Something felt different at MWC 2026. Mobile World Congress has always been the place to find the next big thing in connectivity, devices, and digital infrastructure. However, this year, the atmosphere has completely changed. Walking through the halls of Fira Gran Via, it became clear that AI is no longer something that can only be accessed through apps and chatbots. It’s starting to reach us through our phones, smart glasses, and the networks themselves.
I spent three days on the show floor trying out products, listening to keynotes, and talking to executives. What’s striking to me is how quickly AI is moving from a tool we use to something like a work partner. We are starting to see, hear, interpret, and respond in real time, and the changes will have a huge impact on our business.
The following devices monitor, listen and respond
The most impressive demo I saw was Honor’s Robot Phone. It’s equipped with a motorized camera that tracks your movements, senses what’s going on around you, and reacts in an incredibly lifelike way. When the camera turns on, it feels less like turning on a cell phone and more like turning on a little robot ready to interact with the world.
The broader implication here is that devices are becoming much more context-aware. They are beginning to understand what is going on around them and are beginning to respond without the need for constant direction. For consumers, that might mean better videos and more informative interactions. For businesses, the impact could be even more significant.
Consider jobs that require people to move around, use both hands, and work in rapidly changing environments. Field engineers don’t want to stop and look at manuals. Nurses in busy wards cannot keep walking away to look up something on a screen. In such an environment, a device that understands context and provides help in the moment becomes a completely different kind of productivity tool.
A key question for leaders is which roles benefit most when devices can understand their surroundings, and what rules need to be in place when devices can see and hear more of the workplace.
Smart glasses are starting to become important
The most interesting context-aware interface trend I saw at MWC 2026 was smart glasses, especially display glasses from companies like Meta and Alibaba.
One of my most memorable experiences at the show was trying out the Qwen AI glasses during a live conversation with someone who spoke Chinese. As they spoke, I heard their words translated into English, and the English captions appeared in front of me in real time. It was one of those moments where you suddenly see the practical value of technology.
This is where the opportunities for companies become very real. Work is more continuous when information is displayed directly in your line of sight. You don’t have to stop, pull out your device, unlock the screen and look for what you need. Information comes to you while you are focused on the task at hand.
The use cases are easy to understand. Real-time translation can remove friction in global collaboration. Hands-free navigation helps staff find the right location or asset on a large property. Contextual Q&A means technicians can look at equipment and ask what to do next, rather than having to stop work to find support. Each of these are pain points that organizations are already aware of, but what’s new is how technology can solve them.
Even more important than context-aware phones, smart glasses move AI from the screen to the environment around us. AI will become something that people will work with all day long. There are also some important questions that arise regarding glasses. What can be recorded? Where is the data stored? What does consent actually look like? How do these tools work with existing systems, knowledge bases, and workflows?
Physical AI shows where the cloud falls short
Robotics and autonomous systems have been part of MWC discussions for many years. This year, I felt we were much closer to actual commercial use. This also brought to light a major problem. Cloud-only AI is often too slow or too limited for physical environments.
When decisions need to be made in milliseconds, or when sensitive operational data cannot be exfiltrated from the site, it is not always practical to send everything back to a distant cloud platform. Real-world examples include robots in warehouses and production sites. If you have to wait for the cloud to respond before reacting to an obstacle, that delay can create risk.
Edge computing is not new, but MWC 2026 showed that the discussion is moving forward. Today, we’re seeing networks and infrastructure designed to strengthen the collaboration between connectivity and compute and run AI workloads closer to where the data is created.
Satellites are more important than ever
Another area of focus was satellite connectivity. Ground-based networks will continue to do most of the heavy lifting in locations where coverage is strong. Satellites become important, especially in remote areas where reach and resiliency become more important.
As AI-powered work moves into the physical world, network failures become a bigger problem. If your operations rely on real-time data or autonomous systems, loss of connectivity can quickly become a business risk. Satellite connectivity is starting to serve as part of the resiliency layer that makes edge AI more reliable in real-world situations.
Agentic AI is becoming the operating model
If there was one theme that came up again and again in my conversations at MWC 2026, it was agent AI. These are systems that can plan, decide, and act with some degree of autonomy. They are moving from experimental projects to actual use. Networks are a natural place to start because they are complex and constantly changing. AI agents can monitor conditions, identify potential failures, and initiate or adjust maintenance with far less human effort.
The same pattern is emerging in supply chain and cybersecurity, where work often relies on rapid decision-making across multiple connected systems.
Agenttic AI changes the role of AI in business. Traditional AI often helps people analyze information and make recommendations. Agent AI is further evolved to perform actions. This also means that governance challenges will become even more acute. Businesses need an audit trail and accountability for the operation of their systems. You also need to define where human oversight will occur and which decisions can be handled independently.
The organizations that get the most value from agent AI will be those that view agent AI as an operating model, not just a feature. This means redesigning workflows, connecting systems properly, setting clear rules, and measuring results in meaningful ways.
What leaders should do now
MWC 2026 sent a very clear message. Enterprise AI is on the verge of becoming a reality. We are moving towards devices that understand their surroundings. Glasses are evolving into glasses that support working people. We are moving to networks designed to handle AI workloads as a core requirement. Agentic AI is becoming the coordinating layer that connects all of this and turns capabilities into outcomes.
For leaders, the starting point must be practical. We’ll focus on a few workflows where contextual AI can create clear value. Build the infrastructure needed to make these workflows reliable. Put governance in place before autonomous systems start operating at scale. Set up policies for new interfaces, such as smart glasses, early so your organization is ready when the pilot project begins. What MWC 2026 made clear is that AI is moving from interfaces to the real world of work, and that change is only just beginning.
