The EU Parliament accepted the Synthetic Intelligence (AI) Act at the moment. Member states agreed upon the regulation in December 2023. At present, members of the European Parliament endorsed the act, with 523 voting in favor, 46 voting towards, and 49 abstaining from the vote.
It’s no secret that AI is a double-edged sword. For each optimistic use case, there are a number of methods people can use the expertise for nefarious functions. Regulation is usually efficient in creating safeguards for the adoption of recent applied sciences. Nevertheless, delineating the boundaries of AI’s purposes and capabilities is difficult. The expertise’s huge potential makes it troublesome to get rid of unfavourable makes use of whereas accommodating optimistic ones.
Due to this, the European Union’s new Synthetic Intelligence Act could have each optimistic and unfavourable impacts on banks and fintechs. Organizations that be taught to adapt and innovate inside the boundaries will see probably the most success in relation to leveraging AI.
That mentioned listed here are 4 main implications the brand new legislation could have on banks:
Prohibited AI purposes
The brand new legislation prohibits using AI for emotion recognition within the office and colleges, social scoring, and predictive policing based mostly solely on profiling. This can affect how banks and fintechs use AI for buyer interactions, underwriting, and fraud detection.
Compliance and oversight
The ruling particularly calls out banking as an “important personal and public service” and categorizes it as a high-risk use of AI. Due to this fact, banks utilizing AI methods should assess and scale back dangers, keep use logs, be clear and correct, and guarantee human oversight. The legislation states that residents have two main rights in relation to using AI of their banking platforms. First, they will need to have the power to submit complaints, and second, they’ve the proper to obtain explanations about selections made utilizing AI. This can require banks and fintechs to boost their threat administration and replace their compliance processes to accommodate for AI-driven companies.
Transparency
Banks utilizing AI methods and fashions for normal functions should meet transparency necessities. This contains complying with EU copyright legislation and publishing detailed summaries of coaching content material. The transparency reporting is not going to be one-size-fits-all. In response to the European Parliament’s clarification, “The extra highly effective normal function AI fashions that might pose systemic dangers will face further necessities, together with performing mannequin evaluations, assessing and mitigating systemic dangers, and reporting on incidents.”
Innovation help
The legislation stipulates that regulatory sandboxes and real-world testing might be accessible on the nationwide degree to assist companies develop and practice AI use earlier than it goes stay. This might profit each fintechs and banks for help in testing and launching their new AI use circumstances.
General, the EU AI Act isn’t requiring something outdoors of banks’ current capabilities. Monetary establishments have already got processes, documentation procedures, and controls in place to adjust to current laws. The act will, nevertheless, require banks and fintechs to both set up or reassess their AI methods, guarantee compliance with new laws, and adapt to a extra clear and accountable AI ecosystem.
Picture by Tara Winstead
