Artificial intelligence (AI) – the technology that’s revolutionising our everyday lives

AI has made great strides in recent years, leading to a boom in activity, attention and hype. In many areas, it is already a given in our daily lives, e.g., in the form of voice-operated assistants such as Apple Siri and Amazon Alexa, or extremely powerful translation services such as DeepL or Google Translate. Often, we don’t notice how much AI is involved in the form of extensive neural networks, for example, when we simply take a photo on a smartphone. AI is here. AI is developing rapidly, and it’s up to researchers, politicians and society to ensure that this technology – like so many other developments – serves humanity, individuals and the environment, rather than harming them.

Integration instead of separation

For ARS, AI is becoming an increasingly natural and organic element of software engineering. Particularly in cases where it’s an integral part of applications and functions. And where AI was previously only a specialist tool in the hands of data scientists, it can now, for instance, be the first step towards operationalising machine learning models in larger system contexts. In the last few years, there’s been a great deal of progress in the area of machine learning, to the extent that these tools can now also be deployed in traditional businesses. AI is no longer ‘rocket science’.

Every journey begins with a single step

Here at ARS, we help our customers take a neutral look at how they could start using artificial intelligence in their projects. Together, we check which projects can be implemented easily with a reasonable amount of effort (the low-hanging fruit), and which steps need to be taken to make them happen. And we always keep an eye on the big picture.

Data is the key

The basis for using AI is data to train the model. And the quantity and quality of the data play a key role in project success. Two classic starting points for AI projects:

  • You want to solve a problem using AI, but there is no data available yet. So the first thing to do is evaluate which data is needed, what format it should have and how it can be obtained.
  • You have data from various projects and sources that hasn’t been used yet. AI methods can reveal patterns that hadn’t yet been spotted, which can be used to optimise existing use cases or even create completely new ones.

Case studies

Customer projects we’ve worked on range from a simple automated evaluation of emails to voice-operated assistants.