Recent news indicates that the State Administration for Market Regulation and the National Development and Reform Commission have jointly issued the "Guidance for Artificial Intelligence Measurement System and Capacity Building (2026 Edition)," systematically planning the development of AI measurement capabilities. This document directly addresses the industry pain points of AI technology performance being "unmeasurable, incomparable, and untraceable," promoting a shift in the industry from "building computing power and expanding scale" to "improving quality and strengthening foundations."
Just days before this, XUNCE Technology officially launched the world's first TokenOS operating system—TokenONE. With data tokenization and end-to-end trustworthiness at its core, it pioneers a new paradigm for measurement and cost governance in the AI era, precisely aligning with the policy guidance direction.
The product's release coinciding with supportive policy tailwinds has allowed the industry to keenly sense a signal of "playing the same strategic game"—this is not a mere coincidence in timing but a convergence of industrial transformation and top-level design.
Currently, enterprise AI applications are transitioning from single-point pilots to large-scale deployment. However, issues such as opaque billing, uncontrollable costs, difficulties in data rights confirmation, and challenges in verifying effectiveness are becoming increasingly prominent. The "Guidance" explicitly calls for establishing reliable, secure, and trustworthy measurement standards for artificial intelligence to achieve "measurable, comparable, and traceable" AI technology performance.
As a next-generation operating system designed for the AI-native era, TokenONE, through ten major product highlights, drives the transformation of enterprise data into data tokens that can be directly invoked by models. It ensures that each model invocation generates quantifiable and traceable business value, constructing an industrial production system from raw data to high-energy scenario tokens. This directly tackles the bottleneck of scarce data in enterprise scenarios, supports deployment on multiple computing power foundations and all leading large models in the market, makes large model outputs measurable, and promotes the large-scale commercial implementation of scenario token factories across various industries.
Addressing the long-standing industry pain point of opaque "black-box" billing, TokenONE establishes an end-to-end trustworthy billing system encompassing hardware isolation, encryption signatures, and real-time evidence preservation. It provides a secure, compliant, auditable, and tamper-proof measurement solution. Enterprises can clearly trace the source, purpose, and pricing basis of every token consumed, perfectly aligning with the core requirements of the "Guidance."
Regarding the business model, TokenONE has introduced a "Token Pay-As-You-Go" model, which has already gained market validation. Currently, the invocation price for XUNCE's vertical-specific tokens has reached $10–$100 per million tokens, significantly exceeding that of general-purpose models. In April 2026, token invocation Annual Recurring Revenue (ARR) showed a 300% quarter-over-quarter growth, with token-based paid revenue accounting for approximately 5% of total revenue. The company aims to increase this to 20%–30% by the end of 2026. As TokenONE is fully launched into the market, the process of enterprise data tokenization is accelerating significantly, with demand for high-value vertical scenario tokens continuing to surge, opening up a second growth curve for the company.
Industry observers widely believe that XUNCE's TokenONE fundamentally reconstructs billing credibility from the ground up, filling a critical infrastructure gap for the large-scale implementation of AI and resonating in sync with the national AI measurement standard construction. In the future, TokenONE is expected to become the standard operating system for enterprise AI, helping XUNCE solidify its leading position in AI data infrastructure.