In this blog series, we’re sitting down with top Algolia voices to explore the innovations transforming commerce, consumer expectations shaping industries, and the technologies redefining shopping in the age of AI.

For this installment, we spoke with Bharat Guruprakash, Algolia’s Chief Product Officer, about how AI agents and other trends he’s watching will evolve in 2026, what technology executives should be prioritizing in the new year, and how Algolia views the future of retail.

Q: Aside from agentic AI, what were some of the top retail tech trends you saw last year that may continue into this year?

A: Agentic AI is the headline, but within that, agentic commerce is the big story. We saw Stripe and OpenAI launch the Agent Commerce Protocol, and Shopify integrate directly into OpenAI. These moves highlight a shift: shopping and discovery are happening off-property, in places where consumers already spend time, like OpenAI or Perplexity.

Another major development is AI companies launching their own browsers. They’re trying to be where consumers enter the internet. I expect this trend to carry forward, with more consumers aggregating into spaces where they can automatically reach into LLMs. Retailers will need to figure out how to show up inside these environments.

Q: Do you think consumers are ready for AI browsers?

A: It’s still very early. The UX isn’t there yet. The speed is slow, and watching an agent type out fields can be frustrating. Security is also a huge concern. We don’t yet know what kinds of injections could happen when agents operate inside a browser. So while the potential is big, we’re still in the early stages.

Prior to ChatGPT announcing its Atlas browser, Algolia actually asked consumers whether they ever purchase items directly through generative AI platforms. Only 22% responded in the affirmative, suggesting the potential is there but needs to be nurtured before it hits the mainstream.

Q: What trends didn’t live up to expectations this past year?

A: Two come to mind. First, retrieval-augmented generation (RAG) was huge early in the year but quickly faded. Second, the rush to put chatbots into every application. Companies realized the UX wasn’t great, and those efforts didn’t really take off.

Q: If 2025 was the year of agentic AI everywhere, what will 2026 be?

A: I think 2026 will be the year of memory and scale. Most agentic AI applications today are prototypes or proofs of concept that don’t scale. The only ones with real product-market fit are coding apps. For agents to be useful, they need memory; otherwise, they’re just goldfish. And enterprises need to figure out how to deploy at scale and see ROI.

I also expect retrieval to become a hot topic. As Aaron Levie from Box recently posted on LinkedIn, many AI agent problems are really just information retrieval problems. If agents can find and comb through data more effectively, results will improve dramatically. That’s encouraging for Algolia, since retrieval is our core strength.

Q: What are the smartest investments tech leaders can make heading into 2026?

A: I would divide these investments into two main buckets: improving operational efficiency and boosting employee productivity. That applies across go-to-market, product, and engineering teams. Leaders should lean into finding where AI can drive those gains.

Another smart bet is hiring more junior AI engineers. They bring a fresh mindset compared to established cohorts, and they can challenge existing processes and assumptions. Yes, it requires cultural adaptation, but the benefits of absorbing that new perspective are significant.

Q: Has agentic AI lived up to expectations in retail this year?

A: Absolutely not. Just ask the average shopper if their experience changed dramatically. Mine sure hasn’t. 

Ecommerce is a tough nut to crack for agentic experiences. It’s dynamic, with many interconnected pieces—from inventory to checkout to returns. Retailers have built static, rules-driven workflows for years, and layering agentic AI on top feels like a bolt-on rather than a fresh experience.

Even Amazon’s Rufus hasn’t dramatically changed the shopping experience, despite claims of big revenue impact. I think off-property experiences will be more interesting, but they also challenge retailers’ business models. If ChatGPT recommends the top five items to buy, suddenly the retailer’s website isn’t the primary shopping destination anymore.

Q: What do you make of Instacart’s new agentic AI tools for grocers?

A: It looks like Instacart is focusing more on the retailer side—helping grocers with operational efficiency rather than consumer-facing experiences. That’s smart, because every commerce vertical is different. Grocery dynamics are very different from fashion, for example. This kind of investment in operations and efficiency is exactly where I expect AI to have the most impact in the near term.

Q: How do you see agentic AI impacting brand loyalty if transactions move off-property into AI browsers?

A: We’ve seen this before with new distribution channels—Amazon, Google Shopping, and others. AI browsers are just another channel. The difference is that they shift the point of consumer entry. Retailers will need to rethink how they build loyalty when the shopping journey starts outside their own properties.

Closing thoughts

Agentic AI may continue to dominate headlines, and rightfully so, but the next phase of innovation will be defined by memory, retrieval and scale – without which enterprises will never be able to fully realize the benefits of AI agents. For retailers and tech leaders, the challenge is not just experimenting with AI, but deploying it in ways that truly improve efficiency, productivity, and consumer experience.

Want to learn more about AI Search? Schedule a time to speak with our team in the new year, or signup for a free account today.