Research brief · Robotics · October 2026 · 22+ reports & forecasts

Home robots:
when?

The first consumer humanoid is on sale for your house. In practice, a person in a headset is driving it from an office. We read the market reports, the bank forecasts and the roboticists, and separated what these machines can actually do from what the stage demos promise.

By Christos Papadimitriou · House of 47 Research
Last updated:

See the timeline
$20,000First home humanoid (1X NEO), shipping 2026
TeleopHow today's home humanoid actually works
~2032Consensus for a reliable general home robot
3BHumanoids BofA sees by 2060, most in homes
A humanoid robot against a dark backdrop
The arc · 2002 → 2030s · consumer home robots

Four waves

Every wave of general-purpose home robots has failed on the same three points: price, an unclear job, and novelty that wears off. The only robots that ever reached mass adoption do one dull task, out of sight. The new humanoid wave is betting that AI finally breaks that pattern. It has not yet.

2002 · Roomba

The one that worked

50M+

iRobot puts a robot in homes at scale. Vacuums, then mowers and pool cleaners, win by doing one dull job out of sight. Still the only truly autonomous home machines.

2014 → 2019 · Social bust

The companions died

4 dead

Jibo, Anki, Kuri, later Amazon Astro. Beloved, pricey, and unable to justify themselves against a smart speaker. The novelty wore off before the business worked.

2025 → 2026 · Humanoids

On sale, teleoperated

$20k

1X NEO ships to US homes, Optimus and Figure 03 chase the same dream. But the IFR counted just ~7,000 humanoids sold worldwide in 2025, and the tasks still run through a human operator. Autonomy is marketing.

2030s · Autonomy?

If the hands catch up

~2032

A reliable, unsupervised home robot needs cheap dexterous hands, models that generalize to your mess, and proven safety. First real use likely lands in subsidized eldercare.

Two curves that have to cross

Cost to build a home humanoid

Cross-check · market reports, bank forecasts, roboticists

Where they agree

  • All
    Single-task robots already won

    Robot vacuums are a real mass market, roughly $7B and about one in five US homes. One dull job, done out of sight, is the only formula that has ever worked.

  • All
    General-purpose companions have failed every time

    Jibo, Anki's Cozmo and Vector, Kuri, Astro. Too expensive, no clear job a smart speaker couldn't do, and the novelty faded fast.

  • All
    Today's home humanoid is teleoperated

    The 1X NEO on sale now runs its tasks through a remote human in a headset, not autonomy. This is why evidence beats demos, the gap each model's real skills are scored at Robots Comparison.

  • All
    Dexterous hands are the bottleneck

    Moravec's paradox: lifting is easy, folding a shirt in an unseen kitchen is brutally hard. Hands are both the hardest engineering and roughly half the build cost.

  • Most
    The honest demand is demographic

    Aging populations and care-worker shortages, sharpest in Japan and Korea, are the real pull. Eldercare, not chores, is the first serious market.

  • Most
    Factory wage parity comes first

    BMW reportedly pays about $25/hr for a Figure unit, matching its lowest-paid worker. Work robots pay off before home robots do.

Where they disagree

When does a reliable general home robot arrive?

A machine that does most chores, unsupervised, at a price households pay. Placed by forecast year, 2026 to 2040.

On sale, driven by a humanAutonomous & affordable

How big do the banks think humanoids get?

Headline ambition from each forecast, placed by scale. Figures mix units and revenue, shown by magnitude of the bet.

A nicheBigger than the car industry

Will the winning home robot even be humanoid?

Human shape for human spaces, or specialized machines that abandon the body plan.

Task robots winHumanoids win

1X NEO 2025 ($20k / $499/month, teleop-assisted) · Metaculus general household robot ~2032 (range 2029–2038) · Rodney Brooks 2025 ("at least a decade", likely not humanoid) · Goldman 2024 ($38B/2035) · Morgan Stanley ($5T, ~1B units by 2050) · Citi ($7T, ~650M units by 2050) · Bank of America (3B units by 2060) · Musk (10B+ long term, ~$20k) · IFR World Robotics 2026 (~7,000 humanoids sold in 2025; mass home adoption unlikely near or medium term).

