Agentic AI: the next battleground for Chinese brands

Chinese brands must adapt to the rise of agentic AI, which is increasingly shaping consumer decisions by autonomously curating recommendations based on personal preferences. This shift challenges traditional brand influence, as AI agents now determine what consumers discover and buy, requiring new strategies beyond social media and search optimization. Security risks, cultural biases, and misinformation pose additional challenges.

Agentic AI: the next battleground for Chinese brands To unlock further success in brand awareness, Chinese firms must appeal to the artificial intelligence agents that are increasingly shaping consumer experience artificial intelligence AI agents https://www.scmp.com/tech/tech-trends/article/3353747/chinas-tech-companies-are-looking-rewrite-e-commerce-playbook-ai-agents?module=inline&pgtype=article that will increasingly shape what consumers discover, consider and ultimately buy. These personal concierges are starting to determine the best choices for each user by balancing price, quality, reliability, sustainability and individual preferences. Tell an AI agent you want to go “somewhere”, and it can autonomously choose the destination and book transport, lodging and restaurants based on its deep, constantly updated understanding of your tastes, values and lifestyle – so-called personality pairing. That’s a quantum leap from generative AI. Yet, AI of both the generative and agentic variety share a common dynamic: commerce is embedded into every aspect of each user’s experience. Weary of laborious click-throughs, consumers are seeking curated recommendations from AI agents. As a consequence, brands are losing influence over the answer that shapes choice, speakers argued at the recent Cannes Lions festival in France, attended by marketing executives and business leaders. Brand identity is drifting from what companies intend to what AI agents generate and reinforce – often without marketers’ awareness. The central challenge is how to regain control. Strategies and tools for agentic AI are vastly different from those for social media engagement and search engine optimisation. Security risks https://www.scmp.com/tech/tech-trends/article/3346513/inside-openclaw-mania-gripping-china-security-fears-surge-alongside-enthusiasm?module=inline&pgtype=article range from agents’ privilege and authority to vulnerabilities for would-be attackers. Agents may bypass privacy safeguards when accessing and sharing personal data to customise actions. They can cause harm unintentionally, just as humans can. cultural biases https://www.scmp.com/opinion/world-opinion/article/3259465/we-must-decolonise-ai-overcome-cultural-bias-classroom?module=inline&pgtype=article pose problems that some believe are insurmountable. Governance frameworks may lack ways to counter misinformation https://www.scmp.com/news/world/united-states-canada/article/3273324/five-us-states-push-elon-musk-fix-ai-chatbot-over-election-misinformation?module=inline&pgtype=article , address biases or prevent transactional hijacking by agents.