What are the Artificial Intelligence project topics

Artificial Intelligence

"Every process and every user interface must be AI-enabled in the future," demanded Frank Riemensperger at the Big Data and AI Summit of the Bitkom industry association. "The entry barrier for artificial intelligence is low," said the chairman of the Accenture country group Germany, Austria and Switzerland, encouraging the approximately 1200 participants. Companies today could get into the topic without major upfront investments, for example with Google's TensorFlow framework.

At the moment, however, when it comes to artificial intelligence (AI), music doesn't necessarily play in Germany. Riemensperger pointed out that China invests the most in AI initiatives in an international comparison. German companies ran the risk of missing the boat.

"Artificial intelligence will play a key role for German industry and usher in the second wave of digitization," predicted Andreas Dengel from the German Research Center for Artificial Intelligence. But many see it as a purely technical issue. The scientist advises companies to experiment more and focus on quick wins as a first step.

Lufthansa: "Don't start with AI technology"

As is so often the case in this country, there is less research activity than practical implementation. "Don't start with AI technology," warned Marcel Kling, who heads the "smile" personalization program at Lufthansa. The search for the "golden AI method" has already stalled many projects. Instead, decision-makers should start with a specific application and a clearly identifiable economic benefit.

The "smile" program that Lufthansa has been pursuing since 2014, for example, aims to improve the customer experience. To do this, all touchpoints have to be seamlessly connected, explained Kling. The same information base must be available everywhere in the group where customer interaction takes place. Today there are often breaks in customer dialogue, which often lead to angry passengers. Above all, Lufthansa wants to use artificial intelligence to respond more individually to customers.

The first "smile" applications include, for example, personalized flight offers and individual additional services. For example, a customer on the way to the terminal could be offered a place in the lounge at the right time via an app. So-called "Customer Retention Services" go one step further. An AI-based system should learn from negative service experiences such as a delay and make compensation proposals to the passenger based on their individual preferences and preferences. This can be a bottle of champagne or an upgrade to a higher class.

Of course, a program like smile also needs a technical infrastructure. To this end, Lufthansa relies on a microservices architecture. Kling's team also sees itself in the Lufthansa Group as a service provider who uses AI to make "Decision as a Service" available to other areas for different customer scenarios. In the ideal case, smart systems could automatically decide which products are offered to a customer, when and in which form.

At the Otto Group and Consorsbank, too, technology was not primarily the starting point for AI initiatives. For example, Otto offers its customers aggregated product reviews from other customers as a service on the website. For this purpose, the company has developed a machine learning system that is based entirely on open source software. The Nuremberg-based online bank Consorsbank uses machine learning to try to find out more about its online community and, for example, to identify the most important influencers in specialist discussions.