Exploring govt partnerships in generative AI space: Google Cloud India MD

We are working with one of the state governments, to help them with population immunization and vaccination programmes based on the data available

Google Cloud is exploring the possibility of working with the Indian government to help it in its vision of 'Make AI in India' and ‘Make AI work for India', said Bikram Singh Bedi, managing director at Google Cloud India. In an interview with Peerzada Abrar in Bengaluru, Bedi said the company is also exploring partnerships with the government in areas such as healthcare, e-commerce, and cybersecurity, where generative artificial intelligence (GenAI) will play a key role. Edited excerpts:

How is GenAI impacting your business?

For the first time, we are seeing that the ability to apply it (GenAI) from a commercial cost perspective is now starting to make sense. As we work with our customers, partners across SMBs (small and medium business) enterprises, the public sector, and even consumers, what we're starting to see is that we are making a significant and real-world impact. No longer are we at the stage where it used to be about let's do a small trial or proof of concept. For example, we are working with HDFC. We have set up a centre of excellence for them around GenAI for insurance. The impact is beginning to materialise. We are also working with Apollo Hospitals, helping doctors decide the next best action after seeing the patients.

What is Google Cloud doing to help the Indian government translate its vision of ‘Make AI in India’ and ‘Make AI work for India’?

We are exploring partnerships with the government to see how we can participate and help them deliver on this vision. Such (partnerships) include a model where you have to train it with data and use it in various business applications. We are in the midst of exploring this with the government. It is about providing secure data architectures. We use our Vertex AI product where we will be exposing not just Google models, but third-party and open-source models, coupled with enterprise search, as well as conversational capabilities. The objective is to take data, which is there in India and train these 100-odd models. Then we start training the model in the local languages. For this, we are exploring a partnership with Bhashini (the native AI-based language platform). We've also announced the training of 1 million (professionals and students) to learn GenAI and related machine-learning concepts. We have also announced a chair with the IIT Mumbai, where we are going to be working with students and developers to help them innovate on Google Cloud and Google AI technologies. We are also exploring a partnership with the government to help India improve its cyber defence posture. We're training 1,000 government employees in the cybersecurity space. AI is now in all our products, including security. Consumer data intelligence company, Axis My India, is building a one-of-a-kind platform leveraging Google technology that aims to bring about a change in people's awareness, accessibility, and utilisation of a slew of services.

What are the innovations in the GenAI space for India?

India is an important market for us. It is one of the key markets outside the US where we have all functions deployed, like go-to-market, professional services, engineering and customer support. We're trying to bring them together to make sure that we are able to drive these innovations (in these areas) for our customers. For example, we are working with the Open Network for Digital Commerce (ONDC). We have built a piece of code that we have now open-sourced and a large part of the technology service providers (TSPs) for ONDC have picked it up, who have started to build on top of it. It is going to help onboard the buyers and sellers quickly. This leads to a better consumer experience. We are now working with these TSPs if we can build (features) such as recommendations and search. A lot of these are AI-driven. Building on ONDC is also driving the world's first initiatives and innovations.

What are some of the big opportunities, including large government projects that Google Cloud is tapping in India?

We are working with one of the state governments to help them with population immunisation and vaccination programmes based on the available data. We are working with the transportation department of one of the large cities, where we're helping them with traffic management. There are projects where we are using technology to help provide citizen services. One of our partners is CoRover, (a human-centric conversational and generative AI platform), which is building BharatGPT. That partnership is being built on top of Google's Vertex AI. There is also a collaboration with ONDC. We believe commerce in India is a large opportunity to impact individuals at a social level. India has already shown this through the Unified Payments Interface (UPI) and CoWIN. Now, India has established that it has the ability to scale up digital assets really well. Now programmes like Gati Shakti (national master plan for multi-modal connectivity) and ONDC are the pipeline examples of what is going to be scaled up. ONDC and Google Cloud will launch an India-wide hackathon aimed at catalysing innovation and addressing critical challenges for the (next billion digital users) in the country.

Experts say that AI systems with human-competitive intelligence can pose profound risks to society and humanity. How is Google addressing these concerns?

As Google and Google Cloud, we have taken a socially responsible approach to address these concerns. When you train the models, they have to be trained within your secure network with your data. We recently announced an enterprise grounding service that works across (Vertex AI foundation models, search and conversation) that gives customers the ability to ground responses in their own enterprise data to deliver more accurate responses. The second part of it is security, which is why it's necessary that it has to be trained in your environment with your data. We do penetration testing and put rigorous checks on applications. Our view is to adopt GenAI boldly, but in a socially responsible fashion.