Industry Experts Propose Diverse Standards for Assessing Artificial Intelligence Implementation

China News Service reported that attendees at the Global Asset Management Forum 2026 Shanghai Suhewan Conference round table on October 17 examined the criteria used to evaluate artificial intelligence industrial progress. Song Haitao, head of the Shanghai Artificial Intelligence Research Institute and member of the UN Independent International Scientific Panel on Artificial Intelligence, noted that evaluation methods have transformed. He observed that clients and markets now assess technical capacity with models rather than relying on rankings for self-proof, marking an advance toward industrialization. Legend Capital president Li Jiaqing stated that large models form an essential element across the end-to-end industrial chain, advising a portfolio approach spanning chips, infrastructure, models, and practical applications accompanied by pragmatic, long-term capital. ModelBest co-founder, chairman, and chief executive officer Li Dahai described the current era as a technological explosion, stating that model knowledge density doubles roughly every 100 days while edge models expanded from 1B to 35B running on mobile phones within a single year.