Warum Ihr Unternehmen KI und Datenschutz ernst nehmen muss

We recently discussed some of the ethical challenges facing the development and deployment of artificial intelligence (AI) systems. One area of AI ethics in particular that demands more attention and scrutiny is data privacy.

This is especially true in the age of Facebook and its connection with Cambridge Analytica. In case you’re unfamiliar with this situation, Cambridge Analytica was involved in a scandal of harvesting 50 million Facebook profiles and using a sophisticated machine learning system to analyze the data and deliver highly targeted, personalized messages to change (it’s since been shut down). The scandal is that it collected user data illegally, and that data was used to manipulate voters in both the 2016 United States presidential election and the 2016 Brexit vote in the U.K. As a result, Facebook was recently fined $5 billion USD for poor data privacy practices that allowed the breach in the first place.

To be sure, that’s just a drop in the bucket for Facebook, a company valued at over $500 billion USD. Most businesses aren’t that big, of course, and most probably couldn’t weather a fine of that size. What’s more, with the introduction of Europe’s General Data Protection Regulation (GDPR), which requires companies doing business in Europe to comply with a number of data protection requirements or else face fines.
With serious legal and financial consequences at stake for companies using artificial intelligence, it’s essential to develop policies that ensure your business maintains compliance at all times. Following are two key areas of consideration for developing such a policy.

Understanding Your Data

The first step a data privacy policy is to know what data you’re collecting, how you’re analyzing it, and how you’re using it. A big part of this is knowing whether or not data is anonymized to provide insights into customer behavior, or if it’s not anonymized, how it’s being used to deliver a personalized experience. Collected data could be as simple as what web pages a customer visits on your website to far more detailed personal data such as lifestyle choices, political preferences, sensitive data such as health conditions, and more. What’s more, it’s a good idea to not only understand what data your business collects and uses, but also what data it obtains from third parties. Lastly, in terms of usage, it’s important to know if customer data is being used only internally or if it is being packaged and sold to third parties, as that can have an impact on compliance issues.

Understanding Legal Compliance

With more and more regulations on data collection and usage being developed, it’s essential to develop policies that ensure compliance, else face the fines of non-compliance. This includes examining the legality of:

  • Collection & warehousing practices: Does data collection and storage comply with legal requirements around privacy and data security?
  • Notification and opt-out practices: Do you sufficiently notify users of data collection practices and do you provide sufficient opportunity to opt-out of data collection?
  • User control: Can users whose data you are collecting control their data, what is used and who it is shared with? Can they change their preferences at any time or request their data to be deleted from your system?
  • Audit practices: What is your company doing to ensure ongoing compliance?
    What’s more, from a legal standpoint, it’s essential to ensure that users are made aware of and agree to whatever your data collection and usage practices are. This might be as simple as a privacy statement on your website, or it can be a more complex opt-in/opt-out system.

If your company is thinking about developing an AI program, it’s important to think through all the different implications of data collection and usage. And we can help. Contact us to discuss how we can work with you to ensure your AI program meets both your business needs and your customers’ privacy expectations.

Best regards,
Florian Horn

Gefällt Ihnen der Artikel?

Share on linkedin
Share on Linkdin
Share on xing
Share on XING
Share on twitter
Share on Twitter
Share on facebook
Share on Facebook
Ihre Daten werden gemäß unserer Datenschutzerklärung erhoben und verarbeitet.
Künstliche Intelligenz Partials
Data Analytics

Time Series Data Clustering Distance Measures

As ubiquitous as time series are, it is often of interest to identify clusters of similar time series in order to gain better insight into the structure of the available data. However, unsupervised learning from time series data has its own stumbling blocks. For this reason, the following article presents some helpful time series specific distance metrics and basic procedures to work successfully with time series data.

Weiterlesen »
Künstliche Intelligenz Parts
Künstliche Intelligenz

Unsupervised Skill Discovery in Deep Reinforcement Learning

Scientists from Google AI have published exciting research regarding unsupervised skill discovery in deep reinforcement learning. Essentially it will be possible to utilize unsupervised learning methods to learn model dynamics and promising skills in an unsupervised, model-free reinforcement learning enviroment, subsequently enabling to use model-based planning methods in model-free reinforcement learning setups.

Weiterlesen »