Can AI agents disrupt Zomato and Blinkit? Deepinder Goyal explains why their business models may be protected
AI agents could change how consumers discover products and place orders, raising questions about the future of platforms such as Zomato and Blinkit. Deepinder Goyal argues that their fulfilment infrastructure, real-time coordination and accountability could protect them from disruption, even as AI reshapes customer acquisition.
- Oct 21, 2026,
- Updated Oct 21, 2026 4:30 PM IST
AI agents could change how consumers discover products and place orders, but Deepinder Goyal believes food delivery and quick-commerce platforms such as Zomato and Blinkit are relatively well-positioned to withstand the disruption. In Eternal's Q2 report, Goyal argued that their competitive advantage lies not just in their apps, but in the complex fulfilment infrastructure that turns digital orders into real-world deliveries.
The distinction could prove important as artificial intelligence increasingly enables users to delegate shopping, booking and other transactions to automated assistants. For internet companies, the central question is whether these agents will replace existing platforms or simply become another channel through which customers access them.
Why Zomato and Blinkit may be less vulnerable
Goyal argued that food delivery is not simply a transaction that an AI agent can complete by selecting a restaurant and placing an order. Zomato's value also comes from coordinating restaurant kitchens, delivery partners and customer demand in real time.
Blinkit's quick-commerce business depends on a similarly complex operating model, including dark stores, real-time inventory management, picking and dispatching orders within a tight delivery window.
An AI agent could potentially initiate an order, but it would not automatically solve the operational challenges involved in fulfilling it. Someone must ensure that the product is available, the order is prepared or picked, a delivery partner is assigned and the customer receives it as promised.
In Goyal's assessment, platforms that control this infrastructure retain an important role even if consumers increasingly interact with AI agents rather than opening their apps directly.
MUST READ: After Temple, Deepinder Goyal teases Dark.ai: A bet beyond food delivery services
Food discovery is more than a transaction
Goyal also challenged the assumption that consumers will routinely hand over food choices to AI assistants. According to him, more than 65% of Zomato app opens are exploratory, with users browsing an average of 12 menus and more than 100 food images before ordering.
This behaviour suggests that food ordering is not always a repetitive task that consumers want to automate. Browsing restaurants, comparing dishes and discovering new options can be part of the experience itself.
AI agents could nevertheless capture some routine transactions, such as reordering familiar meals or restocking frequently purchased essentials. Goyal says these reorder carts account for a low-single-digit share of activity on Zomato and Blinkit, respectively, and argues that in-app assistants could also cater to such requirements.
The bigger opportunity for AI may therefore be to make existing platforms easier to use rather than eliminate the need for them.
ALSO READ: OpenAI's AI agents disabled safeguards meant to monitor them. One allegedly lied about it
Why fulfilment could matter more than the interface
The potential disruption is not uniform across consumer internet businesses. Goyal noted greater vulnerability in District, where restaurant discovery, reservations and event ticketing can involve inventory that also exists independently of the platform.
If an AI agent can find an available restaurant table or ticket through another channel, it could potentially bypass the discovery platform. Content-driven advantages, including reviews, ratings and curated recommendations, may also become less durable as AI changes how consumers search for information.
Food delivery and quick commerce, by contrast, depend heavily on physical operations and coordination. Goyal points to the challenges of disaggregating discovery, ordering and fulfilment, citing ONDC as an example of how fragmented systems can create latency, accountability gaps and service failures.
DO READ: Is AI squeezing entry-level tech jobs? Anthropic’s junior technical hiring drops from 27% to 9%
AI could change customer acquisition, not eliminate platforms
The emergence of agents could still reshape competition. If consumers delegate decisions to AI assistants, transactions may increasingly flow towards whichever platform offers the best combination of price, availability and fulfilment. That could weaken established user habits and intensify competition.
However, Goyal noted that platforms may continue to earn from commerce even when they no longer own the primary customer interface, provided they remain the infrastructure powering transactions.
For Zomato and Blinkit, the strategic priority may be to make their fulfilment networks accessible to new interfaces while maintaining reliability and operational efficiency.
The broader lesson is that AI could erode advantages built primarily on content or consumer attention while strengthening the position of companies with difficult-to-replicate physical infrastructure. But Goyal's assessment is not a guarantee of immunity: as AI agents evolve, the balance between customer ownership and fulfilment capability will determine which platforms retain their competitive edge.
