Keep (03650) launches Super AI membership, bringing industry-specific AI into a new phase of productization.

On August 5th, sports technology company Keep (03650) officially released version 9.1 of the Keep App and introduced a new benefit, the "Super AI Membership." This marks another significant product launch following the announcement of the self-developed sports health model Keepace.ai. The Super AI Membership integrates capabilities such as AI voice interactive running, analysis of eating, training, sleeping, and body age analysis into an independent service, providing an all-encompassing sports health experience that covers planning, execution tracking, and results review, specifically aimed at users seeking deep personalized guidance. To lower the entry barrier, from August 7th to 27th, Keep will initiate an "Check-in for Super AI Membership, signing up for health 'gold'" campaign and a charitable donation initiative to encourage more users to personally experience the changes brought by AI services in real sports scenarios. The newly launched Super AI Membership is not a replacement or mandatory upgrade for existing member services. Regular members will continue to enjoy basic benefits such as exclusive courses, personalized plans, and body assessments; the Super AI Membership builds on these basic rights by adding features like AI voice interactive running, eating, training, sleeping analysis, body age analysis, and gym tools (to be updated). The two types of memberships cater to different needs, allowing users to choose based on their training habits and the depth of service they require. In specific scenarios, the AI voice interactive running allows for two-way dialogue during running, providing guidance and encouragement based on pace, cadence, and real-time conditions; personalized schedules will arrange training rhythms based on users' basic information, fitness goals, and recent performances; photo-based diet analysis can identify food items and calculate calorie gaps and nutritional ratios based on exercise goals; sports data and sleep analysis will translate dispersed metrics into more understandable and actionable adjustment suggestions. The body age feature will integrate sleep, cardiovascular, and other metrics to help users see the gap between their physical condition and actual age. The upcoming fully launched gym tools will also support image identification for recommended exercises and provide voice assistance for arranging and recording training sessions. Thus, the AI membership provides not a static set of courses or a one-time tool usage right, but rather a continuously generated and dynamically iterated health plan based on personal data. As users continuously generate new exercise and health data, the content of the service will also be updated in real-time. In April of this year, Keep announced its self-developed large model Keepace.ai, specifically designed for sports health scenarios. Since then, the model has been continuously iterated and has gradually implemented capabilities such as AI voice interactive running and more comprehensive data analysis. The launch of the Super AI Membership is a concentrated productization of these already established capabilities and marks a new phase in Keep's AI development from underlying technology investment to productization and service orientation. Unlike simply integrating a generic large model, Keepace.ai combines a decade's worth of sports data assets accumulated by the Keep platform, supported by a team of experts in sports health from Capital Sports University, and integrates their deep research in exercise physiology, training science, and injury prevention. It is designed to understand users while strictly maintaining safety standards, providing highly personalized and risk-averse exercise guidance. Keep has stated that it will continue to make long-term investments in AI research and productization, gradually transforming technological capabilities into perceptible service experiences for users and exploring new commercialization paths. It is expected that in the second half of this year, further applications of AI in the sports health domain will continue to be released.