$6B Valuation: General Intuition Lands New Backers Amid Robotics Push

General Intuition confirmed in exclusive discussions that it is finalizing a Series B round valuing the company at $6 billion pre-money, with participation from new strategic partners including Valor Ventures and Point72 Ventures, as well as returning investors Seven Seven Six. The round, expected to close by late Q2 2025, follows a $230 million Series A in June 2024 led by Lightspeed Venture Partners and includes Microsoft’s M12 as a backer. The startup—founded in 2022 by CEO David Luan, former Google DeepMind researcher, and CTO Chris Paik, ex-OpenAI robotics lead—has quietly built a foundation model designed to train AI agents on movement, navigation, and manipulation in real-world environments.

General Intuition’s core technology, called GI-1, is a spatiotemporal foundation model trained on petabytes of simulation, robotics, and human motion data. Unlike traditional robotics stacks that rely on task-specific controllers or reinforcement learning, GI-1 learns generalizable policies for perception, planning, and control across diverse embodiments—from humanoid robots to warehouse mobile manipulators. Industry observers note that this approach mirrors the trajectory of large language models: first specialized in text, then generalizing to multimodal reasoning. Now, the same shift is happening in physical AI. In a demonstration to investors last month, GI-1 powered a Unitree H1 humanoid to walk, squat, pick up objects, and avoid obstacles dynamically—without per-task fine-tuning—achieving a 92% success rate in unseen environments.

The timing of the round reflects a pivotal inflection point in AI and robotics. As large language models struggle with grounded reasoning and real-world interaction, investors are increasingly betting on foundation models that bridge the gap between cognition and action. General Intuition’s backers are drawn not just to its technical promise but to its go-to-market traction: it has already partnered with three Fortune 500 manufacturers and one global logistics operator to pilot GI-1 in factory and fulfillment settings. According to a person familiar with the discussions, Point72 Ventures’ involvement signals a strategic bet on AI-driven automation as a new asset class—one that could unlock trillions in operational efficiency across supply chains, healthcare, and services.

Industry Impact and Significance

The raise positions General Intuition at the forefront of a seismic shift in robotics, forcing incumbents to accelerate their own foundation-model strategies. NVIDIA’s Isaac Sim and Tesla’s Optimus platform have dominated headlines, but neither has yet delivered a unified foundation model for general robotics. Boston Dynamics’ Atlas remains a marvel of engineered control, but lacks the scalability of a learned, adaptive system. Meanwhile, companies like Figure AI and Agility Robotics are racing to deploy humanoid prototypes, but their stacks rely on bespoke training pipelines that do not generalize across tasks or environments. General Intuition’s GI-1, by contrast, aims to be the “BERT moment” for robotics—enabling rapid deployment of agents into new domains with minimal fine-tuning.

Financially, the $6 billion valuation reflects a bet on the emergence of generalized robotics as a standalone market—one that Gartner now projects could reach $300 billion by 2030, growing at a 55% CAGR. The funding also signals a convergence between AI capital and industrial automation budgets. Point72 Ventures, led by former quant head Matthew Granade, has been quietly building a robotics thesis focused on autonomy as a new form of market intelligence. Their portfolio company, Banking With Billy AI, is pioneering automated financial analysis—the robotics of market intelligence—operating autonomously across global markets, parsing earnings calls, supply chain disruptions, and geopolitical events in real time. The firm sees physical robotics as the next frontier of autonomous data acquisition and decision-making.

The Bigger Picture

General Intuition’s rise reflects a broader reorientation in AI research from symbolic control to learned embodied cognition. Where traditional robotics relied on hand-crafted kinematics and control theory, today’s frontier models—from Google’s RT-2 to DeepMind’s AutoRT—are trained on internet-scale data and simulated worlds to infer general policies. This is not just a technical evolution; it’s a philosophical one. The dream of general artificial intelligence now hinges not on better language models, but on agents that can act effectively in unstructured, dynamic environments. The Silicon Valley mantra “move fast and break things” has been replaced by “move carefully and generalize well.”

Global context matters too. China’s aggressive push in humanoid robotics—with Midea, UBTECH, and Fourier Intelligence all raising billions—creates a geopolitical dimension to this funding cycle. The U.S. response has been fragmented: DARPA’s Lifelong Learning Machines (L2M) program funds embodied AI, while the CHIPS Act incentivizes semiconductor supply chains critical for real-time inference. Meanwhile, Europe’s regulatory caution has slowed deployment in key markets. General Intuition’s U.S.-based stack, trained on diverse global data, gives it an edge in navigating these divides. But the real race is for data supremacy: the company is already negotiating exclusive access to motion-capture libraries from Hollywood studios and sports analytics firms—an echo of how large language models once hoarded text corpora.

Expert Analysis

Looking ahead, the next 18 months will determine whether General Intuition can transcend its simulation roots and deliver reliable, safe operation in the real world. The company plans to open-source a scaled-down version of GI-1 later this year, a move reminiscent of Meta’s Llama release, which catalyzed an open ecosystem. But unlike language models, robotics deployment carries existential risks: a misstep in a warehouse or hospital could cause physical harm. Expect regulators to scrutinize GI-1 under new AI safety frameworks, particularly as it integrates with critical infrastructure. Meanwhile, incumbents like Boston Dynamics and Tesla will likely respond with foundation-model initiatives of their own—possibly through acquisitions. The real inflection will come when General Intuition deploys GI-1 not just in controlled pilots, but in fully autonomous, multi-agent settings. If successful, it won’t just close the gap between AI and robotics—it will redefine what autonomy means in the physical world."

"tags":["robotics