AI integration will enhance efficiency, magnify abilities, concur experts

AI integration has penetrated into nearly all sectors, concurred AI tech evangelist Anil Phulpagar, and Yogesh Agarwal, Founder and CEO, Unthinkable Solutions during a panel discussion at DevSparks 2025 held in Pune on 1 March.

Discussing the AI integrations across sectors, Phulpagar said that finance, legal domain and retail have seen maximum adoption of AI with the key objective being improving efficiency. 

Agarwal added that every industry has some use case of AI integration. Taking the example of healthcare, he explained, “When a doctor is trying to prescribe medicine or doing a diagnosis, AI can help them. If somebody has a cold, the complementing symptoms could be a cough or a fever. AI systems can keep suggesting (follow up questions). Based on that, it can also recommend medicines.”

“These suggestions however are relevant to a doctor…but they should not be used for directly prescribing medicines,” he said.

Why do companies integrate AI?

According to Phulpagar, a blend of two factors is driving companies towards integrating AI—adopting because everyone else is doing so and adopting AI to solve for a specific goal.

“Everybody wants to achieve better performance, better optimisation and better results. So when we are driving a business at a senior leadership level, the strategy that you're forming will be specific to your business. Where exactly you can leverage AI to achieve your goals will be different, and that depends on the business use case,” He said.

When it comes to customer relationship management (CRM), Agarwal explains that with drip campaigns, as opposed to the similar messages sent to customers, AI helps make the mails more personalised. This helps improve conversion rates, efficiency and better results.

Also Read

AI’s role in medical breakthroughs: How GPUs are transforming genomics and drug discovery

Privacy, security and safety

Maintaining security measures will become more challenging with the implementation of GenAI-based solutions, according to Phulpagar.

“Today, if I'm using a personal application where I'm going to a banking website, and I develop an agent application, a system, which will do my regular transactions with the bank. In case I develop a small bot, because we've got so many open-source LLMs as well, I have to host that LLM model myself. Only then will my data remain secure, because the moment I'm using something which is there on the internet, I'm passing all my data to the third party, or the provider of that LLM,” he added.

The challenges are domain specific, reiterated Agarwal. Elucidating on the healthcare example, he said, “If a patient starts to use it (AI) to self-diagnose and find medicines, it is not the right thing. But if a doctor is being suggested using that, it becomes right.”

“It's not so much about the industry or sensitivity, it's more about how it's being used,” said Agarwal.

What the future entails

Discussing the possibility of developers launching and running their own products, Phulpagar said, “If I'm an engineer, I don't have any finance knowledge or I don't have any administrative skills. All the answers can be achieved from these LLMs. Knowledge is abundant today, the opportunities are abundant, and it is across industry. So for developers, I think it's a good time. It is just that you need to start off, you need to have that idea, and just go forward.”

AI helps magnify one’s creative and other abilities, believes Agarwal. “Right now, in engineering, value is counted mostly based on their (people's) experience, and then a little based on what they actually do. But it has to change, and it would magnify, because in AI, it would become so much more clear that people who are creative can do more,” he concluded.