The prospect of asking AI to shop on our behalf no longer seems far-fetched, as technology companies large and small are making no secret of their ambitions in this space.
The growing use of AI for product search and shopping reflects a broader shift in how people search for information online. Photo: shutterstock.com.
A Recurring Need
In 1991, the internet was commercialized, websites became increasingly widespread, and their number exploded from just one to approximately 2.4 million within seven years. That was also when Google emerged to make finding websites easier.
The rise of artificial intelligence (AI), together with its rapid evolution, is now taking search to a new level. Users can describe the products they are looking for conversationally, set up recurring shopping lists, and even ask AI to complete payments on their behalf.
This capability is often referred to as a “virtual shopping assistant.” In 2025, according to Modern Retail, nearly every major retailer and established AI company in the U.S. had entered the race to develop AI-powered shopping assistants. Amazon expanded its Rufus shopping assistant with a new automatic purchasing feature. Target, Instacart, and DoorDash enabled users to shop directly through their services within ChatGPT, while Perplexity launched an AI-powered browser.
According to research conducted by OpenAI’s Economic Research team together with a Harvard economist, around 2% of all ChatGPT queries are related to shopping, including questions such as “recommend a laptop under $1,000” or “how much do running shoes cost?” That translates into roughly 50 million shopping-related queries per day. Even if only a small proportion of those interactions turn into transactions, the potential volume becomes enormous at scale.
The growing tendency to rely on AI for product search and shopping reflects a broader change in how people search the internet.
Traditional search engines operate primarily through keywords, meaning that longer and more detailed queries can often return results that are mechanical or poorly aligned with the user’s actual intent. Generative AI works differently. Users can provide much richer context through conversational interfaces.
People initially experimented with AI tools by asking questions designed to test their intelligence. Over time, however, they have increasingly turned to AI for actual advice and decision-making — from purchasing products and seeking health information to planning travel.
BrightEdge tracking data from June to August 2025 showed that AI-based search was growing at double-digit rates month over month. For example, Claude grew by 58% in July, Grok recorded a remarkable 1,279% increase, while ChatGPT emerged as the leading source of referral traffic among AI platforms.
Although AI search still accounts for less than 1% of total search traffic and Google continues to hold more than 90%, this marks the first time in three decades that Google has begun losing its near-absolute market dominance.
Gartner predicts that traditional search engine traffic will decline by 25% by 2026 as AI chatbots and other virtual assistants become increasingly popular.
“This will force companies to rethink their marketing channel strategies as GenAI becomes increasingly embedded across all aspects of business,” said Alan Antin, Vice President at Gartner.
Is Vietnam Ready?
In early 2026, Google announced Project Jarvis, which has since been integrated into Chrome under its official name, Project Mariner. The initiative aims to move AI from simply “thinking” to actually “taking action.”
Jarvis can directly control a web browser and perform complex, multi-step tasks on behalf of users. It can independently conduct research, manage shopping carts, and even arrange an entire travel itinerary with minimal human intervention.
Industry observers believe AI-powered shopping will soon become mainstream, supported by global technology companies with significant technical capabilities and financial resources. Given the rapid adoption of AI in recent years, the question is whether Vietnamese businesses should follow this trend.
Võ Quốc Hưng, Growth Director at Tonkin Media, believes that in e-commerce, the shift is inevitable. However, Vietnamese businesses still face several challenges, including how much they should invest and, more importantly, which technologies they should choose in an AI landscape that continues to change rapidly.
Against this backdrop, businesses are increasingly prioritizing investments that respond to clear changes in user behavior. One obvious example is SEO (Search Engine Optimization), which is already being disrupted as more people turn to AI tools for search.
Google itself has recognized this shift by introducing AI-generated summaries through AI Overviews in its search experience.
According to Hưng, this changing search behavior will significantly affect the retail and fast-moving consumer goods (FMCG) industries, where purchasing decisions occur frequently and repeatedly. As a result, traditional SEO will increasingly give way to optimization around entities and authority.
Businesses are therefore beginning to provide AI systems with clean, structured data that can be easily read and understood.
Instead of focusing solely on ranking No. 1 on Google to drive users to their websites, Vietnamese businesses are increasingly looking for ways to ensure that AI systems can understand, reference, and cite their content directly in AI-generated answers.
“Website traffic may decline, but conversion quality could improve because AI has already filtered the information and identified what is most relevant to the user,” Hưng said.
Meanwhile, Trung Nguyễn, founder of Advertising Vietnam, argues that AI Search or AEO (Answer Engine Optimization) remains fundamentally rooted in traditional SEO. Therefore, businesses should continue prioritizing strong SEO practices while adapting their websites to make them easier for AI systems to understand.
Generic content such as “Top 10” lists, “What is…” articles, and basic how-to guides will increasingly be replaced by AI-generated answers. Content based on real-world experience, proprietary data, expert perspectives, and detailed comparisons will become significantly more valuable.
AI agents capable of helping businesses close sales are also seen as an inevitable development, although AI is unlikely to completely replace human interaction in certain markets.
In Vietnam in particular, customers still prefer speaking with a real person when they need detailed consultation, help resolving problems, or greater reassurance before making a purchase.
For this reason, a hybrid model is likely to be the most suitable approach: AI handles basic questions and provides immediate responses, while human employees step in when customers require deeper assistance.
“This trend has already begun in Vietnam. For example, Meta has launched Business AI on Messenger to help businesses automatically respond to customers while still allowing conversations to be transferred to human staff when necessary. Over the next two to three years, AI agents are expected to become much more widely used across retail, e-commerce, education, travel, and customer service,” Trung Nguyễn said.
