Approaches to trustworthy Artificial Intelligence

Great news this week, that the European Union  (EU) presented the draft proposal on the EU Artificial Intelligence Regulation and also the US Federal Trade Commission (FTC) made a clear statement that it shall take enforcement actions against companies that are selling and using biased algorithms.

In my former blogs I have stated that the future success of Artificial Intelligence (AI), depends not only on the further development of its current success factors. The growth of AI shall also be influenced by the ability of people to understand, to trust and being able to participate in AI-systems. Governments are currently taking action to limit the negative aspects that could limit the innovation capabilities and are providing guidelines to the use of AI-systems.

The EU presents in its draft proposal that AI-systems can be qualified as:

  • lower risk systems: these system can be subject to transparency requirements or are not regulated at all. For example chat bots they need to make clear that people are talking to a computer.
  • high-risk systems, where there can be a high risk of misuse as the system can be used to select and facilitate specific services (social security, insurance etc). High-risk systems are subject to strict requirements, where these systems need to be transparent and that human oversight is possible.
  • unacceptable risk systems should be banned when they are manipulative, are used for ethnic profiling and when it violates the European fundamental values.

The European regulation on AI is expected to be operational in 2023 or 2024. However a lot of details and definitions need to be specified and compliant to the European Court of Justice. When companies do not comply with the regulation they can expect a fine of € 30 million or 6% of their global revenue (highest number counts).

It seems that the FTC statement from last week can be operational in the immediate future. According to the announcement, the FTC plans to go after companies using and selling biased algorithms.

In my opinion it is good news that there is going to be more oversight on AI and serious regulations is under way.

Norbert Bol

Literature

European Commission: Europe fit for the Digital Age: Commission proposes new rules and actions for excellence and trust in Artificial Intelligence

FTC Issues New Guidance, Warning That Bias in Artificial Intelligence Could Create Potential Liability for Enforcement Actions

Photo by mikemacmarketing on Foter.com

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s