China's e-commerce export engine is experiencing significant challenges as low-cost e-commerce exports fell 10.9% in April to $9.81 billion, marking the fifth consecutive month of declines compared to a year ago, according to analysis by Luxembourg-based consultancy Trade and Transport Group. The business models, based on flying $5 dresses from Chinese factories to shoppers around the world, were already under pressure after US President Donald Trump introduced tariffs and axed customs waivers on low-value parcels last year. Soaring logistics costs stemming from the West Asia conflict are adding to the strain, with shippers like DHL Express imposing hefty fuel surcharges. Diana Qiao, a Shenzhen-based seller of women's clothing on Temu, said she had raised her selling prices by $2 because her shipping cost per garment had increased on average by $1, with the final burden ultimately borne by consumers.
A new invisible middleman is taking over digital storefronts, threatening to strip e-commerce giants of their data, dollars, and dominance. ChatGPT, operated by Open AI, allows users to connect various applications including Innovist (personal care), MakeMyTrip (travel), Zomato (food delivery), Cars24 (pre-owned cars), and Square Yards (real estate). When users type queries like "Help me find a shampoo for wavy hair that can control frizz for under ₹1,000," AI chatbots respond with product widgets and explain why recommended items suit their needs. Harshil Mathur, Razorpay's co-founder, expects the top 1,000 D2C brands to begin integrating AI-led commerce journeys over the next six-eight months, with the company currently having around 15–20 active pilots where agentic commerce has been rolled out. High-frequency categories such as food delivery, groceries and quick commerce are where conversational commerce could scale first, with users showing repeat behavior for specific use cases.
According to the Gustavson School of Business research, online shopping platforms rely heavily on consumers spending time browsing websites, clicking ads and comparing products, generating billions of dollars through advertising, subscriptions, commissions, sponsored placements and transaction fees. However, "agentic AI" could reshape online commerce by allowing AI to shop on behalf of consumers, fundamentally disrupting this revenue model. As Yuanyuan Gina Cui explains, "When you think about the way AI is already being used to authorize payments, it's already affecting our wallets. It really points to a major shift away from the platform revenue stream." The researchers warn that "agentic AI agents will be able to draw data together from everything we do, and as a result, the filters will be much clearer and will be much more based on personalized shopping rather than segment-oriented shopping," potentially reducing the need for traditional platform engagement. Recent data shows up to 45% of shoppers already use generative AI tools to compare products while searching online, with a significant chunk completing purchases with agentic assistance.
McKinsey estimates that by 2030, agentic commerce could influence $3–5 trillion in global commerce value, with the US B2C retail market alone potentially accounting for nearly $1 trillion. These AI agents can browse, compare, and initiate purchases based on user's goals, preferences, and limits, potentially replacing much of the manual work consumers currently do. Amazon has introduced AWS Agentic Shopping Assistant, with Kate Spade already using the solution to build an AI gift concierge, while other retailers are testing the platform. The shift represents a fundamental change in how retailers compete, as shoppers may rely on AI tools to decide purchases rather than traditional search results, social media ads, and loyalty programs. This could force retailers to optimize not only for Google search and social media algorithms, but also for AI recommendations, potentially changing how retailers approach loyalty programs.
The convergence between digital commerce and physical experiences is accelerating, with stores functioning as brand showrooms, customer acquisition channels and experience centres where transactions may ultimately occur online. Placer.ai's retail outlook found that 44% expect agentic AI to increase the share of online retail, while 34% said it could drive broader growth across commerce overall. The competition extends beyond who completes transactions to who influences purchase decisions in the first place. GCC brand directors face a new strategic imperative: securing space within flagship Vision 2030-backed developments is becoming critical as occupancy rates in Riyadh's leading lifestyle destinations approach full capacity. Brands that establish early presence in premium developments stand to benefit from higher footfall, stronger brand visibility and association with Saudi Arabia's most ambitious urban projects, while those that delay risk being pushed into secondary retail locations with weaker customer engagement opportunities. As Jan Kietzmann notes, "The future depends on what the leaders of these platforms decide they want to do about it. Initially, there might be some big court cases challenging the technology. Others may try to build their own agents. The big ones, with deep pockets, might be able to survive a little longer, but eventually, everybody will need to figure out how to optimize for agentic clients rather than human clients."
