AI News Feed
Figure 03: Mass-Produced Humanoid Home Robot
13 Oct 2025- Figure AI unveiled the Figure 03 humanoid home robot with Helix vision-language-action learning, hardware upgrades, BotQ factory plans, $1B funding, and mass-production targets despite demo reliability issues.
13 Oct 2025
Figure AI unveiled the Figure 03 — a humanoid home robot the newsletter dubs “the Model T of robots.” It’s been designed for mass production (Figure is building a factory called BotQ) and is positioned as a consumer-focused helper for homes, with demos targeting tasks like folding laundry, loading dishwashers, clearing tables and watering plants.
Under the hood is Figure’s Helix system, a vision-language-action stack that learns by watching humans; the team says the robot learned towel folding after about 80 hours of video. Hardware upgrades over Figure 02 include cameras with 2× frame rate, ~60% wider field of view and ~75% lower latency, palm cameras, tactile fingertip sensors sensitive down to ~3 grams, wireless charging through its feet, and improved speakers/microphones. The body uses soft, washable fabric and is ~9% lighter than prior designs.
Figure recently raised $1B at a reported $39B valuation from investors including NVIDIA, Jeff Bezos, OpenAI and Microsoft, and plans initial BotQ output of ~12K robots/year with a multi-year target of 100K. The product earned TIME’s Best Inventions nod, but CEO Brett Adcock admits they aren’t “there yet”: the demo machine was finished only a week before the demo and had trouble reliably handling laundry. Roboticists also debate whether humanoid is the right consumer form factor versus alternatives like hexapods.
Source
Figure AI unveiled the Figure 03 — a humanoid home robot the newsletter dubs “the Model T of robots.” It’s been designed for mass production (Figure is building a factory called BotQ) and is positioned as a consumer-focused helper for homes, with demos targeting tasks like folding laundry, loading dishwashers, clearing tables and watering plants.
Under the hood is Figure’s Helix system, a vision-language-action stack that learns by watching humans; the team says the robot learned towel folding after about 80 hours of video. Hardware upgrades over Figure 02 include cameras with 2× frame rate, ~60% wider field of view and ~75% lower latency, palm cameras, tactile fingertip sensors sensitive down to ~3 grams, wireless charging through its feet, and improved speakers/microphones. The body uses soft, washable fabric and is ~9% lighter than prior designs.
Figure recently raised $1B at a reported $39B valuation from investors including NVIDIA, Jeff Bezos, OpenAI and Microsoft, and plans initial BotQ output of ~12K robots/year with a multi-year target of 100K. The product earned TIME’s Best Inventions nod, but CEO Brett Adcock admits they aren’t “there yet”: the demo machine was finished only a week before the demo and had trouble reliably handling laundry. Roboticists also debate whether humanoid is the right consumer form factor versus alternatives like hexapods.
Source
The method
The prompts
Helping entrepreneurs and business owners use AI tools for growth and productivity.
Wawa Llama Ltd GB447340593