Qualcomm released its newest flagship mobile chipset this week, the Snapdragon 8 Elite Gen 6, built to run large AI models directly on a phone rather than routing requests to the cloud — a launch timed to a smartphone market that is shrinking even as device prices climb.
The new chip, including an "Extreme" variant, is Qualcomm's first built on Taiwan Semiconductor Manufacturing Co.'s 2-nanometer process, and the Extreme version is designed to run AI models with as many as 30 billion parameters entirely on-device, without a network connection. Qualcomm said the chip carries a dedicated neural processing unit alongside its CPU and GPU cores, along with security controls meant to govern what data an AI agent running on the phone can access. Motorola, Xiaomi and ZTE are among the manufacturers expected to build premium phones around the chip.
A pricier chip for a shrinking market
The launch lands at an awkward moment for the industry. Memory chip prices have surged this year as manufacturers redirect production capacity toward the AI data-center boom, squeezing supply for phone makers and driving up the cost of finished devices. Analysts project the global smartphone market will contract by roughly 14% in unit shipments during 2026, with a further decline possible in 2027. The Snapdragon 8 Elite Gen 6 is expected to cost more than $300 per unit — the highest price ever for a mobile processor — a cost that phone makers will likely pass on to buyers.
Rather than compete on volume, Qualcomm and its phone-maker partners are leaning further into the premium tier, betting that buyers with more disposable income will pay up for devices marketed as personal AI hubs capable of running agents without relying on remote servers.
"We're going into this transition from what is a very phone-centric model to now an agentic-centric model for new experiences."
Cristiano Amon, Qualcomm CEO
That framing puts Qualcomm chips in direct competition with Apple's latest A-series silicon, which Apple has similarly marketed around on-device AI processing rather than raw benchmark speed. Both companies are racing to convince consumers that running AI locally — rather than through a subscription-based cloud assistant — is worth a premium, at a time when data privacy and network latency have become selling points in their own right.
For carriers and retailers, the shift toward costlier, AI-capable flagships arrives alongside slower replacement cycles, as many consumers hold onto existing phones longer given rising prices. Whether the Snapdragon 8 Elite Gen 6 can revive demand — or simply raises the entry price for an already-shrinking premium segment — is likely to become clearer once devices built on the chip reach store shelves in the coming months.
