Humanoid Robots Face a Scaling Reality Check: Demand, Not Manufacturing, Is the Real Bottleneck

The Hype vs. Reality of Humanoid Robot Scale

Humanoid robotics companies have raised hundreds of millions of dollars at billion-dollar valuations by promising to transform work. Tesla plans 50,000 Optimus robots in 2026. Figure targets 100,000 by 2029. Morgan Stanley predicts a $5 trillion market with over 1 billion units by 2050.

But according to Melonee Wise, former chief product officer at Agility Robotics, the industry faces a fundamental challenge: nobody has found an application that requires thousands of robots per facility.

Manufacturing Isn't the Problem

Building tens of thousands of humanoid robots is physically feasible. About 500,000 industrial robots were installed worldwide in 2023. Since a humanoid is roughly equivalent to four industrial arms in component terms, existing supply chains can handle even optimistic production projections.

"Simply building the robots is arguably the easiest part of scaling humanoids," says Wise. "The bigger problem is demand."

The Real Bottlenecks

1. Battery Life

Agility's Digit robot can run for 90 minutes with a 9-minute recharge (10:1 ratio). In practice, it charges after every 30 minutes of work, keeping 60 minutes as reserve. Without this buffer, robots could run out of power mid-task in logistics and manufacturing environments.

"No one wants to deal with" several hundred robots weighing over 100 kg each running out of power mid-task, Wise notes.

2. Reliability

A factory at 99% reliability experiences ~5 hours of downtime per month. Industrial customers expect 99.99% because production line downtime can cost tens of thousands of dollars per minute. Agility has demonstrated this in specific applications but not in multipurpose scenarios.

3. Safety

Unlike autonomous vehicles and drones, which benefited from immature regulatory environments, humanoid robots operate in already heavily regulated industrial settings. Compliance with existing industrial safety standards is mandatory.

4. AI Capability

"I think what a lot of people are hoping for is they're going to AI their way out of this. But the reality of the situation is that currently AI is not robust enough to meet the requirements of the market."

Two Paths Forward

• Volume approach: Deploy thousands of robots doing one job (requires a killer application that doesn't exist yet)
• Versatility approach: Deploy hundreds of robots that can each do 10+ jobs (most companies are betting on this, but AI isn't ready)

The Uncomfortable Truth

The market for humanoid robots is "almost entirely hypothetical." Even the most successful companies have deployed only small numbers in carefully controlled pilot projects. Future projections are based on an "extraordinarily broad interpretation" of what a capable humanoid robot — which doesn't currently exist — might be able to do.

As Wise puts it: onboarding any new client takes weeks or months. Large deployments are the most realistic path to scale, but large deployments require finding jobs where thousands of robots make economic sense. That application hasn't been found yet.