Rephrase the title:Covariant Develops Language Model That Can Make Robots Think Like Humans

Covariant, the artificial intelligence spinout from UC Berkeley, has unveiled RFM-1 (Robotics Foundation Model 1), positioned as a “large language model (LLM) for robot language” by CEO Peter Chen.

Developed using data collected from Covariant’s Brain AI platform, RFM-1 aims to extend beyond warehouse applications and empower robots across diverse industries such as manufacturing, food processing, recycling, agriculture, and homes. Peter Chen envisions RFM-1 as a revolutionary solution capable of propelling billions of robots into various sectors, emphasizing the need for adaptable systems.

According to TechCrunch, the platform’s launch aligns with ongoing discussions in the robotics industry regarding the future of “general-purpose” systems, particularly with the emergence of humanoid robotics firms. Covariant’s software, currently deployed on industrial robotic arms for tasks like bin picking, exhibits hardware agnosticism, promising compatibility with various robotic systems.

Robots with Human-Like Reasoning

RFM-1 is heralded for providing robots with a “human-like ability to reason,” a departure from traditional robotic systems programmed for repetitive tasks. In contrast to single-purpose robots, RFM-1 processes real-world data, determining optimal actions based on a deeper understanding of language and the physical world. The platform’s interface allows users to input text commands, fostering natural interactions with robots.

Covariant aims to bridge the gap between traditional programming and a more intuitive method for human-robot interaction, promoting efficiency and adaptability. While the claims of a “human-like ability to reason” warrant careful consideration, Covariant’s RFM-1 showcases a new paradigm in robot decision-making, with potential applications revolutionizing training, programming, and deployment across various industries.

Read Also: Real-Time HD Imaging Revolutionizes Spinal Cord Procedures

Big Tech Firms Investing in AI-Powered Robotics

TechTimes earlier reported that Tech mogul Jeff Bezos, along with major players like Nvidia Corp., spearheaded investments in Figure AI Inc., a Sunnyvale-based startup focused on humanoid robot development.

Amid a funding round targeting around $675 million, with an estimated pre-money valuation of $2 billion, Jeff Bezos commits $100 million through Explore Investments LLC. Microsoft invests $95 million, while Nvidia and an Amazon-affiliated fund contribute $50 million each.

Last year, Parkway Venture Capital led the $70 million initial external funding round for Figure AI, which also included $20 million from the company’s CEO and founder Brett Adcock. The ongoing funding round includes investments from Intel Corp.’s venture capital arm, LG Innotek, Samsung’s investment group, Parkway Venture Capital, Align Ventures, ARK Venture Fund, Aliya Capital Partners, Tamarack, Boscolo Intervest Ltd., and BOLD Capital Partners.

While the final details of the funding round are pending confirmation, these substantial investments underscore the increasing interest in the AI robotics sector. Figure AI focuses on advancing AI technology, concentrating on creating humanoid robots capable of functioning in diverse environments, such as warehouses and retail spaces, and handling hazardous tasks, with support from OpenAI and Microsoft.

The broader AI robotics industry witnesses significant activity, with companies like 1X Technologies AS and Sanctuary AI securing substantial funds for humanoid robot development. Elon Musk, the company’s CEO, is actively developing the humanoid robot Optimus in anticipation of a billion humanoid robots populating the planet by the 2040s, thanks to the expanding potential of AI-powered software to improve robot-human interaction and collaboration.

Related Article: Elon Musk’s ChatGPT Competitor Grok Will be Open-Source This Week