The East Liberty-based AI robotics startup Skild AI has raised $1.4 billion in fresh funding, boosting the company’s valuation to over $14 billion.

Launched in 2023, Skild AI is building a general-purpose AI model designed to power a wide range of robots and tasks with a single system.

“The Skild Brain can control robots it has never trained on, adapting in real time to extreme changes in form or environments. The model is forced to adapt rather than memorize — much like intelligence in nature,” Deepak Pathak, CEO and cofounder of Skild AI, said in today’s announcement. “We believe that a unified, omni-bodied brain is the fastest way to establish a continuous data flywheel where the model gets better with every single deployment, no matter what the hardware or task.”

The round includes a long list of investors. It was led by SoftBank, with participation from big-name investors like NVIDIA’s venture capital arm NVentures and Jeff Bezos (through Bezos Expeditions).

The company will use the new capital to continue scaling its model training and “grow the future deployment of its technology,” it said in a press release. 

Skild AI did not immediately respond to Technical.ly’s request for comment for further details and on whether it will expand facilities or hiring in Pittsburgh.

See the full list of Skild’s latest investors

• SoftBank
• NVentures
• Jeff Bezos (through Bezos Expeditions)
• Unnamed entities administered by Macquarie Capital
• Disruptive
• 1789 Capital
• Lightspeed
• Felicis
• Coatue
• Sequoia Capital
• Samsung
• LG
• Schneider
• CommonSpirit
• Salesforce Ventures
• TF Capital
• Andra Capital
• Palo Alto Growth Capital
• KIC
• Alpha Square
• Mirae Asset
• Destiny

Skild’s journey to becoming a unicorn

Last summer, Skild unveiled its Skild Brain. Videos showed the company’s AI model being used in humanoid, dog and tabletop arm robots completing various tasks, like walking a dog or washing dishes. The company trains its model on videos of humans and physics-based simulations.

Skild says, unlike traditional models, its AI is “omni-bodied,” meaning it can control any robot without prior knowledge of its exact body form. It does this through in-context learning, or when the model is put into an unknown environment and it adjusts its behavior based on its lived experience in real time.

“We believe this omni-bodied learning is essential for building [artificial general intelligence] that works reliably in the physical world, paving the way for robots that can safely help humans in everyday environments,” Abhinav Gupta, cofounder and president of Skild AI, said in a prepared statement. “This enables robots to operate dynamically in complex environments, without requiring preprogrammed instructions for each scenario.”

Skild AI became profitable last year, making about $30 million in 2025, according to a press release from the company. Its model is currently being tested in a variety of settings, including security inspections, package delivery, warehouses, manufacturing, data centers and construction tasks.

“The company plans to ultimately deploy robotics in consumer homes, with enterprise tasks as the first application,” it said in the release.

The startup raised $500 million in April 2025 and $300 million in July 2024. However, there are no corresponding Form D filings under the Skild AI name with the US Securities and Exchange Commission, which startups are required to file, though they can do so under an alias.

The company has also expanded into new markets over the last year, including offices in San Mateo, California, and Bengaluru, India.