Probability of outcome · House of 47 synthesis

The odds

Synthesis estimates across the forecasts and reports below, scenario estimates, not market forecasts. The through-line: single-task robots keep compounding, humanoids stay supervised for years, and the YouGov read is that only ~38% of Americans even want a household robot.

Capability map · what a robot can actually do in your home

What it can do, and when

The cruel twist: the chores people most want automated, laundry, dishes, tidying, are exactly the dexterous, unstructured tasks that are hardest. The one already solved, floors, is done by a non-humanoid puck.

TaskToday~2030~2035

"Teleop" = works but with a human in the loop. "Demo" = shown in controlled settings, not reliable in a random home. Directional, synthesised from analyst and roboticist forecasts; ~2035 full-autonomy remains contested.

House of 47 read · How to think about it

Buy on evidence, not on stage

Separate the robot from the demo

A curated clip and your actual kitchen are different problems, and the distance between them is the whole story. Score each model on what it has really shown, autonomous versus teleoperated, which is exactly what robotscomparison.com is built to do.

Single-task still wins

The money in the home today is in robots that do one dull job well. The vacuum, the mower, the pool cleaner outsell every humanoid dream, and will keep doing so until autonomy is proven.

Watch the hands and the price

Dexterous hands are the capability chokepoint and roughly half the bill of materials. The curve to watch is build cost falling toward ~$15–20k while reliability climbs. Both have to happen.

Eldercare is the real first market

The honest pull is demographic, not chores. Supervised, subsidized eldercare in aging economies is where a capable home robot lands first, a decade before it folds your laundry unwatched.

Base case

Through the late 2020s, humanoids stay supervised and expensive while single-task robots keep compounding. Treat any "autonomous home helper" claim as teleoperated until a credible, repeated, unsupervised track record says otherwise.

Tail

Robot foundation models, Gemini Robotics, Physical Intelligence's pi-0, could collapse the timeline and make general chores real by the early 2030s. Cheap to track, expensive to be surprised by.

A humanoid can do your chores today. Someone is just driving it from a headset.

Sources · figures checked against primary and direct reporting

Sources

1X NEO 2025, the teleoperation reality, Tom's Guide

IFR World Robotics 2026, ~7,000 humanoids sold in 2025, report

IFR 2025, Humanoid Robots, Vision and Reality position paper, paper

Figure AI 2025, Figure 03 & $39B valuation, overview

Tesla Optimus, timeline & $20k target, overview

Goldman Sachs 2024, $38B humanoid TAM by 2035, report

Morgan Stanley, ~$5T market / ~1B units by 2050, Humanoid 100

Citi, ~$7T / 650M humanoids by 2050, Global Insights

Bank of America, 3B humanoids by 2060, via Fortune

Rodney Brooks 2025, the sceptic case, Fortune

Toyota Research Institute, manipulation as the hard problem, TRI

Physical Intelligence 2024, the pi-0 generalist policy, pi0

Google DeepMind 2025, Gemini Robotics, announcement

Metaculus, reliable household robot ~2032, forecast

YouGov 2025, 38% want a household robot; top wishes floors/dishes/laundry, survey

iRobot, Roomba history & 50M+ units, overview

NYSOFA / ElliQ, eldercare, 95% less loneliness, NY State

Amazon Astro, business unit discontinued 2024, TechCrunch

Social-robot failures, Jibo, Anki, Kuri, IEEE Spectrum

Household robot market, Mordor ($11.7B 2026 → $23.5B 2031), report

Personal & household service robots, Grand View ($18B 2025 → $91B 2033), report

Changelog · what changed in this brief, newest first

Changelog