Must read: Snapdragon 8 Elite Gen 6 series is here: Qualcomm’s new chips target flagship AI smartphones
Two chips, two kinds of flagship buyers
According to Sonal, the two flagship chips give buyers options to choose between performance and a full agentic AI experience. “There are two types of flagship customers,” he said. The Elite version is for users who play games and want great battery life and a good overall experience. Whereas Extreme is for users who “want all the brand new things,” such as running an agent, best cameras, and a powerful gaming experience.
Extreme can run large language models on the device itself, he said, and it matters for India for three reasons: No internet requirement, faster performance, and greater privacy.
“Because you are running the model locally, you can now have everything working out in your favour without actually having a stable internet connection,” Sonal said. “The most important thing is privacy, because devices are now running the language models on itself. Your data is not leaving.
Your data is all secure within the phone.”
He further highlighted that the “biggest change” will be AI agents running on smartphones. Sonal said, AI will take care of “all your tasks, mundane tasks, anything complicated.” He also shared an example, saying the agent can plan your itinerary, schedule, and travel without manual effort. “We are moving from the reactive era of AI into the proactive era of AI,” he said.
Must read: IMC 2026: Snapdragon 8 Elite Extreme Gen 6 to power three next-generation flagships from iQOO, Motorola, and Xiaomi
Shopping and travel as biggest use case in AI era
When asked about major use cases of agentic AI in smartphones, Sonal named three use cases likely to drive the first wave in India. The first is agentic commerce, and she also talked about Qualcomm’s partnership with Mastercard. Travel and productivity planning follow. “You can now define your entire day,” he said. “Productivity, shopping, and travel — the first three big use cases where you will see the agent shining.”
Mass-market rollout
Sonal revealed that AI agents could reach affordable devices over the next three to five years as advances in chip technology bring on-device AI capabilities to a wider consumer base. He said that the features will gradually trickle down as the technology matures and gets cheaper, first to its mid-premium tier and then to lower tiers.
“In the next two, three years, you will see maybe a 7th generation, which is lower than our flagship; you will have some feature sets coming into that,” he said. “In the next 3 to 5 years, we want to bring agentic AI into more and more phones so that the entire world can experience what the agents can do for you.”
However, Qualcomm does not want to accelerate the rollout without first understanding the practical use cases and potential security threats. “When you do or play on a flagship phone, you work on a very limited set of people. And then you take all your learnings and expand it into the lower tier,” Sonal said.
Must read: Microsoft launches new Surface Pro 12 and Laptop 13 with Snapdragon X2 Plus: Key upgrades explained
‘India is a land of scalability ': Sumit Sonal
Qualcomm is optimistic about India’s acceleration of agentic AI adoption, highlighting the country’s technology-aware consumers and the opportunity to scale new digital services across a large market.
“We are technologically inclined; the number of young kids who understand technology today will actually be the 1st base that will trust us (Qualcomm) with agent AI workflows, and then once that trust is built, you will see more and more adoption coming in.”
He added that education would play a key role in building consumer trust and encouraging adoption. Qualcomm is working with OEM partners, like Motorola, to introduce agentic AI solutions and familiarise consumers with their capabilities.
Qualcomm backs AI guidelines
When asked about the global AI slowdown debate, Sonal backed guidelines governing AI as language models and AI systems evolve rapidly. “I think it is in India's interest to put guidelines,” he said. Clear rules would also help companies like Qualcomm “make products much more trusted for the Indian ecosystem.”
“We are in the right direction. I think we are still a year or 2 years away from the maturity of that ecosystem, but I think we'll reach there soon,” he added.