We study the European Union’s (EU) recent responsible AI strategies and stakeholder responses to them. Responsible AI refers to AI that is designed and used in accordance with human values, such as transparency and accountability (Dignum, 2020). In recent years, the EU has positioned itself as a central player in striving for responsible AI (e.g., European Commission, 2020). Our empirical material includes document material (key EU strategy papers) and expert interviews on the EU’s responsible AI approach. Our main argument is that anticipation promoting responsible AI requires appropriately designed technical artifacts as the necessary condition, while the sufficient condition is provided by the social component, namely stakeholders’ technological frames, foresight frames, and expectation work. Thus, our study theorizes technical artifacts and anticipatory frames as crucial for anticipatory processes.