Figure Raises $675M at $2.6B Valuation and Signs Collaboration Agreement with OpenAI

Figure Raises $675M at $2.6B Valuation and Signs Collaboration Agreement with OpenAI | Just Think AI
May 21, 2024

Figure, the leading company developing humanoid robots, recently announced a $675 million Series B funding round led by Eldridge, Lupa Systems, Microsoft's M12, NVIDIA and VentureFriends. This massive round of funding values Figure at $2.6 billion and brings its total funding raised to date to over $1.5 billion. Alongside the funding news, Figure also revealed a new partnership with leading AI lab OpenAI to collaborate on developing next-generation AI models tailored specifically for embodied humanoid robots.

Overview of Figure and Their Humanoid Robots

Founded in 2013, Figure is a robotics startup based in Sunnyvale, California that is focused exclusively on developing humanoid robots. Unlike industrial robot arms which are fixed in place, humanoid robots are able to move around on two legs and interact with the environment using their arms, hands and sensors.

Figure's humanoid robots are powered by proprietary hardware and software that gives them coordinated and dynamic mobility allowing them to walk, balance, and manipulate objects using both arms. Some of the key capabilities of Figure's humanoids include:

  • Bipedal walking: Figure robots can walk on two legs, navigate environments, go up stairs, etc. This allows them mobility to transport items or get to locations inhuman operators can't easily access.
  • Object manipulation: Using vision systems, depth sensors and AI models, Figure humanoids can pick up, transfer, and place items with a high degree of precision using both arms in coordination.
  • Fast adaptation to new situations: The robots continuously observe environments, detect changes, and rapidly assimilate novel scenarios. This intelligence allows deploying them into uncontrolled real-world settings.
  • Safe interaction: With embedded vision and perception abilities, Figure humanoids can detect people and objects and modify behaviors to operate carefully around humans.

Figure is positioning their humanoids for commercial use cases like retail operations, inventory and warehouse management, last mile delivery, healthcare settings and more.  The recently raised funding and OpenAI collaboration aims to accelerate their ability to bring this technology to market.

Key Details of Figure's $675M Series B Funding Round

The $675 million Series B funding Figure recently secured represents a massive vote of confidence from top institutional investors in their humanoid robot technology.

The round was led by Eldridge, the firm led by Todd Boehly focused on media, sports and technology. Other major investors include:

  • Lupa Systems: Backed by James Murdoch, focusing on media and emerging technologies.
  • Microsoft's venture capital fund M12: One of the largest corporate VC funds globally.
  • NVIDIA: Leader in AI computing infrastructure. Prior investor in Figure.
  • VentureFriends: Backed by Greek entrepreneur Spyros Magiatis, focused on deep tech.

“This latest funding round will allow us to take the next giant leap toward deploying humanoids at scale into commercial environments,” said CEO Michael Degroot.

Bringing the total raised by Figure to date to over $1.5 billion, this latest round gives Figure substantial resources to meet their goal of having humanoids deployed into real businesses within the next two years.

It also reflects strong investor confidence that Figure's humanoid technology will transform how work gets done across many industries.

Figure Partners with OpenAI to Advance Next-Gen AI Models

In tandem with their new funding round, Figure also announced a new partnership with OpenAI, the leading non-profit AI research organization.

Figure will collaborate with OpenAI researchers to develop new state-of-the-art artificial intelligence models tailored specifically for deployment on Figure's humanoid robot hardware platforms.

Some areas this partnership aims to push forward include:

  • Sim2Real transfer: Training models in simulation and transferring learnings to the real world. Dramatically cuts experimentation time.
  • Multimodal perception: Combining vision, depth sensing, proprioception for advanced environmental understanding.
  • Long-term autonomy: Enabling robots to operate independently over long time periods.
  • Scene understanding: Interpreting objects, people and dynamics in uncontrolled settings.

By combining Figure's expertise in robotics hardware and autonomy software with innovations coming out of OpenAI's labs, this collaboration aims to take embodied AI to the next level and pave the way for real-world deployment at scale.

