AI Governance in Financial Groups: Blueprint from the Sparkassen-Newsroom

S-Communication Services GmbH (S-Com), the communication unit of the Sparkassen-Finanzgruppe, produces guide articles, campaign pages and consumer information for sparkasse.de on behalf of about 340 German savings banks. Two years ago, the project manager completely dismantled this editorial process, examining 17 individual steps, more than half of which were manual handovers between Jira, Word, Slack and the CMS. Common cloud-based automation tools failed early in the optimization process due to a lack of control over server location and no self-hosting capability. Every new tool at S-Com must pass an internal compliance framework - an assessment with over 100 questions covering IT security, works council and data protection. This practice was already part of the corporate culture before the AI Act became relevant for most companies. In the end, only one automation tool that can be operated entirely within the company's own IT infrastructure remained. On this basis, an architecture was built that automates editorial work steps without relinquishing control over sensitive data. More than half of the steps now run automated, with part still in test operation. Depending on the content piece, the process is 10 to 20 percent faster, and an external quality assurance service provider has already been saved. However, automation does not replace everything; some steps remain deliberately with humans. The regained time mainly flows into editorial diligence: tone, context and a brand voice that cannot be arbitrarily replaced by a language model. In this t3n PRO use case, the company shows how AI governance in a regulated corporate environment becomes ongoing practice: why no tool at S-Com goes into production without an assessment over the entire lifecycle, which criteria decide which AI model may be used for what, and why only part of the process steps are deliberately kept automated.