Moonshot AI Targets $2 Billion in Annual Revenue Amid Allegations of Unauthorized Model Training

Chinese AI lab Moonshot AI is targeting $2 billion in annualized revenue by the end of 2026, according to a Bloomberg report published Friday — double the company’s revenue run rate recorded in August. The projection follows the summer release of its K3 model, which the company credits for the growth.

Moonshot AI is the maker of Kimi, an AI assistant built on the company’s open-weight models. K3, the lab’s latest model, is currently generating as many as 300 billion tokens per day on OpenRouter, though usage figures have declined slightly in recent months.

Despite the revenue target, Moonshot’s figures remain well below those of its closed-weight competitors. Recent reports put OpenAI’s annualized revenue at $40 billion and Anthropic’s at $65 billion. Because Moonshot’s model weights are freely available, the company operates on lower margins than those rivals. The rising projections suggest open-weight AI models can still generate meaningful revenue, even if they are less lucrative than closed-weight alternatives.

The revenue news arrives alongside serious allegations against the company. Earlier in the week of the Bloomberg report, Anthropic accused Moonshot AI of conducting a long-running model distillation campaign in which nearly 300,000 requests from Kimi were routed directly to Claude Opus — effectively serving Anthropic’s model in place of Kimi’s own. Anthropic alleged that more than 23 million responses from its models were collected and used in Moonshot’s training data. Moonshot’s model development practices, the source notes, “remain controversial — if not downright illegal.”

The combination of rapid revenue growth targets and unresolved legal and ethical questions places Moonshot AI at a pivotal and contested moment in its development.

Source: TechCrunch

This article was generated by AI and cites original sources.
Scroll to Top