What This Funding and OpenAI Deal Means for Figure

The staggering $675 million Series B round and high-profile OpenAI partnership serves as resounding validation of Figure's technical strategy and the promise of their humanoid robots.

Some of the key implications for Figure include:

Investor Confidence in Figure's Approach

Top tier investors are betting big that Figure's focus on developing generalist humanoid robots represents the future of how work will get augmented by machines over the coming decades.

The $2.6 billion valuation cements Figure as a leader in the red hot AI robotics sector.

War Chest to Accelerate Timelines

With over $1.5 billion in funding raised to date, Figure now has ample capital to pour into perfecting their core humanoid platform and speeding up their go-to-market timeline.

They are aggressively expanding hiring across all functions from hardware engineers to business development roles. This will enable launching pilot programs with major enterprises within the next 12-18 months.

Pole Position in Humanoid Robot Race

While other startups and academic labs are also working on humanoids, none have yet reached Figure's level of demonstrated capability combined with such strong financial backing.

This deal firmly places Figure at the front of the pack to be the first to get these futuristic robots into mainstream commercial use.

Overview of Figure's Humanoid Robot Technology

While general purpose humanoid robots have been a staple of science fiction for decades, building machines that combine mobility, manipulation skills, autonomy and efficiency has proven incredibly difficult from an engineering perspective.

Figure stands at the cutting edge of making humanoids a reality thanks to core innovations across the technology stack:

Hardware Platform

Figure's humanoids feature custom designed hardware optimized specifically for dynamic mobility and dexterous object manipulation. Key hardware components include:

  • Actuation system: Proprietary electro-mechanical design for smooth and efficient movement comparable to humans. Enables walking, squatting, bending.
  • Sensors suite: Includes cameras, LiDAR, proprioceptive sensors. Provides 360 degree environmental perception for localization, mapping and obstacle avoidance.
  • Computation stack: Leverages NVIDIA GPUs for energy efficient on-board processing and inference.
  • Object manipulation: Custom designed arms and grippers tailored towards dexterous grasping and fine motor control.

By designing hardware systems holistically around the capabilities needed for versatility and adaptability in uncontrolled real-world settings, Figure robots push the boundaries of what's possible in physical capability for humanoids.

Software Architecture

On top of the custom robotic hardware, Figure has invested heavily into the proprietary software intelligence that brings these machines to life:

  • Sensor fusion algorithms: Fuse data from vision, LiDAR, noise and vibration to achieve accurate proprioception and situational awareness.
  • Sim2Real transfer learning: Models trained extensively in high fidelity simulations using machine learning techniques transfer seamlessly to the real world through domain adaptation and fine-tuning.
  • Behavior trees: Blend reactionary reflexes with higher level planning and task execution modules to balance safety and productivity.
  • Graph based autonomy: Maps environments and represents object relationships, spatial connections and situational context for long term autonomy.
  • Analytics engine: Logs vast amounts of data during operation for assessing utilization, reliability and areas needing performance improvements.

This layered software architecture equips Figure humanoids with functional intelligence comparable to humans. Paired with the mechanical versatility of the hardware, it empowers use across diverse uncontrolled settings beyond the capacity of conventional industrial robots.

Real World Use Cases for Figure's Humanoid Robots

While still early days, Figure has identified a number of promising applications where their humanoid's unique capabilities can transform workflows:

Retail Operations

In retail environments, humanoids can stock shelves, handle inventory, transport packages, and assist customers. Benefits include:

  • Adaptability to rearrange products, shelves, fixtures as needed
  • Handling variety of objects of different sizes and textures
  • Free staff to focus on higher judgement tasks
  • 24/7 productivity without fatigue

Healthcare Settings

Healthcare spans applications like hospital logistics, pharmaceutical warehouses, senior home assistance. Figure robots offer:

  • Deliver food, laundry, medical supplies around facilities
  • Manipulate fragile medicine boxes, vials, syringes safely
  • Better sterilization tolerance than human workers
  • Friendly assistant to patients alongside staff

Supply Chain and Logistics

Figure humanoids are positioned to optimizing logistics in warehouses, 3PLs and container yards via:

  • High mobility to navigate tight spaces filled with obstacles
  • Lifting and moving boxes, pallets up to 30 kgs
  • Sorting irregular shaped items even if unorganized
  • Handling traceability and inventory auditing

These settings involve high physical variability, dexterous manipulation skills and situational adaptability playing directly into the strengths of Figure's generalist humanoid approach.

