From production to the product, from sales to the legal department: artificial intelligence (AI) opens up new ways for the automotive industry to further improve the customer experience and make processes more efficient. To make responsible use of these opportunities, Daimler AG is the first automotive manufacturer to establish principles for handling AI. “Responsible use”, “Explainability”, “Protection of privacy” and “Safety & Reliability” are the four guiding principles according to which the company develops and uses AI.
“Our four principles for dealing with artificial intelligence create a clearly defined scope for action. We thus make it clear that the human being will remain the pacemaker of technical progress,” says Renata Jungo Brüngger, Member of the Board of Management of Daimler AG for Integrity and Legal Affairs. “The principles are not mere theory, but are practical and responsible guidelines for day-to-day use,” Jungo Brüngger continues.
The four principles for dealing with artificial intelligence
An inter-disciplinary team that included experts from Development, IT and Legal Affairs developed the four principles for dealing with artificial intelligence at Daimler, so as to take into account technical, legal and social aspects.
Behind the principle of “Responsible use” is the maxim that the opportunities and effects of AI must be carefully weighed up. The use of AI at Daimler is therefore preceded by a transparent assessment process: Will products, services and processes actually become faster and more efficient with AI? What are the benefits for different target groups? These questions are examined during a risk-based technology impact assessment. Responsible use of AI also means creating a high-quality, representative database. This means that the data used must take the factors relevant to the application into account. This fosters fairness and inclusion.
The principle of “Explainability” creates the transparency necessary to have trust in new technologies. Customers and employees of Daimler AG must be able to recognise when they come into contact with AI – e.g. by information in the General Terms & Conditions, or by direct messages in the system. The principle also stands for comprehensibility: it must be clear which parameters are included in data analyses. This makes it easier to evaluate the results of AI-based calculations, and avoid unwanted bias.
With the principle “Protection of privacy”, Daimler is reaffirming that data protection is a feature of quality for the company. The user must have control over access to and use of their data – as is the case with e.g. the MBUX multimedia system. Artificial or anonymised sets of data are used where possible when testing and training AI.
The principle “Safety & Reliability” means that the high quality standards of Daimler also apply to applications in the digital world. The company develops and tests AI to the latest scientific and technical standards. The basis for this is conscientious training of learning systems, the use of scientifically validated algorithms and up-to-date technologies.
AI principles and data responsibility
The AI principles are part of a comprehensive approach for the responsible handling of data at Daimler. Alongside a guideline for data and a data-related compliance management system, this includes a culture that sensitises employees to the importance of data. In this way Daimler is shaping the mobility of the future in a proactive and legally secure way.