When ali baba announced on may 11, 2026 that he was fully connected to the big questions model and the treasure, many saw a “tech + scene” performance. (blockview://markdown-image-tos-cn-i-tt/87ced6d78d68d69838381c9b1cd91058cb8) however, dismantling its commercial inner core is not a simple product upgrade, but ali has found a clear major commercial battleground for a costly model whose central theme is:** how to transform ai's technological potential into a quantifiable and sustainable increase in power-trading. It's a commercial landing at the heart of the "callboard" before that, thousands of people had been connected to such events as the take-out, the liquor brigade, but those were more like “trainers”. As a source of revenue for the ali core, **951 million live monthly users** and **4 billion commodity banks** provide the most commercially valuable landing soil for large models. The depth of integration with poaching marks the move of the commercialization of ai from “exploring scenes” to “harvest scenes”. This access was not a simple interface, but rather achieved a closure of the whole-process transaction** from demand understanding, commodity matching to billing, performance sale. Users can complete the purchase only by describing their needs in thousands of questions to app, and order information is synchronized in real time to treasure hunting. This is essentially an upgrade from a “dialogue tool” to a “new consumption portal” that directly drives transactions. Its commercialization path is clear:** by increasing the efficiency and experience of shopping, it stimulates more consumer behaviour, which is ultimately reflected in the treasure-hunting gmv and income growth. Behind the 40 per cent increase in efficiency was the official 20-year disclosure of commercial barriers to electricity company data, the increase in decision-making efficiency by more than 40 per cent from traditional search methods** by the purchase of ai, the average response time was less than 1. 2 seconds, and the intended identification accuracy rate was **98. 7 per cent**. This data-driven efficiency gains are key to establishing business models. Its core support is the 20-year accumulation of real shopping scene data** combined with the depth of the big questions model. This constitutes a short-term competition barrier that is difficult to replicate: - ** understand complex needs**: users describe “buying softer sneakers with large v-beds and gtx waterproofs, colours, shoelaces,” and ai is able to screen commodities that meet six conditions at once. - ** paradoxical fuzzy intense**: ai can extract live football 6 with precision and provide a link to the purchase simply on the basis of the memory of the fragments of “the cover of adriano's football game”. - **providing scenario**: for “what should expect pregnant mothers buy”, ai may recommend a combination of commodities such as pre-natal kits in conjunction with later pregnancy conditions. This precision referral capability, which is based on big data, directly reduces the cost of decision-making by users, transforms “walking” random consumption, in part, into “questioning” end consumption, and increases the efficiency of transaction conversion. # industry competition focus: the shift from “search ranking” to “ai recommended eligibility” is reconfiguring the rules for distribution of traffic in the electrician sector. Future competition is no longer “who can appear on the first three pages of the search results”, but rather “** who can be the standard option in the ai recommended answer**”. This has led directly to a shift in brand competitiveness strategy - from seo (search engine optimization) to geo (generation engine optimization). Brand needs: - optimizing structured data (parameters, scenes, etc.) for commodities to be “readable” accurately by ai. - accumulation of high-quality positive content in small red books and knowledge platforms as a “credible source” for ai. - actively covering the usual shopping and question-and-answer scenes of users and seizing the list of nominees for ai. In the face of ali's move, the competition has developed a differentiated layout: - **kingdong**: the independent application of the "kyotoai purchase" was launched by the end of 2025, with the main hit on the non-romanized primary ai shopping entrance. - **byte bean pack**: the peak of dau was **145 million in spring 2026,** with direct placing of a single function within the internal measurement application, following an open ecological route. (blockview://markdown-image-tos-cn-i-t/933a53ee917043f791a59a9111b998c28)) - ** 100-degree statement**: realization of the “search+ lead purchase” partial closure, but lack of capacity to pay on-site. Ali's path is closed to ecology, data is not drained and experiences are consistent, provided that users remain within the ali system. ** 618 in 2026 will be the first key window to test the operational capability of the ai electrician model.** # ari's strategic value: the consolidation of the fundamentals and the construction of the next generation of infrastructure, for ali baba as a whole, is at two levels: ** first, the consolidation of the electrician's basics to deal with the diversion of flows. ** in stock market competition, traditional search flows are being encroached on content platforms such as tremors and little red books. Through ai's new experience of “dialogue is trade”, ali aims to lock users more firmly into their own ecosystems and to enhance user stickiness and life-cycle values. ** second, validate commercial returns on large model inputs and build up the next generation of business infrastructure. ** the calculator cost of training the hundreds of billion-scale parameter models is high, and capital markets have always been concerned with the return on the huge ai inputs. Embedded deep into the core of the treasure hunt, the “cash cow”, is a crucial step in validating the viability of the commercialization of ai. In the long run, ali is trying to create an ai hub that integrates ecological capabilities such as poaching, paying, flying pigs, and promotes the transformation of companies from an “electrician platform” to an “**ai-driven consumer service provider**”. **conclusion: ** queries and poaching are a key bet for ali baba to scale up the al technology. It stimulates gmv growth in the short term by improving the efficiency of its purchases, consolidates industry in the medium term by reconfiguring the rules governing the distribution of flows, and serves the strategic goal of a group-driven transition to ai in the long term. Its commercial nature is fueled by 20-year-old sunken electrician data, driving this “ai engine”, which ultimately realizes its value in the largest “bankboard” of treasure hunting。
