 At IBM Research we're doing work across all aspects of IoT from sensors, how they're connected to the platform, building the platform, and then turning that sensor data into solutions for industries. In order to do this, we have to be able to experiment and understand how these systems are going to work and how they're going to interact with people without actually building out the full solution. We are building what we call a virtual testing platform. At its core, what this means is that we can take some sensors and some people and integrate them into a much larger virtual platform where we're simulating the behavior of the hundreds of thousands to millions of sensors that are in the system. So in collaboration with University College Dublin, our lab has created a platform where we can test out and understand how large-scale sensor systems are going to react in the real world as they interact with humans. There are many possible applications for testing services on our virtual testing platform. Take as an example a car sharing service. A group of friends are going to a concert and want to meet and share a car to get to the event. The car sharing service adapts the meeting place dynamically based on current traffic to ensure that people will not wait for each other too long at the meeting place. So this combination of simulated and real-world feedback can help us understand if the optimization achieved in this service is perceived as reasonable by the users. Until now, cars have been designed solely for the benefit of the owner. In the future, they will be designed to minimize pollution in densely populated areas like city centers or automatically switching between combustion and electric mode. The virtual testing platform allows us to design these protocols and measure driver response and acceptance. Imagine your electric bike being able to detect polluted areas in advance and to help you by increasing the level of assistance to reduce your effort. As a result, your breathing rate will lower and this would mitigate the harmful impact of pollution. The virtual testing platform can be used to connect to the bike and to test how the cyclist would actually react to this new service, investigating the interactions between the cyclists and the bike. Imagine a city where accidents don't cause traffic jams, where each car is being rerouted around obstacles while minimizing the impact for each driver. In some cases where congestion is building, individual cars can be directed through or around an area of limited access and the flow can be maintained without creating knock-on effects downstream. The virtual testing platform can simulate these large-scale environments and networks while providing a way to perform controlled user acceptance tests. This combination of simulated and real-world data generate valuable insights that are critical to system development. These applications of the virtual testing platform are just the tip of the iceberg. By developing the ability to combine real-world experience with virtual testing and simulation, we will be able to bring new IT services to market faster than ever before.