Where has AI-powered e-commerce reached?

```

We are at a critical turning point in e-commerce—the dawn of the so-called “Agent Commerce” era. Morgan Stanley believes this trend is not empty hype but is turning into real transaction data. Although adoption seems slow at present, the underlying logic is being reconstructed.

According to Wind Trading Desk, Morgan Stanley’s latest research report on February 1st points out that for investors, the most important number is this: by 2030, the total gross merchandise value (GMV) of agent commerce is expected to reach about $190 billion (base case) to $385 billion (optimistic case). In other words, by then, 10% to 20% of the US e-commerce market will be driven by AI agents.

The current market landscape shows a “dual-track” competition: on one side are general-purpose large-model platforms represented by ChatGPT and Gemini, boasting huge traffic and young user bases; on the other are retail giants like Amazon and Walmart launching dedicated AI tools, which, though having smaller user bases, have extremely high conversion rates. For targets like Google, Amazon, Meta, and Walmart, whoever can gain mindshare and payment entry in this early stage will secure enormous excess returns over the next five years.

Large-model platform penetration approaches 50%—young users dominate

The report states that general AI platform adoption is steadily climbing. As of January 2026, ChatGPT and Gemini’s monthly usage rates reached 47% and 33% respectively, a slight increase from 45% and 32% in October 2025. While the growth isn’t explosive, it demonstrates an irreversible trend.

Generation gap is even more critical. ChatGPT dominates among younger users, with usage rates among those aged 16-34 exceeding 65%-75%. By contrast, Gemini’s growth mainly comes from the 35-44 age group. This “age gap” is a clear strategic signal to Google: If it does not quickly bridge its product gap among younger users, its user base may face loss in the coming years.

From “chat” to “transaction”: surprising conversion rates for commercial behavior

What investors care about most: Are people really shopping with AI? The report’s data gives a definite answer.

Current commercial behavior is highly stable. About 40%-55% of AI platform users are using these tools for product research and price comparison. More importantly, about 30%-40% of agent users ultimately made purchases.

This is an extremely explosive early signal. If we extrapolate this ratio to the entire US population, it means that just last month, about 18% of Americans shopped via ChatGPT, and about 10% via Gemini. Given that current product forms are still at a preliminary stage, this early purchase penetration shows immense future growth potential.

Retailers strike back: small base, high conversion

Although general large models dominate traffic, retail giants are not sitting idly by. Walmart’s Sparky, Amazon’s Rufus, and Target’s shopping assistant have monthly user penetration rates in the low double digits, far below general-purpose platforms, but their users have very strong purchasing intent.

Research data shows that purchase conversion for retailers’ dedicated agent tools matches or exceeds general platforms. This shows that for Amazon and Walmart, the opportunity lies in leveraging their vast existing user base—using AI tools to increase core users’ shopping frequency and spending. As long as they can drive agent tool adoption, they can take back the initiative from general platforms.

Category landscape: grocery and FMCG are the “Trojan Horse”

Looking at specific consumer categories, high-frequency items are becoming the breakthrough point for AI e-commerce. Data shows that among users who shop via AI platforms, 50% bought groceries, a proportion that has stayed stable since last October. Household goods and personal care items come next.

This is a crucial strategic pivot. Morgan Stanley believes that grocery will be agent commerce’s biggest unlocking category in the next five years. Whoever captures the high-frequency grocery entry point will capture user habits.

Risk Warning and DisclaimerThe market carries risks; investment requires caution. This article does not constitute personal investment advice, nor does it take into account any individual user’s particular investment goals, financial situation, or needs. Users should consider whether any opinions, views, or conclusions in this article fit their specific circumstances. Investments made on this basis are at your own risk. ```