AI can be defined as computer systems able to perform tasks traditionally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages. This unique technological advancement allows an easier interaction between people and the technological tools at their disposal, and represents an opportunity for development available virtually to everybody.
AI needs a lot of data to learn from to be useful. The extraction of data (from text, images, audio and other form of information) is greatly facilitated by tools that in turn use AI in order to be able to extract information more accurately. So, it is possible to reliably extract text from scans and pictures, as well as videos, or to identify in real time which objects are present on a picture or video, or again to extract the text and sentiment information from audio and video.
Examples of tools which extract information: apps that read and extract information from invoices or visit cards, software capable of recognising wine labels or plants from a picture, picture auto-tagging functionalities.
Arguably the most used AI tools are those that help users finding what they are searching. Most prominent example is Google search, which is very often able to give us the answer we were looking for within the first results, despite the unimaginable number of possible results. This is done thank to algorithms (computer software) that can interpret what the user is asking (as opposed to look simply for the words used in the search) and find the content that best relate to what is searched and, if data is available, to the person searching. This technology can be used by single websites (shopping, and content suggestion, such as streaming), but also internally by firms that have large amounts of documents to manage (contracts, knowledge, processes, etc.).
Other examples of tools which enable search: Netflix suggestion, newsfeed population from various social apps, shopping website’s search functions.
Chatbot and reply support
Chatbots became most useful with the introduction of AI. Instead of having to give the program a fixed set of questions to choose from and answers that can be provided (such as the FAQ), chatbots can understand the topic of the question asked by the user and look for an available answer. If this is not there, it can redirect the user to a support person. In a similar way, AI tools can analyse past correspondence records and suggest answers to received emails and letters. This process can even be automated if enough information is available.
Examples of chat bots and reply support: autocorrect/autocomplete functions, chatbots of different service providers, such as banks and insurances, or utility operators.
The ability of AI to analyse vast amounts of data and find meaningful correlations help users to make more reliable predictions. This is useful in a wide range of activities, such as prediction of visits to the emergency guard of an hospital, maintenance need of complex machines such as plane motors and the vastly complex folding of proteins in chemistry.
Examples of systems using AI based predictions: traffic prediction, maintenance of wind turbines, prediction of demand for ride sharing apps.
Widely available assistants (such as Siri or Cortana) support users by translating requests and commands received in digital actions such as performing searches, managing calendar, providing reminders, translating and much more. These tools function as generally voice activated entry point for other programs. More and more programs are developed with the option of being integrated with assistants.
Other examples of assistant bots: Viv, Google assistant, Bixby.
Autonomous robots gained, with the help of AI, an impressive ability to move in and act on the ever changing physical world. Autonomous transportation vehicles are already widely used (if not on the public roads, in slightly less crowded and more controlled areas such as warehouses, campuses, but even hospitals and hotels) to transport people and objects. Furthermore, they ability to interact with objects became more flexible, as they can adapt to a varying reality without the need of painstakingly exact instructions in order to perform tasks.
Other examples of autonomous robots: Tesla car, Roomba vacuum cleaner, Boston Dynamics robot (the dancing robots!).
The potential of artificial intelligence (AI) is enormous. But in order to be able to exploit it, in addition to technical understanding, it requires above all the courage to embrace change.
The development in data collection, data processing and modelling capabilities shows the increasing tension between progress and the current philosophy based on individuals’ independence and privacy. A profound societal and cultural transformation is happening. In my opinion, there is currently not a critical need for addressing the Personal Data Laundering through legislation, also because it is generally relatively easy to collect abundant personal data in a compliant manner. This topic is however an important aspect to keep in mind when developing best practices.
Thanks to available data, it is possible to get insight on hidden patterns, predict situations, create an intelligent ambient able to sense and act upon its surroundings, or support workflows by automating standard tasks. All this require some prior consideration on the data that will be used for the implementation of AI-Tools.