CEOInterviews.AI
Start App
Rajesh Magow
Co-Founder & Group Chief Executive Officer, MakeMyTrip Limited

Will AI Replace MakeMyTrip? The Founders Respond 🤖 #nikhilkamath #makemytrip #tajhotels #travel

📅 Sep 09, 2026 Bharat Breakout 8 MIN 12 VIEWS 70 SEGMENTS · 3 SPEAKERS
Could ChatGPT and Artificial Intelligence completely change the way we plan and book our travel? In this conversation, Nikhil Kamath, Co-founder of Zerodha, is joined by Rajesh Magow, Co-founder of MakeMyTrip, and Puneet Chhatwal, MD & CEO of IHCL (Taj Hotels) to discuss how AI could transform the travel and hospitality industry. The discussion explores the role of ChatGPT, OpenAI, Large Language Models and AI-powered travel planning, along with the future of personalised recommendations and online booking platforms. As travellers increasingly use AI to research destinations and plan their...

What Rajesh Magow said

Written from the verified transcript and checked against it. Every figure links to the moment it was said.

Rajesh Magow, Co-Founder and Group CEO of MakeMyTrip, discussed AI's impact on travel. He said AI is huge but traffic from ChatGPT is minimal, citing their presence on the platform via '@MakeMyTrip'. He explained they block scrapers but focus on protecting deep funnel data like user preferences and drop-off, not superficial data. Magow confirmed using frontier models like Gemini, Claude, Cursor, and GPT, grounded in proprietary data, with plans to possibly build own models due to token costs. He argued replication is hard in deep transaction businesses due to payments, fulfillment, and after-sales, citing Google's failure to go deep and Flipkart's independent Cleartrip. He acknowledged AI is a powerful revolution and never say never, while a co-panelist added that AI is an enabler and their business is AI-proof, emphasizing real human connections.

Key takeaways

  1. Traffic from ChatGPT via '@MakeMyTrip' is minimal; it's just taking time.
  2. MakeMyTrip uses frontier models like Gemini, Claude, Cursor, GPT, grounding them with proprietary data.
  3. Token costs of big LLMs are prohibitive, possibly pushing MakeMyTrip to build own models.

Chapters

  1. 0:00AI's impact on travel
  2. 0:40Data protection strategy
  3. 2:01Using LLMs and grounding
  4. 3:10Token cost pressure
  5. 3:29AI replication challenges
  6. 4:56Flipkart and Cleartrip
  7. 5:15AI as enabler and AI-proof claim
  8. 6:01AI's potential and uncertainty

