Get the latest best-practice stories, news and white papers straight to your mailbox
The selection process for the market's third operator was a spectacle to behold
Artificial Intelligence, AI, had a lot of its hype in 1990s. Then it became quite an unpopular term, especially among serious researchers and data scientists. Now it has come back and maybe this time it is something real. Deep learning in particular has become an important area for research and development. Applications are from personal assistants to self-driving cars and digital finance services.
Machine learning, data science, big data are all linked to AI, or a part of it, depending on definitions. AI was originally defined to mean "the science and engineering of making intelligent machines.” In practice nowadays it means a machine needs a lot of data from its environment, it must then analyze data and using machine learning models to operate in an optimal way.
Industry robots typically execute simple task in a controlled environment. But if we make robots to work at home or streets, or self-driving cars, it is not anymore possible to properly pre-define rules for all environments and situations. Then it becomes more important that the robot is able to learn from its environment and past, and also share its learning to other machines. Deep learning is a part of machine learning and it especially focuses on more abstract models to learn, when it is not possible to define exact learning targets, or rules for optimal behavior.
Apple has acquired two AI companies, Perceptio and VocalIQ, this fall. The target might be to make Siri smarter, but it can be linked to Apple’s future products too. Google has also acquired AI companies, for example, DeepMind last year. Twitter and Facebook have also been active in this area. These are just a few examples, how big internet companies that have a lot of data are now investing in AI.
Fintech and digital finance services are also active in developing AI solutions. Trading is one area where AI tools have been developed for a longer time. The focus has especially been to utilize market data and news feeds and make optimal sell and buy decisions. The next generation fintech AI is able to do much more. When the whole finance sector has a transition now to digital services and processes, the transition opens many new opportunities to use data and machines that can learn.
In the future fintech machines will have a larger role to analyze and optimize risks, manage portfolios, price finance instruments, help people with their investment and lending needs, and build longer terms scenarios. With deep learning it is not necessary to limit tasks to limited data input feeds and simple trading transaction optimization, instead it is possible to basically take any data and make long term planning for different needs.
Internet of Things, IoT, is accelerating needs to have AI solutions. IoT itself is particularly to collect data and control actions. IoT is important for robots and self-driving cars, and for many applications and services like fintech, smart cities, and smart homes. It has been said IoT is especially about data and utilizing data. Soon we start to have so much data and so many things to control that it just becomes impossible to handle all that without machine learning. Then we see a new era when AI really comes to everyday life everywhere.
Big data has been a hot topic for a few years. But the real question is, how we can really utilize the data, otherwise it is quite useless. The utilization needs applications that can turn data into applications that help people, improve business and create new, more effective ways to do things. Machine learning, data science and AI have an important role in that work, especially when the amount of data and complexity of tasks grows rapidly.