How MMTs are using Gen AI to reduce travel anxiety

The IT teams led by Sanjay Mohan and Ankit Khanna at MakeMyTrip are coming up with creative methods to use generative AI to enhance the travel experience. For the hotel and growth & development companies, Ankit is the chief product officer and Sanjay is the chief technology officer. Additionally, they employ around 1,200 individuals in the engineering team and another 1,000 in the data science and product departments.
A significant use case for Gen AI, according to Ankit, has been hotel evaluations.

Reviews for hotels might number in the hundreds or even thousands. Ankit claims that choosing which reviews to read and which not to read is difficult for the user. To save the customer time from having to go through several evaluations, they have used Gen AI to generate summaries of hotel reviews. “Ankit, who was product-in-charge at Careem, Freecharge, and Snapdeal prior to joining MMT in 2019, says that you can get a complete understanding of why people choose a hotel and what they like and dislike about it in just one paragraph.”
It is only lately that Amazon has begun summarising product reviews on its marketplace. The essence of any review is contained in a single paragraph.
A chatbot named Myra that assists users in organising their travels was also developed by Ankit’s team using Gen AI. The chatbot speaks English, Hindi, and Hinglish. Many individuals find that typing is uncomfortable, yet speaking comes naturally to them. Additionally, there are much too many options when it comes to flights, and the hotel search is so content-driven that you have to read a little bit before making a choice, according to Ankit.
By giving more detailed suggestions depending on what the user says to her voice assistant, Myra makes all of this simpler.
According to Sanjay, this has also aided in connecting with people in tier-III and tier-IV cities who feel uncomfortable communicating or completing forms in English. “These models are now much more proficient and precise in translating languages. Thus, you may begin shopping, use voice control to arrange a flight, and input all the data,” he explains.
Setting the Scene for Travel
The ability of GenAI to extract themes based on context is also being used by MMT teams. When selecting hotels, for example, it is important to consider the user’s travel context. She may be selecting hotels for a business or family vacation, or she may be going alone or with friends. The user will only be drawn to accommodations and reviews that are relevant to that specific journey. In order to contextualise the trip, Sanjay explains, “We slice and dice user-generated content (UGC) accordingly, and Gen AI extracts relevant content from the UGC.”
By using tags to organise the material, GenAI may give thematic context, such as how the hotel performs in terms of location, facilities, and cuisine. “The summary varies based on the type of search,” adds Sanjay, who joined MMT in 2015 after positions at Infosys and Yahoo.
In addition to synthesising the text, GenAI also writes a paragraph centred on the major ideas that are important to the traveller. Additionally, this holds true for subcategories. What makes, for example, a certain hotel in a community unique? To do this, Sanjay explains, “We use data-science models, again created by Gen AI, to ask what it is about the hotel that makes it stand out among all the similar hotels in a particular neighborhood—clubbed with star rating, price point, etc.”
According to him, Gen AI will identify three features that set each hotel apart for the traveller: cuisine, wheelchair accessibility, and/or a children’s area. Making smarter judgements is aided by this.