack. Production rate increases and customer named pilots give buyers something contracts can attach to.
Kang told Benzinga that Figure is the only U.S. company he tracks that has crossed the 1,000 robot production milestone in the humanoid category. Milestones alone do not guarantee long term wins, but they matter for supply chain learning and field reliability data.
Tesla side of the scoreboard
Tesla narrative centers on scale economics copied from automotive manufacturing. After ending Model S and Model X lines at Fremont, the company is converting space for Optimus assembly. JPMorgan factory tour notes cited in industry coverage describe tarped production areas and a phased rollout: Optimus Academy training first, internal factory tasks next, external sales potentially in the second half of 2027.
On Tesla August 5, 2026 earnings call, leadership pointed to late July or August 2026 for production start, without publishing external customer counts or verified operating hour tables comparable to Figure public claims.
Two strategies, one market
The contrast is strategic, not moral. Figure is selling software maturity and enterprise proof points now. Tesla is betting that automotive grade mass production and vertical integration will overwhelm early movers once hardware design finalizes.
Investors should treat production milestones and paying deployments as different filters. The former shows manufacturing intent. The latter shows someone signed a purchase order.
Risks ahead
Humanoid economics remain fragile. Unit costs, safety certification, and task specific reliability can erase headline autonomy clips. Both companies must convert pilots into repeatable workflows, not one off livestreams.
Warehouse operators will also compare humanoids against specialized arms that already work today. General purpose robots must justify flexibility premiums with measurable throughput.
Sources
Benzinga interview with RoboStrategy CEO Andrew Kang (August 25, 2026)<br />Figure AI public deployment announcements at BMW and Catalyst Brands (June 2026)<br />Industry coverage of Tesla Fremont Optimus line conversion and JPMorgan tour (August 2026)