Qualcomm CIO Atilla Tinic is rethinking the IT playbook for the AI era

With over 30 years of IT experience, Qualcomm CIO Atilla Tinic has seen a few shifts in his time. But like many executives and analysts I’ve spoken with recently, he says AI is changing how he manages IT more significantly than anything that came before it.

By now, we know all too well that a model is only as good as the data it was trained on. At Dreamforce last month, Salesforce executives went so far as to say that the company's data is much more important than the interface, and that perhaps the interface could give way to an AI front end like ChatGPT or Claude.

Much like Salesforce, Tinic says that he is willing to reconsider some of his long-held assumptions in the face of AI. At the same time, the company is also shifting and changing as it expands beyond handsets into other areas like automotive, IoT and data centers. The company is projecting $40 billion in non-handset revenue by 2029, up from about $11 billion in FY 2025, on top of the handset business that still anchors the company today.

I spoke to Tinic about how these changes are altering his thinking, as he begins to see how AI and the company's strategic push into other areas are closely intertwined.

The changing use of data

One of those fundamental assumptions that has changed is how a business uses the data it stores in a data warehouse or data lake.

He said that if you asked him 10 years ago about how you use that data, he would have a completely different answer, and that's due to how AI uses the data inside a company. "If you'd asked me about your data warehouse at that time, or data lake, I would have said never include transactional processes in your data warehouse or lake. It's there for BI," Tinic said. In those days, systems of record like Salesforce handled the transactions.

But when it comes to AI and how work gets done these days, that line between analytics and operational systems is blurring. Agents can now access data from across the business, which all comes together in the data lakehouse, to carry out tasks that previously depended on individual transactional systems.

"I've done a 180 now, and I really believe your data lake or lakehouse is really your source of record for unified data going forward," he said. "And when you start to incorporate AI agents on top, they are more powerful in their capabilities when they have unified data that's trusted and curated that they can leverage."

Agents require rethinking workflows

In many ways, the coming age of AI feels like what companies went through with digital transformation a decade ago, but Tinic says there is a fundamental difference. "With digital transformation, we spent a lot of time automating the existing workflows that, to be frank, have been there for a long period of time," he said. "We didn't really change the workflows, we just automated the tasks within them."

What agentic AI is doing is giving companies like Qualcomm the opportunity to embed AI into the way they redesign systems and workflows, rather than simply slapping it on top of what already exists. “With AI, we now have to think about a digital worker who's going to perform tasks differently in combination with your internal workforce than maybe a human does,” he said.

Governance is another area where AI has challenged Tinic's traditional approach to IT. "Governance in my vernacular and historically, it was sometimes a dirty word because it implied that you might be moving too slowly," he said.

But he says Qualcomm has developed a centralized approach to AI governance, which actually helps the company move faster. Instead of allowing AI tools to proliferate across the organization and trying to bring them under control later, as often happened with cloud and SaaS, the company has established oversight from the get-go, allowing it to adopt new technologies quickly while better managing the associated risks.

All of these technology changes are tied inextricably to the strategy shift the company is going through. "I think I'm lucky in the sense that the two go hand in hand because as you're going through a transformation that's changing some of the IT ecosystem for our diversification strategy, we're now able to embed our AI strategy into that as we go forward as well," he said.