From factory hunt to chatbot prompt
Artificial intelligence is moving deeper into the mechanics of commerce, and one of the clearest examples is product sourcing. MIT Technology Review reports that US-based small online sellers are using tools such as Alibaba's Accio to compress what used to take weeks or months of supplier research into a far shorter process built around conversational prompts.
That shift matters because sourcing has traditionally been one of the slowest and least transparent parts of running a small e-commerce business. Entrepreneurs often had to search through factory listings, compare suppliers manually, send inquiries, negotiate costs, and test whether a product idea could work before committing time and money. AI is now being inserted into that sequence as a front-end decision tool.
A case study in compressed decision-making
The source centers on Mike McClary, an Illinois entrepreneur who previously sold a flashlight called the Guardian LTE through his outdoor brand. When he decided to revisit the product in 2025, he did not begin with the usual directory and outreach process. Instead, he used Accio, describing the flashlight's original design, production cost, and profit margin.
According to the report, the tool suggested several changes to the product, including making it smaller, slightly less bright, and battery-powered rather than rechargeable. It also identified a manufacturer in Ningbo, China, that McClary said could reduce manufacturing costs from about $17 per unit to roughly $2.50.
McClary then took over the process himself, contacting the supplier to discuss the revised design. Within a month, the new version of the flashlight was back on sale through Amazon and his own website. That timeline is the real headline. The AI system did not manufacture the product or close the deal, but it appears to have condensed the search and comparison stage that often slows small sellers down.
Why small merchants care
For large companies, sourcing has long involved teams, established supplier networks, and repeatable procurement processes. Small merchants rarely have those advantages. Many rely on speed, improvisation, and a willingness to test products quickly. The article describes that approach as being scrappy: spot demand, adapt an existing design, find a factory, market modestly, and move fast.
AI tools are now being layered onto that operating model. Instead of replacing the entrepreneur, they appear to be lowering the cost of exploration. Sellers can ask for product revisions, margin scenarios, or manufacturing options in a single conversation rather than searching across many fragmented sources.
That has consequences beyond convenience. Faster research cycles can change what kinds of businesses are viable, who is able to enter a category, and how quickly a seller can respond to changing demand. It also increases the speed at which product ideas can move from concept to market, especially in sectors where differentiation comes from pricing, packaging, or incremental design changes rather than deep technical invention.
Alibaba's role and the new sourcing interface
MIT Technology Review notes that Alibaba is best known globally for other parts of its commerce empire, but Alibaba.com was originally built as a marketplace for Chinese factories open to bulk orders. That business is now being updated with AI tools designed to guide users through product and supplier discovery.
The significance here is not just that AI is being added to an existing marketplace. It is that marketplaces are becoming advisory systems. The interface is shifting from lists of suppliers to chat-based workflows that suggest designs, identify manufacturers, and highlight cost opportunities. That turns sourcing from a search problem into a recommendation problem.
For small merchants, that can be a major advantage. For suppliers and marketplaces, it may become a competitive necessity. The businesses that can reduce friction earliest are likely to capture more sellers, more experiments, and more transactions.
The limits of automation
The report also makes clear that AI is not removing human judgment from the process. McClary still contacted the supplier himself and handled the next stage directly. That distinction matters. Sourcing remains a real-world activity involving trust, negotiation, specifications, and execution risk. AI can accelerate the funnel, but it does not eliminate due diligence.
Even so, the balance of effort is changing. If entrepreneurs can now move through ideation, product refinement, and factory discovery in a single tool, then the biggest bottleneck may no longer be finding options. It may be choosing among too many of them and validating which ones are real.
That is a meaningful change in how small business gets built online. AI is not just helping sellers write ads or generate images. It is beginning to shape the physical products they decide to make in the first place.
This article is based on reporting by MIT Technology Review. Read the original article.
