The article discusses the potential impact of the European Union's General Data Protection Regulation (GDPR) on the use of machine learning algorithms. The GDPR, set to take effect in 2018, will restrict automated individual decision-making that significantly affects users and will create a "right to explanation," allowing users to request an explanation for algorithmic decisions made about them. The authors argue that while this law poses significant challenges for industry, it also presents opportunities for computer scientists to design algorithms and evaluation frameworks that avoid discrimination and enable transparency. The GDPR's Article 22 specifically prohibits automated decision-making based on personal data if it significantly affects the data subject, with exceptions for contract performance, legal authorization, and explicit consent. The regulation also requires data controllers to provide appropriate safeguards, including the right to human intervention and the ability to contest decisions. The article highlights two key issues: non-discrimination and the right to explanation. Non-discrimination is a core principle in the EU, and the GDPR's requirements on profiling aim to prevent discriminatory effects. However, achieving this goal is challenging due to the complexity of data correlations and the potential for uncertainty bias in underrepresented groups. The right to explanation is another critical aspect. The GDPR mandates that data subjects have the right to access information about their data and receive meaningful information about the logic involved in automated decisions. This raises questions about how to explain complex models like neural networks, which are difficult to interpret. The authors suggest that research is needed to develop methods for quantifying the influence of input variables on outputs, making algorithms more transparent and interpretable. Overall, the GDPR presents both challenges and opportunities for machine learning, emphasizing the need for algorithms to be transparent, fair, and explainable.The article discusses the potential impact of the European Union's General Data Protection Regulation (GDPR) on the use of machine learning algorithms. The GDPR, set to take effect in 2018, will restrict automated individual decision-making that significantly affects users and will create a "right to explanation," allowing users to request an explanation for algorithmic decisions made about them. The authors argue that while this law poses significant challenges for industry, it also presents opportunities for computer scientists to design algorithms and evaluation frameworks that avoid discrimination and enable transparency. The GDPR's Article 22 specifically prohibits automated decision-making based on personal data if it significantly affects the data subject, with exceptions for contract performance, legal authorization, and explicit consent. The regulation also requires data controllers to provide appropriate safeguards, including the right to human intervention and the ability to contest decisions. The article highlights two key issues: non-discrimination and the right to explanation. Non-discrimination is a core principle in the EU, and the GDPR's requirements on profiling aim to prevent discriminatory effects. However, achieving this goal is challenging due to the complexity of data correlations and the potential for uncertainty bias in underrepresented groups. The right to explanation is another critical aspect. The GDPR mandates that data subjects have the right to access information about their data and receive meaningful information about the logic involved in automated decisions. This raises questions about how to explain complex models like neural networks, which are difficult to interpret. The authors suggest that research is needed to develop methods for quantifying the influence of input variables on outputs, making algorithms more transparent and interpretable. Overall, the GDPR presents both challenges and opportunities for machine learning, emphasizing the need for algorithms to be transparent, fair, and explainable.