Eligible locations include Albania, Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Kosovo, Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, Moldova, Monaco, the Netherlands, North Macedonia, Norway, Poland, Portugal, Romania, San Marino, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, the United Kingdom, and the United States. Selected professionals will review AI-generated responses across sensitive and ambiguous scenarios, apply structured safety policies and rubrics, identify behavioural failures, and provide detailed feedback that supports safer and more reliable model performance.