Qualcomm CEO Cristiano Amon Says 2026 Is the Year AI Agents Replace Apps, Betting the Chip Giant on Smart Glasses and Data Centers

Qualcomm CEO Cristiano Amon is making a definitive bet that the smartphone's reign as the center of consumer computing is ending. In a wide-ranging interview on CNBC's The Tech Download podcast, Amon said AI agents will become the new app, fundamentally changing how consumers interact with their devices. The vision is stark: instead of tapping icons on a screen, users will delegate tasks to autonomous AI systems that operate in the background, make decisions, and carry out complex work without constant human supervision.

At Computex 2026 in Taiwan, Amon declared 2026 the year of agents, marking what Qualcomm sees as the end of the passive chatbot era. According to Ynetnews, the chip giant describes agentic AI as independent systems that turn phones, PCs, cars, and wearables into endpoints rather than central hubs. The smartphone doesn't vanish, but it becomes just one of many surfaces where AI agents surface information and take action on behalf of the user.

Amon told Fortune that the smartphone's reign as the primary device is ending, and he is particularly bullish on smart glasses as the natural successor. This isn't a distant prediction. Qualcomm is already working on over 40 designs of new AI-powered devices, including jewelry, earbuds with embedded cameras, wearable pins, and watches, according to CNBC.

Qualcomm isn't just talking about a post-smartphone future; it's shipping silicon for it. The company disclosed it has a $65 billion design win pipeline in automotive cockpit and ADAS chips, according to an investor analysis from Rijnberkinvestinsights. More critically, Qualcomm has already signed custom silicon commitments with Microsoft, Meta, and two unnamed hyperscalers for data center inference, before shipping a single Dragonfly chip.

At Computex 2026, Qualcomm formally introduced its Dragonfly data center processors, positioning them as direct challengers to Nvidia and Intel in inference workloads. Ynetnews reports that the company sees data center AI as one of multiple inflection points over the next three to five years. The company's official statement confirms a comprehensive data center strategy aimed at accelerating diversification beyond mobile handsets.

Fortune notes that Amon credits Qualcomm's 40-year survival to a culture of reinvention. Under his five-year tenure, the company has shifted from smartphone dependency to a portfolio that spans automotive robotics, wearable AI jewelry, pins, pendants, and other form factors that don't even have established category names yet.

Despite the bold pivot, Amon isn't writing off smartphones entirely. Bloomberg reports that he said the smartphone market will remain suppressed, not because demand has collapsed, but because soaring memory prices and supply shortages are keeping device costs elevated. The demand is there; the economics are squeezing it.

This matters for Qualcomm's near-term revenue. The company still derives a substantial portion of its income from smartphone chips and licensing. But Amon's framing treats this as a managed transition rather than a cliff. The smartphone becomes one node in a distributed AI agent network, not the center of gravity. The company's licensing business, QTL, continues to generate revenue from the vast majority of its patent portfolio, while Qualcomm Technologies handles the engineering and R&D for the new compute categories.

Bloomberg's interview with Amon emphasized that the company expects memory price pressure to ease, but the structural shift in consumer computing means Qualcomm's growth won't depend on a smartphone rebound. The bet is that AI agents running across glasses, earbuds, cars, and data centers will create a larger total addressable market than smartphones ever did.

In a Time magazine interview, Amon drew a historical parallel to explain his conviction about AI's scale. He pointed out that in 2000, people thought the internet would be big, but 26 years later, the internet proved far larger than anyone predicted. He applies the same logic to AI: whatever the consensus expectation is, the reality will exceed it.

The gladiator metaphor Amon uses to describe Qualcomm's competitive posture reinforces this. He told Time he tells his team they are in the gladiator business. You enter the Colosseum with a 33% chance of winning, a 33% chance of losing, and a 33% chance of mutual destruction. Winning only buys you the right to fight again. This isn't swagger; it's an acknowledgment that the AI hardware market will be brutal, and survival requires constant reinvention.

Fortune reports that Amon is prepared to bet the farm to stay ahead. That bet is now visible in the simultaneous pushes into consumer wearables, automotive AI, and data center inference. The Rijnberkinvestatings analysis frames Qualcomm as a contrarian pick precisely because the market is pricing the Apple smartphone headwind while ignoring the multiple new revenue streams that are already contracted.

Qualcomm's entry into data center silicon represents the most significant diversification move in the company's history. According to its official statement, the company sees multiple inflection points over the next three to five years, with data center serving as the centerpiece of a comprehensive strategy to strengthen long-term computing leadership.

The Dragonfly chips target inference workloads, not training. This is a deliberate choice. Inference is where the volume and latency requirements align with Qualcomm's core competency in low-power, high-performance edge computing. Rijnberkinvestinsights notes that Microsoft, Meta, and two hyperscaler custom silicon commitments are already signed, which means revenue is contractually locked before the first chip ships.

Ynetnews frames this as a direct challenge to Nvidia and Intel in data centers. While Nvidia dominates training, inference at scale is a different game, and Qualcomm's power-efficiency advantage could matter in cost-sensitive deployments. The company is not trying to win the training market. It's positioning itself as the inference platform for the agentic AI era, where millions of agents will need real-time processing across distributed infrastructure.

The timeline is aggressive. Amon told CNBC that 2026 is the year agents go mainstream, and the company's 40-plus device designs suggest partners are already building products for a 2026-2027 launch window. Smart glasses are the category Amon is most bullish on, and CNBC reports he believes they could eventually become as big as the smartphone.

Fortune's reporting indicates that Qualcomm is positioning itself as the default silicon provider for this new category, much as it did for Android smartphones. The difference is that the agentic AI market isn't a single device category. It spans wearables, automotive, PCs, and cloud infrastructure. If the thesis holds, Qualcomm's revenue mix will look dramatically different by 2028.

The company's own statement emphasizes that the diversification is accelerating now, not in some distant future. The combination of a suppressed smartphone market, signed data center contracts, and a pipeline of AI-native consumer devices creates a narrative where Qualcomm's next growth wave is already underway, even if the market hasn't fully priced it in.