OpenAI Partnership to Accelerate Embodied AI Breakthroughs

Figure's recently announced technology partnership with OpenAI has momentous implications for accelerating innovations specifically around embodied artificial intelligence.

OpenAI is famous for leading key breakthroughs in modern AI around large language models like GPT-3, robotic learning algorithms, and solving complex coordination challenges for robot swarms.

By pooling together the expertise both organizations have developed respectively around AI algorithms and physically capable robots, this partnership can catalyze progress on multiple fronts:

Challenges in Real World Robot Autonomy

While great strides have occurred recently in AI, the vast majority of advances remain restricted to success in digital environments like games or simulation.

Unfortunately, many of these techniques fail catastrophically when deployed onto real world robotic systems. This "Sim2Real Gap" remains one of the biggest unsolved challenges obstructing rollout of smart machines.

Figure's extensive datasets and experience around bridging simulation-to-reality transfer will be invaluable for OpenAI to ensure their leading algorithms can be tailored successfully for physically-instantiated robots operating in open environments.

Supervised Autonomy

Another promising direction is combining open-ended autonomous learning algorithms with modular software blocks that provide oversight and stepped intervention if needed.

This "supervised autonomy" approach would build on OpenAI frameworks like TreasureHunt ​while leveraging Figure's hierarchical architectures for balancing safety and productivity.

Meta-learning Techniques

An emerging paradigm in robotic AI is meta-learning - where systems evolve to quickly assimilate entirely new tasks from just a few examples by building on prior knowledge.

Figure's simulators combined with OpenAI's expertise in fast adaptation and few-shot generalization would drive bleeding edge research on next-gen hyper-flexible machine intelligence.

Further down the road collaborations may even explore how self-supervised feedback learned by one Figure robot could be shared peer-to-peer to bootstrap skills acquisition across an entire fleet working in concert.

The recently secured $675 million funding round at a stratospheric $2.6 billion valuation cements Figure as the undisputed leader in cutting-edge humanoid robot technology. With ample financial muscle and links now forged with the world's top AI research lab in OpenAI, Figure is positioning itself to open up a new era where intelligent humanoid machines begin collaborating alongside humans to transform a myriad of industries.

While it will still take time to perfect these robots and choreograph their safe and reliable integration into real-world businesses, this funding and collaboration lays the groundwork for Figure to take their first steps out of factories and into mainstream over the next few years.

What seems certain is that the robotic future is arriving ahead of schedule thanks in large part to Figure's technical vision and their expanding web of high profile backers who believe their generalist approach will define how the next generation of automation gets embedded across our economic and social fabrics.

MORE FROM JUST THINK AI

MatX: Google Alumni's AI Chip Startup Raises $80M Series A at $300M Valuation

November 23, 2024
MatX: Google Alumni's AI Chip Startup Raises $80M Series A at $300M Valuation
MORE FROM JUST THINK AI

OpenAI's Evidence Deletion: A Bombshell in the AI World

November 20, 2024
OpenAI's Evidence Deletion: A Bombshell in the AI World
MORE FROM JUST THINK AI

OpenAI's Turbulent Beginnings: A Power Struggle That Shaped AI

November 17, 2024
OpenAI's Turbulent Beginnings: A Power Struggle That Shaped AI
Join our newsletter
We will keep you up to date on all the new AI news. No spam we promise
We care about your data in our privacy policy.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.