Ambient Intelligence (AmI) lies at the intersection of different technologies, including Artificial Intelligence (AI), Internet of Things (IoT), Big Data, Pervasive-Ubiquitous Computing, and Networks and Human Computer Interaction (HCI).
The technology represents the perfect convergence of the IoT and AI. IoT sensors and devices embedded into the surrounding environment (such as the user’s home and workplace) will gather contextual information and use AI to predict the user’s needs.
For most of the last decade technology has been providing us with reactive assistance. But AmI services will be able to help you both reactively (through user-initiated requests) and proactively (with AI-initiated help). In other words, AmI doesn’t ask the user what they need, but has the ability to make decisions and take action on the basis of user context.
The future of ambient intelligence
Early-stage examples of AmI include Alexas, health monitoring apps etc. However, as ambient intelligence improves, the further integration of AI into the IoT will provide the foundation for what is now being referred to as the Artificial Intelligence of things (AIoT). The AIoT devices will be able to analyze collected user data and make and act on judgements without the involvement of human beings, resulting in greater efficiency and effectiveness.
Many such devices will be supported by robust edge-cloud infrastructures, which will allow them to execute localized processing and reduce communication latency from centralized data centers—thereby further reducing their response time.
To create more advanced ambiently intelligent systems, two things will be critical: (1) multimodal and multisensory intelligence and (2) interaction skills. The first refers to the ability of AI to process multimodal signals (sensory, linguistic, and visual modes of intelligence) simultaneously, while the latter refers to the ability of AI to master natural language and communicate as effectively as human beings.
The potential of Ambient Intelligence
Ambient Intelligence presents endless possibilities towards making our lives easier and better: it has the potential to help us in settings from our living room to the hospital.
Here are just a few examples of exciting things ambient intelligence could make possible:
- AmI could change the way we travel. For instance, once you disembark the airplane, you may receive an alert from one of the devices you carry (it could be a smartphone, a smartwatch, or your AR glasses) informing you which carousel you will be able to find your luggage in. Then, the device could proceed to give you directions to the carousel—and once you’ve picked up your luggage, another message could tell you when and where to find the nearest taxi. On the way to the hotel, you could receive yet another alert letting you know that you’ve already been checked in, and provide you with your room number and a digital passcode.
- AmI could also improve the speed and quality of medical care. For example, if a user with a heart condition is wearing an Apple watch or a device that monitors their biomarkers, it could directly communicate with their cardiologist as soon as it senses arrhythmia or some other abnormality. Then, the doctor will be able to take immediate action, whether that involves calling the patient, scheduling an appointment, or having their prescription delivered to the pharmacy. In hospitals, ambiently intelligent sensors could have multiple uses. For instance, operating room sensors could maintain surgical count to ensure that no surgical objects are mistakenly retained inside the patient.
- AmI could also be used to provide assisted living centers to proactively anticipate the needs of elderly people. An ambiently intelligent house be an alternative to nursing homes for geriatric patients— providing them an environment in which they can live safely and independently.
Privacy concerns
AmI systems, as exciting as they are, have also sparked concerns regarding privacy. Naturally, many people won’t be comfortable with intelligent spaces collecting data without their consent. AI still has a bias problem, and user information falling into the wrong hands could have serious repercussions. Navigating the ethical challenges of Ambient Intelligence is going to be an uphill task.
