LearnerShape Joins Study on EU AI Act

LearnerShape is pleased to have participated in an Open Loop study of the proposed EU AI Act for which a report was recently published – Artificial Intelligence Act: A Policy Prototyping Experiment. Our contributions are quoted in the report.

The EU AI Act is a European Commission proposal for an EU regulation that would regulate AI applications in three categories:

• unacceptable risk – using subliminal techniques, exploiting vulnerable groups, using social scoring leading to unjustified detrimental treatment, or using real-time biometrics in public spaces for law enforcement purposes
• high-risk – in specified sectors, including biometrics, critical infrastructure, education, employment, public services, law enforcement, migration and border control, and administration of justice
• unregulated – not within the above-specified uses.

Unacceptable risk activities would not be permitted in the EU, and high-risk activities would be subject to various regulatory requirements. As the EU AI Act is currently drafted, LearnerShape believes that at least some of our activities would be regulated by the Act, particularly because we focus on the education and employment sectors.

The Open Loop study focused on these EU AI Act requirements for high-risk activities. The study concluded that many of these proposed requirements may be impractical, overbroad or unclear, and we agree. While LearnerShape strongly support a responsible approach to AI – indeed this is central to our approach – we do not support prescriptive regulatory requirements that will make it more difficult for EU AI companies to be global leaders. We look forward to further public dialogue on these issues.

Maury Shenk, Founder & CEO, LearnerShape