MUST READ: Major security flaw triggers action from Google and Apple against eMudhra certificates
For Unparalleled coverage of India's Businesses and Economy – Subscribe to Business Today Magazine
AI agents could change how consumers discover products and place orders, but Deepinder Goyal believes food delivery and quick-commerce platforms such as Zomato and Blinkit are relatively well-positioned to withstand the disruption. In Eternal's Q2 report, Goyal argued that their competitive advantage lies not just in their apps, but in the complex fulfilment infrastructure that turns digital orders into real-world deliveries.
The distinction could prove important as artificial intelligence increasingly enables users to delegate shopping, booking and other transactions to automated assistants. For internet companies, the central question is whether these agents will replace existing platforms or simply become another channel through which customers access them.
Why Zomato and Blinkit may be less vulnerable
Goyal argued that food delivery is not simply a transaction that an AI agent can complete by selecting a restaurant and placing an order. Zomato's value also comes from coordinating restaurant kitchens, delivery partners and customer demand in real time.
Blinkit's quick-commerce business depends on a similarly complex operating model, including dark stores, real-time inventory management, picking and dispatching orders within a tight delivery window.
An AI agent could potentially initiate an order, but it would not automatically solve the operational challenges involved in fulfilling it. Someone must ensure that the product is available, the order is prepared or picked, a delivery partner is assigned and the customer receives it as promised.
In Goyal's assessment, platforms that control this infrastructure retain an important role even if consumers increasingly interact with AI agents rather than opening their apps directly.
MUST READ: After Temple, Deepinder Goyal teases Dark.ai: A bet beyond food delivery services
Food discovery is more than a transaction
Goyal also challenged the assumption that consumers will routinely hand over food choices to AI assistants. According to him, more than 65% of Zomato app opens are exploratory, with users browsing an average of 12 menus and more than 100 food images before ordering.
This behaviour suggests that food ordering is not always a repetitive task that consumers want to automate. Browsing restaurants, comparing dishes and discovering new options can be part of the experience itself.
AI agents could nevertheless capture some routine transactions, such as reordering familiar meals or restocking frequently purchased essentials. Goyal says these reorder carts account for a low-single-digit share of activity on Zomato and Blinkit, respectively, and argues that in-app assistants could also cater to such requirements.
The bigger opportunity for AI may therefore be to make existing platforms easier to use rather than eliminate the need for them.
ALSO READ: OpenAI's AI agents disabled safeguards meant to monitor them. One allegedly lied about it
Why fulfilment could matter more than the interface
The potential disruption is not uniform across consumer internet businesses. Goyal noted greater vulnerability in District, where restaurant discovery, reservations and event ticketing can involve inventory that also exists independently of the platform.
If an AI agent can find an available restaurant table or ticket through another channel, it could potentially bypass the discovery platform. Content-driven advantages, including reviews, ratings and curated recommendations, may also become less durable as AI changes how consumers search for information.
Food delivery and quick commerce, by contrast, depend heavily on physical operations and coordination. Goyal points to the challenges of disaggregating discovery, ordering and fulfilment, citing ONDC as an example of how fragmented systems can create latency, accountability gaps and service failures.
DO READ: Is AI squeezing entry-level tech jobs? Anthropic’s junior technical hiring drops from 27% to 9%
AI could change customer acquisition, not eliminate platforms
The emergence of agents could still reshape competition. If consumers delegate decisions to AI assistants, transactions may increasingly flow towards whichever platform offers the best combination of price, availability and fulfilment. That could weaken established user habits and intensify competition.
However, Goyal noted that platforms may continue to earn from commerce even when they no longer own the primary customer interface, provided they remain the infrastructure powering transactions.
For Zomato and Blinkit, the strategic priority may be to make their fulfilment networks accessible to new interfaces while maintaining reliability and operational efficiency.
The broader lesson is that AI could erode advantages built primarily on content or consumer attention while strengthening the position of companies with difficult-to-replicate physical infrastructure. But Goyal's assessment is not a guarantee of immunity: as AI agents evolve, the balance between customer ownership and fulfilment capability will determine which platforms retain their competitive edge.
MUST READ: Major security flaw triggers action from Google and Apple against eMudhra certificates
For Unparalleled coverage of India's Businesses and Economy – Subscribe to Business Today Magazine