Status quo Artificial Intelligence: Topic track on solutions.hamburg

Humanoid robots that do justice before the criminal court. Machine doctors who operate open-heart surgery. Autonomous weapon systems that determine war and peace. Technologically, this is already conceivable and feasible. It is only a matter of time before the first autonomous systems, thanks to Artificial Intelligence (AI), take on tasks that previously only humans could accomplish. But do we want to?

Norderstedt, September 12, 2017 – Full house at solutions.hamburg: More than 200 visitors came to the theme track AI at the invitation of Lufthansa Industry Solutions. Top-class experts gave answers to current developments in the field in five lectures and a subsequent panel discussion. Neural networks, data analytics and IoT play a special role in this context. But not only know-how and feasibility were topics of the event, but also the question of morality and ethics: “Artificial Intelligence: How far will we go?”

“Neural networks are the basis for machine learning. These network models are among the most successful techniques for learning interactions between humans and robots. In concrete terms, this means that we teach social robots to learn in the same way as children - by imitating adults or, in this case, their teachers,” said Prof. Stefan Wermter from the University of Hamburg. “This is how self-organizing neural networks for the recognition of human actions are created.”

The start-up WUNDER from Berlin is developing a neural network of a completely different kind: the “shopping brain”. WUNDER uses AI playfully to generate content and cognitive personalize communication. Artificial intelligence makes product discovery more vivid, personal, influential, direct, flexible and relevant than ever before. Connecting people and products is done in real time using new learning techniques to ensure that customers receive the right products.

AI in Companies

Whether self-propelled cars, robots or digital assistants like Alexa: they are all based on the use of artificial intelligence. Some of these practical examples were presented by the speakers and demonstrated that AI has arrived in the companies. For example, Lufthansa Industry Solutions has equipped escalators with sensors to detect anomalies based on the audio signals. The recorded and analyzed noises indicate whether the escalator is fully functional or whether a defect has occurred. Self-learning algorithms not only allow the anomalies to be detected, but also to be analyzed with increasing precision. This self-learning condition monitoring makes predictive maintenance possible with audio classification.

Digital Ethics

However, there are still some challenges to be mastered in the area of AI, so that these were the focus of the panel discussion. “We need a digital ethics,” emphasized Wolfgang Dorst, Head of the Industrial Internet division of the Bitkom digital association. “At the moment, there is no legal certainty for AI. The increased use of AI also creates new requirements for employees in companies themselves. People have to learn to collaborate with machines at eye level," said Sven Meier, Digital Transformation Lead at EnBW. For software developers, AI opens up new career opportunities.

“There are already a lot of AI solutions on the market, the topic is not new - but what is missing in the corporate sector, however, are the appropriate problems. We must first identify these,” said Dr. Harald Kolbe, Head of Innovation Management at Lufthansa Technik Logistik Services, summarizing the consensus of the participants. All experts also agreed: “AI takes companies to the next level of development. The risk of missing the development and not being there is simply too high,” says Dr. Lars Schwabe, Associate Director of Lufthansa Industry Solutions.