Retail is on the brink of a digital sea change as agentic commerce slowly makes its way onto consumer-facing platforms.

By 2030, US B2C retail could see up to $1 trillion in orchestrated revenue from agentic commerce, according to a McKinsey report.

• For example, ChatGPT is now making it easier for consumers to query about retailer and brand offerings through new app integrations, opening up the potential for more business to shuttle through agentic commerce.

“With the large language model (LLM) providers coming front and center, there’s an opportunity to centralize everything through themselves,” said Andrei Papancea, CEO and cofounder of conversational technology company NLX.

Positioning for agentic commerce

Currently, Papancea said, the problem for retailers is that most AI technology on their sites, if any, is devoted to discovery and not lower-funnel transactions and purchases. Aiming for those lower-funnel purposes would require more digital maturity.

“So many organizations, when they’re talking about how they adopted AI, they mean they adopted ChatGPT internally, or [Microsoft] Copilot,” he said.

As Papancea sees it, most retailers are largely sitting on the sidelines while LLM providers set themselves up to own the full funnel from discovery to purchase.

Calling all AI assistants

For retailers who want to maintain as much sway as possible over their shoppers' journey, experts say the time is running out.

“There’s a very tiny window for brands to recapitalize and take control,” said Papancea.

The clock is ticking as consumers grow more comfortable shopping with AI tools, leading to a higher likelihood that they will rely on those tools in the future.

• Retail businesses saw a 128% monthly increase in actions on AI tools and assistants throughout the first half of 2025, according to an October Salesforce study.
• Consumers who regularly use AI agents are twice as likely to say their experience with retailers has improved, the same study found.

Across all industries, customer service is the top use case for AI agents, and sales comes in third, per Salesforce. The number of agents built and deployed by businesses grew by 119% in the first half of 2025.

Transactional engagement ups the ante

LLM providers like OpenAI have begun adding transactional components to their consumer interfaces. So, while more consumers are comfortable searching for information about products using AI, soon they’ll be able to buy more of those products that come up in the search results.

“If I’m a mom and pop shop and I want to sell my products through Uber Eats, then yes this will help bring my company and products to the masses,” said Papancea. “If I’m a brand, if I’m McDonald’s, I want to have control over my brand pulse. And while it’s certainly possible to do that through the LLM providers, it’s not going to work just out-of-the-box. It requires tailored engineering and thought.”

If retailers want to continue guiding the shopping experience for their customers, they need to start planning now, according to Papancea.

“It could be an assistant, but it can’t be just a traditional chatbot, a tiny thing in a corner of the site that’s just an afterthought," he said. "Every brand could get their own Gemini-like, ChatGPT-like interface. But then instead of just relying on whatever those companies provide, a generic experience with multiple purposes it needs to service, it could just be theirs.”

The change is retail commerce is coming, but hasn’t arrived yet. Traditional search still dominates, with generative AI search claiming only 3.3% of total time spent searching, according to an August 2025 Comscore data analysed by EMARKETER.

This was originally featured in the Retail Daily newsletter. For more retail insights, statistics, and trends, subscribe here.