A closely watched measure of artificial intelligence token prices, the LLM Token Expenditure Index from Silicon Data, fell to a new record low of 97 cents on Monday, marking its lowest level since inception late last year and more than halving from its peak earlier this summer [1]. This sharp decline reflects mounting competition in the AI sector, with the rise of open-source Chinese models such as Moonshot's Kimi K3 offering lower prices than leading frontier labs, further pressuring market rates [1].
OpenAI responded to these market dynamics by announcing price cuts to two of its GPT-5.6 AI models in late July, while other labs have introduced 'dynamic pricing' that adjusts access rates based on demand, intensifying downward pressure on token prices [1]. Charles-Henry Monchau, investing chief at Syz Group, noted that foundation model labs are most exposed to this deflation, as falling token prices compress revenues while fixed compute costs remain, prompting a strategic shift toward differentiation in distribution, memory, and context rather than raw model capability [1].
The drop in the LLM Token Expenditure Index could create new profit pressures for AI leaders like Anthropic and OpenAI, both of which confidentially filed for IPOs this summer [1]. Investors may need to reassess their expectations for returns on capital invested in AI infrastructure, as megacap technology firms such as Nvidia and Microsoft have committed billions to expanding AI capabilities [1]. According to Steve Hou, Silicon Data's head of research, the current supply of both frontier and cheaper models may already be sufficient for most tasks, suggesting a potential plateau in demand-driven price growth [1].
The broader market reflected these pressures, with technology stocks leading declines on Tuesday: the Nasdaq Composite fell nearly 1%, and the S&P 500 slipped 0.4% [1].
CONCLUSION
The rapid decline in AI token prices signals intensifying competition and cost pressures in the sector, challenging the profitability of leading AI labs and prompting strategic shifts. Investors and market participants may need to recalibrate expectations for returns as the market adjusts to a new pricing environment and sufficient supply. The negative sentiment was echoed in broader technology stock declines.