Research data shows that AI agents consume four times the tokens of regular conversations, and multi-agent systems can reach up to 15 times; a 2025 analysis of AI-assisted software development found that engineers spend more than twice as much time reviewing and revising AI-generated code (9% of programming time) as they do waiting for AI to generate code (4%). The authors cite economic research from MIT, the University of Washington, and UCLA, stating, "The bottleneck to growth is no longer intelligence itself, but human verification bandwidth."
Lightspeed Partner: Token Overuse in AI Is Like APM Grinding in Games—Validation Capacity Is the Real Bottleneck
TechFlowShare
Summary
Lightspeed Partner warns that token overuse in AI mirrors APM grinding in games, with validation capacity as the real bottleneck. On-chain analysis reveals AI agents consume four times more tokens than regular conversations, while multi-agent systems reach 15 times. A 2025 study shows developers spend 9% of their time reviewing AI code, double the 4% spent waiting for generation. The fear and greed index may shift as human validation bandwidth, not AI intelligence, becomes the growth constraint, according to a joint economics paper from MIT, UW, and UCLA.
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.
Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.