Connected mobility is evolving fast \u2014 but the real shift isn’t in the vehicles or infrastructure themselves. It’s in how data, connectivity and compute interact in real time, enabling smarter decisions at scale. This interplay, between network APIs, edge AI, and regulated connectivity, was at the heart of discussions at the Smart Mobility Summit held at MWC25.
Across policing, emergency response, logistics, automotive and port operations, a common thread emerged: real-time situational awareness, resilience, and scalability depend on exposing network capabilities in a developer-friendly way \u2014 and aligning them to actual operational constraints on the ground.<\/p>\n\n\n\n
For years, the industry recognised the technical potential of telco APIs but questioned their commercial readiness. That\u2019s now changing. From remote drone fleets to port operations and fleet routing, practical deployments are starting to emerge. These are not just prototypes, but functioning services. Nokia, for example, showcased its Network as Code platform, which aggregates network APIs from multiple operators and simplifies access for developers and enterprises. The value lies not just in exposing location or identity data, but in controlling quality of service and latency \u2014 especially for safety-critical tasks.
\u201cBeing able to guarantee the bandwidth allocation, ultra-low latency, that\u2019s where the real differentiation is,\u201d said Alex Walling, Head of Business Development, Network as Code at Nokia. He pointed to use cases requiring deterministic performance. For example, controlling a ship remotely or steering a car over the public mobile network. \u201cIn the US, to operate a drone without line of sight, you have to guarantee 50 milliseconds of latency. That\u2019s only achievable if you can allocate the network itself.\u201d Alexandra Reichl, Liberty Global\u2019s VP for Mobile & API Services underscored the need for standardised APIs for mobility. \u201cThe standards are important here, and I think this is particularly the key with mobility. from our perspective as the network provider, this is a fundamental shift. We are putting network exposure layers into our network so that those functionalities, those APIs can be exposed and be made available into the solutions that are out there\u201d.<\/p>\n\n\n\n
Perhaps the most mature example of connected mobility came from Swisscom. Philipp Eder, Head of Drones & Robots at the operator, outlined the company\u2019s move beyond individual drone deployments to an integrated national service. \u201cIf you had been here ten years ago, everyone spoke about drones \u2014 but nobody was really able to scale it,\u201d Eder said. Swisscom\u2019s solution supports a distributed network of 300 drones across Switzerland. Each is connected via 4G or 5G and operated through a service model akin to ride-hailing. The aim is rapid response within five minutes to any location, whether for fire surveillance, security patrol, or infrastructure inspection.
But connectivity alone isn\u2019t enough. Eder emphasised three layers of support that operators must provide: stable aerial connectivity, real-time crowd density data (to ensure safe flight paths), and access to broader airspace awareness tools. \u201cIt\u2019s not just providing a SIM card. It\u2019s making sure you have a stable connectivity in the air, which is quite challenging. We can determine how many people are in a certain location. This is a key factor, which is needed for the industry to grow.\u201d<\/p>\n\n\n\n
If the Swiss model represents scalability, the Madrid police department brings the operational constraints into sharp focus. David Hermoso, Inspector for Policia Municipal de Madrid, explained how drones are already in active use. The city used them for event monitoring, traffic oversight, and emergency response \u2014 but limitations remain.
\u201cOur biggest challenge is transmitting the video feed in real time to the decision-makers. In crowded areas, networks become saturated,\u201d Hermoso said. The team experimented with Starlink, drone-in-a-box solutions, and hard-wired rooftop uplinks, but none offer the portability and dynamic response needed for urban policing. Hermoso emphasised that the team needs prioritised mobile connectivity that performs reliably during city congestion and high-stakes situations. \u201cWe\u2019re counting on 5G to provide the quality of service we need to scale \u2014 without taking unnecessary risks.\u201d<\/p>\n\n\n\n
Public safety and defence sectors only find new capabilities valuable when they can trust them \u2014 especially under pressure. As Magnus Packendorff, Head of Mission Critical Networks for Ericsson noted, \u201cmission-critical services are rarely used on sunny days. They’re used in crisis. It’s crucial these networks are trusted.\u201d
The key, he argued, is making the network programmable in ways that map to real-world operational control. For example, being able to dynamically assign high-priority network slices based on a command centre\u2019s view of an evolving situation. \u201cIf there\u2019s a fire and they don\u2019t have water, they need to call a plumber \u2014 and that plumber should have the highest priority in the network. But in reality, it didn\u2019t work. Now, with 5G and dynamic APIs, we can actually achieve that.\u201d This transition \u2014 from theoretical capability to usable tooling \u2014 is what will define the next phase of deployment.
Security must move at “machine speed”
Exposing the network creates new attack surfaces, particularly as services span public and private domains. Mitchell Rappard, Director – 5G Technical Solutions of Palo Alto Networks highlighted the importance of visibility and automation. \u201cWe have a saying in the security world: you can\u2019t secure what you can\u2019t see. Implicit trust is the enemy of zero trust.\u201d
He noted that mobile networks introduce protocol-level challenges unfamiliar to traditional enterprise IT teams. Security teams must apply zero trust models in ways that account for GTP-U, SCTP, and other telco-specific behaviours. More critically, detection and mitigation need to be autonomous. \u201cWhen something does go wrong \u2014 and it will \u2014 the system needs to respond at machine speed.\u201d Rappard also called for closer integration between network intelligence and security systems, so network APIs could trigger automated actions. For example, to isolate compromised devices or adjusting traffic handling in real time.<\/p>\n\n\n\n
Mobility is inherently cross-border, and so are the networks and vehicles supporting it. The industry is betting on initiatives like GSMA\u2019s Open Gateway to bring coherence across operators, regions and industries. As Alexandra Reichl pointed out, \u201cThese are vehicles. These are devices that will cross borders, that will go in and out of private network fields, and it needs to be straightforward. That functionality, that provision needs to work regardless of where you’re sitting, how you’re behaving, and, you know, that standard functionality will be key here.\u201d
KT echoed that sentiment by presenting use cases for automotive AI services and traffic signal optimisation in Korea. For these to scale beyond local deployments, APIs must be portable, discoverable and supported across operators. The focus, several speakers argued, should now shift from technology readiness to ecosystem readiness.<\/p>\n\n\n\n
Operators are exposing their network APIs. Developers are building. Industries are engaging. The technical foundation is no longer the bottleneck. But scale still depends on three things: shared standards, viable business models, and proof of real-world impact. As Philipp Eder put it, \u201cthe technology is here.\u201d The task now is to make it usable by delivering services that enable customers to fly, monitor, protect and respond at scale.<\/p>\n\n\n\n