Humans&, a three-month-old AI startup founded by veteran researchers from Anthropic, xAI, and Google, has raised $480 million in seed funding at a $4.48 billion valuation, marking one of the largest seed rounds in technology history. The company announced the funding on Tuesday, attracting investments from Nvidia, Amazon founder Jeff Bezos, and prominent venture capital firms including SV Angel, Google Ventures, and Laurene Powell Jobs's Emerson Collective.
The extraordinary valuation for a pre-product company signals intense investor confidence in elite AI research teams pursuing next-generation capabilities beyond current chatbot and agent technologies. The round was led by SV Angel, founded by legendary Silicon Valley investor Ron Conway, alongside Humans& co-founder Georges Harik, who served as Google's seventh employee and played instrumental roles building Gmail, Google Docs, and leading the Android acquisition.
Elite Pedigree Powers Investor Conviction
The founding team represents a concentration of expertise from AI's most influential labs. Andi Peng, formerly at Anthropic, contributed to reinforcement learning and post-training for Claude models spanning versions 3.5 through 4.5. Eric Zelikman and Yuchen He both worked at xAI, helping develop the Grok chatbot and bringing deep experience in reasoning-based reinforcement learning. Stanford professor Noah Goodman rounds out the founding team with academic expertise in psychology and computer science.
The startup's approximately 20 employees include alumni from OpenAI, Meta, Reflection, AI2, and MIT, assembling what investors view as exceptional technical talent capable of advancing fundamental AI capabilities. This concentration of experienced researchers from competing labs reflects a broader trend of breakaway teams securing massive early-stage funding based primarily on founder credentials and vision rather than demonstrated products or revenue.
Human-Centric Vision Distinguishes Approach
Humans& positions itself around empowering human collaboration rather than replacing workers, addressing growing concerns about AI's impact on employment. The company plans to develop software helping people collaborate more effectively, conceptually similar to AI-enhanced instant messaging that goes beyond simple automation to strengthen organizational and community connections.
The technical approach involves training AI systems using novel methods, including programming chatbots to proactively request information from users and store context for long-term utilization. This focus on memory, multi-agent reinforcement learning, and user understanding aims to create AI that serves as connective tissue between people rather than isolated task completion tools.
The company emphasizes innovations in long-horizon reinforcement learning and tightly integrated product development alongside scientific research. Humans& expects to launch its first product early this year, though specific details about capabilities or target markets remain undisclosed.
Strategic Investment Landscape
Nvidia's participation reflects the chipmaker's strategy of taking stakes in AI startups heavily dependent on its computing hardware, ensuring proximity to future demand drivers while supporting the ecosystem powering its revenue growth. The company has emerged as a key backer across multiple next-generation AI labs as demand for its GPUs continues surging.
The funding continues a pattern of investors allocating massive capital to breakaway teams from established AI labs. Similar mega-rounds have valued Safe Superintelligence at $32 billion and Thinking Machines at $12 billion, both founded by former OpenAI researchers. These valuations reflect market belief that the next AI breakthroughs will emerge from small, elite teams with deep technical expertise and substantial computing budgets rather than incremental improvements to existing systems.
For the broader AI industry, Humans&'s funding underscores that competition remains focused on fundamental capabilities and research talent rather than application-layer execution. The willingness of top-tier investors to bet billions on pre-product companies demonstrates sustained confidence in AI's transformative potential despite ongoing debates about current model limitations and return on investment timelines.