Questions asked in this interview

7
  1. 0:03How big is this AI thing?
  2. 0:08Are you seeing it change already?
  3. 0:35Why would you allow ChatGPT to scrape your data? Can't you block that?
  4. 2:50When you say grounding, what do you mean?
  5. 3:07Why don't you run your own model on your own data?
  6. 4:56They bought this company in travel, right?
  7. 6:33It doesn't matter to you, no?
Rajesh Magow 0:00 ↗
And better market cap.
Interviewer 0:03 ↗
How big is this AI thing?
Rajesh Magow 0:06 ↗
I think in our world it's huge.
Interviewer 0:07 ↗
Yeah?
Rajesh Magow 0:08 ↗
Yeah.
Interviewer 0:08 ↗
Are you seeing it change already?
Rajesh Magow 0:10 ↗
Change in terms of what?
Interviewer 0:11 ↗
Like if I am going to MMT's website and looking for a hotel today, are people already going to ChatGPT and saying, "Find me the cheapest price on next hotel?"
Rajesh Magow 0:21 ↗
Not as much. We are also on ChatGPT. They have a feature called "@MakeMyTrip" where you can bring in your brand and you can have your own native experience. We are not getting too much traffic coming in from there. It's just taking time.
Interviewer 0:35 ↗
Why would you allow ChatGPT to scrape your data? Can't you block that?
Rajesh Magow 0:40 ↗
Everybody blocks it.
Interviewer 0:42 ↗
Mhm.
Rajesh Magow 0:42 ↗
They end up still doing it. You will obviously do whatever you can do to block it, but what you do end up protecting is deep data, deep in the funnel. There can be multi-layer security that you can build that can do it. Superficial data is public in any case.
Interviewer 1:06 ↗
Mhm.
Rajesh Magow 1:07 ↗
So what data matters, Nikhil, in this business is not necessarily that super top-layer data.
Interviewer 1:12 ↗
Top layer is price, availability.
Rajesh Magow 1:15 ↗
Price, just the description of the product.
Interviewer 1:18 ↗
Yeah?
Rajesh Magow 1:19 ↗
Under it is user-related data, right? The end traffic, like your customer data, the preferences of consumers.
Interviewer 1:28 ↗
Right.
Rajesh Magow 1:29 ↗
Right? Because what let's say Nikhil has been coming to our platform and what he has been buying over a period of time, different products. Let's say if you were buying a product, you were going with a group, you were buying a homestay or you were buying a hotel, which kind of hotel were you buying, and even the funnel drop-off data, like how much time are you spending? Are you a looker or are you a booker? Are you a serious traveler or high-intent traveler, or a low-intent traveler?
Interviewer 1:59 ↗
When you incorporate AI in your company—
Rajesh Magow 2:01 ↗
That is correct.
Interviewer 2:02 ↗
—you will not use these LLMs from the—
Rajesh Magow 2:04 ↗
No, we will. We will use LLMs.
Interviewer 2:06 ↗
—get that data, no?
Rajesh Magow 2:07 ↗
No, not necessarily. Not all of it.
Interviewer 2:10 ↗
I'm asking, you won't run an open-source model of your own and try to put your data in that so it's—
Rajesh Magow 2:14 ↗
So, I'll tell you, right now we're doing both, and it's an evolution which I believe will happen. In the early days, because these frontier models were very strong, you end up using multiple frontier models, right?
Interviewer 2:29 ↗
Like Claude and OpenAI.
Rajesh Magow 2:30 ↗
All of them.
Interviewer 2:31 ↗
Gemini.
Rajesh Magow 2:31 ↗
Gemini, Claude, Cursor, GPT. But the grounding of that will happen with our proprietary data. That data doesn't go anywhere, like user reviews, for example. That only belongs to us.
Interviewer 2:50 ↗
When you say grounding, what do you mean?
Rajesh Magow 2:52 ↗
Just to make sure that the accuracy and the relevance of the responses to the customer is far more accurate than generic.

35 more exchanges in this transcript

Sign in free to read the rest of this interview. No card required.

Sign in to read the full transcript

Cite this transcript

APA, MLA, BibTeX
APA

Magow, R. (2026, September 9). Will AI Replace MakeMyTrip? The Founders Respond 🤖 #nikhilkamath #makemytrip #tajhotels #travel [Interview transcript]. Bharat Breakout. CEOInterviews.AI. https://ceointerviews.ai/interview/1335069/

MLA

Rajesh Magow. "Will AI Replace MakeMyTrip? The Founders Respond 🤖 #nikhilkamath #makemytrip #tajhotels #travel." Bharat Breakout, 9 Sep. 2026. Transcript, CEOInterviews.AI, https://ceointerviews.ai/interview/1335069/.

BibTeX
@misc{magow2026_1335069,
  author       = {Rajesh Magow},
  title        = {Will AI Replace MakeMyTrip? The Founders Respond 🤖 #nikhilkamath #makemytrip #tajhotels #travel},
  howpublished = {Interview transcript, Bharat Breakout. CEOInterviews.AI},
  year         = {2026},
  month        = {sep},
  url          = {https://ceointerviews.ai/interview/1335069/},
  note         = {Speaker-attributed transcript with timestamps}
}