Bytes are fading; tokens are the future!
In an era when traffic is priced like a commodity, why can tokens become the 'hard currency' of the AI internet? The answer lies in a rebuilt means of production and the logic of commercialization; tokens are not a replacement for bytes but the value anchor the AI era cannot bypass.
In mobile internet days, data became dirt-cheap — from tens of RMB per GB in 2011 to under RMB 2/GB by 2025. Users mostly paid for access and enjoyed services freely on top.
AI rewrites the rules: the new infra stack — compute (with GPUs atop the stack), storage, networking, and power — is costly, while vendors shoulder heavy model training and inference bills. Traffic monetization alone cannot cover the economics. Tokens emerge as the metering and pricing unit for AI services, with ChatGPT subs effectively token bundles and AI companion devices offering monthly plans priced by tokens, aligning exactly with the cost structure.
Data reinforce tokens' centrality: in H1 2025, China averaged 30trn tokens used per day, with ByteDance at a 49% share and its Doubao processing as many as 50trn daily.
While total inference spend remains high, unit inference cost fell 90%+ over two years, making the token model commercially viable.For users, tokens may look like an added fee but are the necessary price for near-human reasoning. For vendors, tokens bridge the cost gap, enable profitability, and serve as the core lever for ecosystem influence.The future of tokens is the future of AI commercialization. As models improve and costs fall further, users will normalize paying, and tokens will shift from 'optional' to 'must-have' pricing.
Tokens will not displace bytes; they complement data plans and support continuous iteration of AI services. They are not just a pricing unit but the value core of the AI internet, with importance rising as AI penetration deepens.
$Alibaba(BABA.US) $TENCENT(00700.HK) $ByteDance(BYTED.NA)
The copyright of this article belongs to the original author/organization.
The views expressed herein are solely those of the author and do not reflect the stance of the platform. The content is intended for investment reference purposes only and shall not be considered as investment advice. Please contact us if you have any questions or suggestions regarding the content services provided by the platform.
