Will AI chat interfaces shift demand away from delivery apps? This is what Deepinder Goyal says

Bengaluru: Will AI chat interfaces become the new front door for commerce? Will it shift the demand away from apps such as Zomato? No, says Eternal founder Deepinder Goyal.

According to him, general-purpose interfaces are suitable for general-purpose queries, but they are inadequate for complex, high-frequency, and habitual transactions.

"Ordering food involves browsing menus, comparing restaurants, ratings and delivery times, substituting items and tracking orders in real time," he said.

Though according to Goyal, chat interfaces will not displace food delivery, there is significant disruption when it comes to the use of AI. Delivery platforms have evolved beyond basic automation and are now leveraging agentic commerce.

Understanding a user's behaviour, an AI agent can itself compare and order food instead of a user browsing for the best rates and options.

Recently, at an event in Bengaluru, Swiggy CTO Madhusudhan Rao said that soon agents will do all the searching for customers and order food according to their preferences.

Will AI assistants manage your food ordering in future? "AI assistants will start playing a role, but not fully replace user control. They can help with repeat orders, recommendations, and decision-making based on past behaviour. Over time, they may handle routine use cases like reordering meals or suggesting options. But food is still a personal and contextual choice, so users will continue to stay involved in the final decision," said Madhav Kasturia, Founder & CEO, Zippee, a quick commerce logistics startup.

Kasturia said AI is already being used across the stack, from demand prediction and inventory planning to routing and delivery efficiency. Going forward, it will improve personalisation, make operations more efficient, and help platforms manage supply and demand better.

"The biggest impact will be in reducing friction, making the entire experience faster, more predictable, and more reliable," he said.

Dravisha Katoch, Investor, Antler India, said AI is already core to food delivery. The visible layer is search, recommendations, and dynamic pricing.

The invisible layer is where the real work happens - rider supply-demand modelling at peak hours, geo-analytics on which restaurants to push discounts to, inventory and material management for cloud kitchens, fraud detection, ETA prediction.

Hyper-personalised menus - your homescreen rebuilt every time you open the app; voice and agent-led ordering; real-time supply matching - riders, restaurants, kitchens, all optimised live; and AI for SMB restaurants, are something that Katoch is watching out for.

"Food delivery is one of the cleanest examples of an industry where AI is the substrate. The companies that win the next decade will treat data and prediction as one of their core products," he said.

In a letter to shareholders, Goyal also explained that they are one of the largest deployers of AI in India. "Demand prediction, route optimisation, supply-chain management, customer experience, fraud detection, catalogue quality, and partner support already use AI across our platforms. These systems are trained on billions of real interactions, where the cost of getting something wrong — a late delivery, a wrong order, a failed refund, an unavailable item -- is immediate and measurable. But the more important question is not how AI helps us reduce costs. That is useful, but not enough. The sharper question we ask ourselves is: how does AI expand the market?" he said.

According to Goyal, for customers, conversational interfaces let them express intent in full sentences rather than two-word search queries. For delivery partners, AI will make onboarding, training, and day-to-day guidance conversational and personalised, thereby expanding the earning opportunity to people who may otherwise struggle with app flows, language, documentation, or lack of confidence.

"For restaurant partners and sellers, AI will make high-quality business guidance available from day one —the kind of growth-guidance that historically our teams could provide to a limited set of partners. AI lets us serve people and businesses that were previously too hard, too small, too fragmented, or too operationally expensive to serve well. Operational efficiency will follow, but it is not the ambition. Our moat remains the physical world. AI makes that moat wider by allowing more of India to participate in it," he explained.

In India, where food delivery and quick commerce are already part of daily urban life, the next shift could be from opening multiple apps to simply telling an assistant what you need. Global moves support this direction.

Alexa+ now lets users order from Uber Eats and Grubhub through natural conversation, while DoorDash and Instacart have brought grocery ordering into ChatGPT-style experiences. For India, the relevance is clear.

If users can order groceries or meals through AI in other markets, similar behaviour can easily fit into Indian use cases like monthly staples, office lunches, family meals, or diet-based ordering, said Somdutta Singh, Founder and CEO, Assiduus Global.

She said the future may not be fully hands-off for everyone, but ordering will become more guided, personal, and low-effort.