The Proposal for an Artificial Intelligence Act COM (2021) 206 from a Consumer Policy Perspective

The Proposal for an Artificial Intelligence Act (AIA) of 21 April 2021, COM(2021) 206 final, is a landmark document for the regulation of AI worldwide. In the light of the massive risks for consumers (alongside opportunities) that come with the mass roll-out of AI systems it is of utmost importance to make sure this future instrument, together with other instruments, provides an adequate level of protection for consumers in the AI context.At first sight, the AIA Proposal seems to turn a blind eye on consumer interests as such. This impression is created, e.g., by the fact that the prohibition of manipulation by subliminal techniques and of exploitation of vulnerabilities is restricted to cases where such AI practices cause, or are likely to cause, physical or psychological harm (excluding mere economic harm). The same impression is created by the way the list in Annex III on high-risk AI systems is currently phrased, as this list includes credit scoring, but not the evaluation of factors similar to creditworthiness (such as complaints history), risk assessment by insurance companies or personalised pricing in general.The full text study can be accessed here: https://www.sozialministerium.at/dam/sozialministeriumat/Anlagen/Themen/Konsumentenschutz/Konsumentenpolitik/The-Proposal-for-an-Artificial-Intelligence-Act-COM2021-206-from-a-Consumer-Policy-Perspective_dec2021__pdfUA.pdf