Eurofound's ERM database on restructuring-related legal regulations provides
information on regulations in the Member States of the European Union and Norway
which are explicitly or implicitly linked to anticipating and managing change.
Netherlands: Algorithmic management
Phase
Algorithm register
Native name
Algoritmeregister
Type
Algorithmic management
Added to database
18 October 2023
Article
-
Description
LAST UPDATE 2023 - THIS CONTENT WILL NOT BE UPDATED
The Dutch government has launched an algorithm register for all government agencies to enter information about their algorithms in one place. The register is accessible to everyone, and is meant to supply transparency about whether or not government algorithms stick to the rules.
The Dutch Digitalisation strategy highlights the importance of safeguarding constitutional rights and public values in cases where automated decision-making (ADM) is applied. The introduction of the Algorithm Register signifies a commitment to transparency and accountability in the governments' use of algorithms.
The government wants the use of algorithms to be responsible. The idea is that people should be able to trust that algorithms adhere to the values and norms of society and that there should be an explanation of how algorithms work. The government aims to achieve this by inspecting how algorithms function and any potential discrimination and arbitrariness in their use. Again, the idea is that when the government is transparent about algorithms and their application, citizens, organisations, and the media can critically monitor and ensure that this complies with the law and regulations.
The Algorithm register in the Netherlands is not yet complete. More and more government organisations are publishing their algorithms in the algorithm register. The goal is to have all government organisations connected to relevant algorithms. More information on which organisations these are and which algorithms they publish can be found in the algorithm register guide. Currently, governmental institutions are encouraged, but not obliged to add algorithms they have used in the registry. Ultimately, registration of impactful algorithms will become legally mandatory. The Dutch government is still working on a legal framework for algorithms. By 2025, the Dutch register will become mandatory once the preparatory work around the EU's AI act is finalised.
The following processes, among others, currently contribute to responsible algorithm use:
The Algorithm Register helps to make algorithms findable, better explainable and their application and results understandable.
The Algorithm Supervisor (the Dutch Data Protection Authority) coordinates the control of algorithms: do the government algorithms comply with all applicable rules?
The Ministry of the Interior and Kingdom Relations is working on the Implementation Framework 'Use of Algorithms'. This makes it clear to governments what requirements apply to algorithms and how they can ensure that their algorithms can meet those requirements.
The first version of the Algorithm Register was launched in December 2022. This forms the basis for further development. The government develops the register openly, together with experts, (potential) users and stakeholders. The Ministry of the Interior and Kingdom Relations is also continuing to develop the Algorithm Register website.
Commentary
Back in 2018, an evaluation by the Dutch Council of State underscored the increasing significance of Algorithmic decision making (ADM), while also highlighting potential risks where citizens cannot verify the rules that are being employed. Citizens are at risk of being profiled and confronted with decisions based on information of which the source is unknown. Citizen’s ought to have confidence that algorithms adhere to public values, legal requirements and standards, as well as being able to comprehend the rationale behind their outcomes.
As per 1 January 2023 a new regulator for algorithms has been introduced in the Netherlands, it has been housed within the Dutch Data Protection Authority (Autoriteit Persoonsgegevens, Dutch DPA).
On 31 August 2023, the Dutch Data protection Authority (AP) published a report on Algorithm Risks in the Netherlands, concerning systems and apps that use algorithms and artificial intelligence (AI), and the risks posed to individuals, groups and society. Furthermore, it demonstrated:
Bias and subsequent discrimination in law enforcement
Oversight of financial transaction, particularly in cases where a ‘false positive’ regarding a suspected illegal transaction has become evident.
The application of algorithms for municipal social benefits to estimate the potential risk for fraudulent activities related to the distribution of social benefits.
The report further recommended the necessity of committing to improve algorithm risk management such as through an algorithm impact assessment, highlighting the need for:
Implementing review mechanisms and frameworks in a manner that aligns with proportional requirements
Risk-oriented phasing in by organisations.
Additional metadata
Cost covered by
National government
Companies
Involved actors other than national government
National government
Involvement (others)
-
Thresholds
Affected employees: No, applicable in all circumstances Company size: No, applicable in all circumstances Additional information: No, applicable in all circumstances
This Eurofound research paper explores key trends in restructuring in recent years, highlighting the companies that announced the largest job losses and job gains in the EU. It builds on an analysis of company announcements recorded in Eurofound’s European Restructuring Monitor (ERM), alongside a new classification of restructuring events involving changes in company location.
Employers increasingly use tools such as email, SMS and messaging apps like WhatsApp or Signal to communicate with employees. While these technologies offer both efficiency and convenience, their use in communicating sensitive information, particularly for notifying employees of dismissal, raises legal concerns. This article explores the legal framework on dismissals across the EU, with a special focus on the use of digital means for communicating employment dismissals. Drawing on examples from various Member States, it examines the legal validity of digital dismissals.
In 2023, thousands of workers in big tech lost their jobs. Meta, Amazon, Google, Apple, Microsoft and Salesforce had been considered to offer good and secure jobs up to this point. Giants of the information and communication technology (ICT) sector, these companies are among the highest paying, with Eurostat data from 2022 indicating that workers in ICT had the second-highest median gross hourly earnings (surpassed only by earnings in the financial sector).[1] These layoffs were a shock, especially as the biggest companies had hired extensively during the COVID-19 pandemic. What happened in the two years after this redundancy wave – was that the end of the cuts or did the companies start expanding again?
In 2024, the automotive sector in the EU came to the fore in public and policy discussions. The focus was on the slowdown in electric vehicle (EV) sales, rising global competition, belated investments in new technologies, and the potential closure of production lines in Europe. A number of European car manufacturers and suppliers announced their intention to make large-scale redundancies and change long-standing collective agreements on job security and wages, while workers raised concerns amid demonstrations and industrial action.