French robotics-focused start-up Genesis AI has released its first foundational model, GENE-26.5, along with a human-like hand built in-house that’s capable of carrying out complex tasks such as cooking or solving a Rubik’s Cube. The announcement marks a significant milestone for the company, which is positioning itself at the intersection of artificial intelligence and physical robotics.
The company, which completed a $105 million (£77.2m) seed funding round in July of last year, began with a focus on a model to power real-world robotics, then built its own hand in order to have more accurate control over the full stack, said co-founder and chief executive Zhou Xian. This integrated approach allows Genesis AI to optimize both the software and hardware components, potentially accelerating the path to commercially viable robots.
Unlike robotic hands built by dozens of other companies, which tend to have two or three fingers, the unit from Genesis AI closely resembles a human appendage, allowing data to be more directly transferred to the device. This anthropomorphic design is crucial for tasks that require fine motor skills and dexterity, as it enables the robot to mimic human movements with precision. Many existing robotic grippers are optimized for industrial environments where objects are uniform, but Genesis AI‘s hand is designed for dynamic, unstructured settings typical of homes or labs.
Delicate tasks and data collection
Genesis has also created lightweight sensor gloves that can be worn by workers in fields such as pharmaceuticals or manufacturing, capturing data that can be passed along to the AI model – although whether workers would be happy to help train a potential robotic replacement remains in question. This ethical dilemma is not unique to Genesis AI; similar concerns have been raised in logistics and manufacturing where human workers are often asked to train algorithms that may eventually displace them. However, the company argues that such gloves can enhance worker safety by reducing repetitive strain and enabling remote operation of robots in hazardous environments.
Data from the gloves can be combined with information gleaned from videos of workers carrying out tasks, Genesis said. This multimodal approach leverages both tactile and visual inputs to create richer training datasets. In a demonstration video, Genesis showed a pair of its hands carrying out complex or delicate processes such as preparing a smoothie or playing the piano. The piano performance, in particular, highlighted the hand‘s ability to coordinate multiple fingers independently, a challenge that has eluded many robotic systems.
The development of GENE-26.5, the foundational model, is built on transformer architectures similar to those used in large language models, but adapted for spatial and kinematic reasoning. The model processes sensor data from the hand and gloves to generate real-time control signals. According to the company, the model can generalize across tasks without needing explicit programming for each new activity, reducing the time required for deployment.
Background and funding
Co-founder and president Theophile Gervet previously worked for France’s Mistral, a leading AI startup known for its open-source language models. Gervet‘s experience at Mistral likely influenced Genesis AI‘s approach to building scalable AI systems. The company retains a substantial presence in Europe that it attributed to the continent’s large talent base and potential market of industrial customers. Europe, particularly France, has become a hub for AI research due to initiatives like the French government‘s “AI for Humanity” strategy and the presence of institutions like INRIA and the Paris-Saclay University.
It said its team of 60 people is split roughly evenly between the US and Europe, with a slightly larger group in the US base in California. In addition to San Carlos, California, it has offices in Paris and London. This transatlantic structure allows the company to tap into the venture capital ecosystem in Silicon Valley while leveraging Europe‘s deep pool of engineering talent.
The seed funding round last year was co-led by Eclipse and Khosla Ventures, with additional backing from Bpifrance, HSG, former Google chief Eric Schmidt, telecoms entrepreneur Xavier Niel, MIT Computer Science and Artificial Intelligence Laboratory director Daniela Rus and Apple distinguished scientist Vladlen Koltun. The involvement of such high-profile investors underscores the strategic importance of robotics and AI. Schmidt, a vocal advocate for AI regulation and development, has backed several robotics startups in recent years. Niel, a major investor in French tech, has been instrumental in nurturing the Parisian ecosystem.
Genesis AI‘s $105 million seed round is among the largest in the robotics sector, reflecting the high capital requirements for hardware and AI development. For comparison, many robotics startups raise smaller rounds at this stage, often focusing on software-only solutions. Genesis AI‘s decision to build custom hardware signals confidence in its technology and market timing.
Competitive landscape
The robot hand market is becoming increasingly crowded, with companies like Shadow Robot Company (UK), Mechanical Engineering at Stanford, and others developing dexterous grippers. However, most focus on industrial applications where simplicity and durability are paramount. Genesis AI‘s human-like hand, with its five fingers and semi-articulated joints, aims for versatility. Other notable efforts include the DLR Hand II (Germany), the BarrettHand, and the Keio Hand from Japan, but integration with a foundation model like GENE-26.5 sets Genesis apart.
The company said it is working to supplement the mechanical hand with a full-body, general purpose robot. This ambitious goal would require solving locomotion, manipulation, and perception in a single platform. Competitors in the full-body humanoid space include Tesla‘s Optimus, Boston Dynamics’ Atlas, and Figure AI, but Genesis AI‘s approach of starting with an advanced hand may provide a unique advantage: if the hand can already perform complex tasks, integrating it with a bipedal body could accelerate overall development. The full-body robot is expected to use the same GENE-26.5 model, possibly with additional modules for walking and balance.
The use of sensor gloves for data collection also parallels research at institutions like the Max Planck Institute for Intelligent Systems, where tactile sensing is being explored for fine manipulation. However, Genesis AI appears to be closer to commercialization, having already demonstrated functional tasks.
One of the key challenges Genesis AI faces is the economic viability of its solution. Building a human-like hand with multiple sensors and actuators is expensive. The company will need to prove that its robot can perform tasks more efficiently than humans or existing specialized machines. The target markets—pharmaceuticals, manufacturing, and possibly healthcare—require high precision and adaptability. For example, in pharmaceutical labs, robots could handle hazardous materials or repetitive pipetting, reducing contamination risks. In manufacturing, the hand could perform assembly tasks that currently require human dexterity.
The ethical considerations around worker data capture cannot be ignored. While sensor gloves may improve ergonomics and safety, they could also be used to monitor worker performance or to replace them entirely. Genesis AI has not yet publicly addressed these concerns, but the company may need to adopt transparent data policies and engage with labor unions to avoid backlash.
Another technical hurdle is sim-to-real transfer. While data from gloves and videos can train the model offline, the real world presents unexpected variations in object shape, texture, and lighting. Genesis AI‘s demonstration videos suggest it has made progress in handling such variability, but scaling the system to a wide range of environments will require extensive testing.
Looking ahead, the company plans to refine the hand‘s sensitivity and add more degrees of freedom. The full-body robot is still in early development, but the team’s background in AI and hardware suggests they are taking a methodical approach. If successful, Genesis AI could become a key player in the next wave of robotics, where machines are not confined to cages but work alongside humans in homes and factories.
Source:Silicon UK News