AI industry shifts toward conversational advertising and rigorous model evaluation

The artificial intelligence industry is seeing significant shifts in advertising, regulation, and technical evaluation. In the e-commerce sector, SHOPLINE has partnered with Criteo to launch a pilot program allowing brands to integrate with ChatGPT advertising. This move aims to capture conversational shopping traffic, with SHOPLINE reporting that global merchant orders from AI channels grew over 35 times in the first half of 2026 compared to the previous year.

Simultaneously, regulatory bodies in China are cracking down on Generative Engine Optimization (GEO) practices. Authorities in Shenzhen, Guangzhou, and Beijing have issued fines to service providers for using AI to fabricate fake rankings, industry ratings, and user satisfaction data to manipulate AI model responses. These actions target unfair competition and the spreading of misinformation within AI-driven search environments.

On the technical front, companies like Manfu Technology are moving away from simple parameter-based competition toward sophisticated data engineering. They are focusing on solving the “hidden shortcomings” of large language models—such as broken reasoning chains and tool-calling errors—by implementing process-oriented evaluations rather than just checking final output accuracy.