Southeast Asia’s sellers spent a decade learning to sell to people. Now they have to sell to AI

For two decades, selling online in Southeast Asia meant captivating a person's attention through eye-catching photos, fair prices, and positive reviews that shoppers could scrutinize. However, the landscape is now shifting to cater to an algorithm's judgment. With the advent of AI agents, consumers can now ask ChatGPT to find "a good baby stroller under SG$150 (US$117)," and the agent handles the tedious task of reading listings, comparing options, and applying built-in criteria.

The only winner is the one chosen by the agent, which is then presented to the human shopper. Consumers seem to be ready for this handoff, particularly in categories like grocery shopping (41% of shoppers) and household supplies (31% of shoppers), which are the mainstay for countless small Shopee and Lazada sellers. Trust, on the other hand, drops significantly for higher-stakes categories like financial services (15% of shoppers).

A June 2026 study by Checkout.com revealed that 89% of merchants are already preparing for agentic commerce, despite the fact that only 3% of transactions currently involve an AI agent. The region's small sellers have spent a decade learning to sell to people and now must learn to sell to AI agents, a shift that presents both opportunities and challenges.

One major concern is that the decision logic of AI agents remains opaque and uncontestable, leaving small sellers without the ability to contest or understand why their shop is being recommended or not recommended. The global AI shopping assistant market is projected to grow exponentially in the coming years, with Southeast Asia already accounting for 90% of purchase decisions being influenced by AI recommendations.

While some platforms are taking steps to help sellers, such as partnering with OpenAI for ChatGPT trials, the focus remains on assisting sellers rather than providing transparency and control to the agents themselves. This lack of visibility and accountability poses a risk to the livelihoods of small sellers who now depend on these AI agents.

The region's AI investment boom has not been distributed equitably, with most funds flowing into infrastructure and foundation models rather than tools that would make these systems more transparent and understandable for small sellers. As AI agents continue to take over the roles traditionally held by humans, it is crucial that the region's sellers are given a fair shot at understanding and challenging the decisions made by these algorithms.