The CEO of Qualcomm (QCOM.US), the world's largest smartphone chip company, stated this week at the Mobile World Congress (MWC) in Barcelona that the upcoming massive wave of "AI Agents" will transform the broader digital ecosystem. A shift is underway from the question-and-answer format of chatbots like ChatGPT towards AI Agents capable of "autonomously executing various tedious and complex tasks." This emergence signifies that artificial intelligence is evolving from an informational aid into a highly intelligent tool for transformative productivity in human society.
In a keynote speech at the mobile industry's largest annual gathering, the CEO projected that 2026 will be the "Year of the AI Agent." These services, which can handle complex, multi-step AI workloads or proactively complete vast repetitive tasks for users, require massive amounts of data and real-time context. This will elevate the importance of various intelligent mobile devices or physical AI devices as data sources, he explained in an interview. "We are moving from a mobile-centric, app-centric digital ecosystem to an agent-centric ecosystem for the new era," he said. "AI Agents will become the center. They don't just respond to you. They observe, interpret, and act."
Unlike the current model where users initiate all actions through specific applications, the future trajectory of AI technology will focus more on empowering agents and multi-step tasks. This could mean that smartwatches, smart glasses, or other wearable micro-intelligent devices will become as crucial as smartphones in providing continuous awareness of the user's environment.
Ahead of MWC, the San Diego-based company introduced its new Snapdragon Wear Elite chipset, designed to power smaller, wearable smart devices with on-device AI processing capabilities. The company envisions its customers creating a range of on-device AI products, from pendants and brooches to smart headphones and watches. These devices will integrate efficient large AI models and sensors, providing vital technical support for personalized agents. Samsung Electronics, Google under Alphabet Inc., and Motorola under Lenovo Group have all committed to developing new consumer electronics products based on the Wear Elite platform. The first significant products are expected to launch sometime this summer.
Lenovo, exhibiting under the banner "AI for All, Devices for You," also signaled that its vision for the future of consumer technology involves a decentralization of the smartphone. "We aim to be the orchestrator of all this incoming data, coordinating and enabling this through our AI hyper-agent, Qira," a Lenovo executive vice president stated in an interview. "AI, and natural language interaction between humans and machines, will change the form and size of devices. We are moving from applications to intent."
From the current technological trajectory, the development of AI software is concentrating on two areas: "generative AI applications" and, building upon generative AI, a shift from chatbots to "AI Agents that autonomously perform complex tasks." The urgent need for businesses to enhance efficiency and reduce operational costs is currently driving the widespread adoption of these two core categories of AI software. Among them, AI Agents are highly likely to represent the ultimate trend in AI applications for the next decade. Their emergence marks the point where AI begins its transformation into a highly intelligent productivity tool.
Recent research indicates that the market size for AI Agents could reach $53 billion by 2030, implying a compound annual growth rate of 46% starting from 2025. The anticipated concentrated emergence of autonomous AI Agents like Anthropic's Claude Cowork and OpenClaw in 2026 is not coincidental. It fundamentally results from the first simultaneous convergence of five key curves: model capabilities, tool protocols, development frameworks, reasoning costs, and on-device contextual capabilities.
At the AI application layer, AI Agents are poised to become the dominant commercial interface because they convert "intelligence" directly into "action." This signifies AI's progression from "knowing how to answer" to "knowing how to execute, collaborate, and complete highly complex multi-step tasks." A recent trend report on intelligent agents projected that by 2026, agentic workflows would spread from engineering teams to non-technical departments such as sales, legal, operations, and marketing. This view aligns with the CEO's perspective—it's not that "Agent will be a specific feature," but that agents are becoming the default interaction paradigm for next-generation software. Agents will be more than just models; they will be "context aggregators + action orchestrators," moving from Apps to Intent, from tools to execution, and from UI to delegation.