A team of former Anthropic researchers has raised $200 million in seed funding just weeks after departing the frontier lab, marking one of the largest seed rounds in the AI industry. The startup aims to build AI systems that can create better AI, a concept that pushes the boundaries of automation in machine learning.
The round was led by Andreessen Horowitz and Kleiner Perkins, with participation from NVIDIA. The size of the round at $200 million and a reported valuation of $1 billion reflect intense investor appetite for talent leaving frontier labs to start their own ventures. This is a seed round, not a Series A, which underscores the unusual scale and urgency behind the deal.
The funding comes amid a broader reckoning over the true cost of AI services. A recent analysis noted that OpenAI booked roughly $13 billion in revenue in 2025 while posting an operating loss of about $21 billion, spending close to $1.60 for every dollar taken in. Frontier labs are deliberately pricing below cost to capture market share, subsidized by venture capital and cloud giants, a dynamic that analysts expect to reverse once the subsidy race winds down.
The emergence of a $200 million seed round for a team that has yet to ship a product suggests that investors are betting on the next wave of AI automation — systems that can improve themselves. That ambition could accelerate the pace of AI progress, but it also raises questions about safety and control as models gain more autonomy.
Counter_argument: Critics caution that such massive seed rounds for unproven teams may repeat the hype cycle of previous tech booms, where valuations outpaced actual product-market fit and led to eventual corrections.