AI Revenue Cannot Sustain Cash Burn Rate, Google Unveils "Self-Disruption" New Narrative

According to Omdia Beating, Google is expected to invest around $200 billion in AI data centers and devices this year. Jasjeet Sekhon, Chief Strategy Officer of DeepMind, admitted that the current AI revenue cannot cover such expenses, leading to a potential "spending first, revenue lagging" scenario in the industry.

He identified the long-term rationale for Google's heavy investment as RSI, which involves using AI to participate in developing the next generation of AI and continuously improving training methods and the models themselves. Sekhon stated that RSI has become a key logic behind this massive investment cycle.

However, the true RSI has not materialized yet. Current AI can only optimize certain algorithms, kernels, and training processes. Sekhon predicts that it may be achieved in the next two years. Google is justifying today's $200 billion investment with a technology goal that has not yet been realized.