AW AlgorithmWatch gGmbH
AW
AlgorithmWatch is a human rights organization based in Berlin and Zurich.
ID: 268308931088-81
Lobbying Activity
Response to Digital package – digital omnibus
14 Oct 2025
Mit digitalpolitischen Rechtsrahmen wie dem DSA, DMA und AI Act konnten wichtige Schutzstandards EU-weit auf den Weg gebracht werden. Deren konsequente und schnelle Umsetzung sollte jetzt die Priorität darstellen. Diese Gesetzestexte vereinen mühsam erarbeitete Standards und Verantwortlichkeiten, deren Rücknahme in materieller Hinsicht nur bedeuten kann, grund- und menschenrechtliche Verantwortungen zu missachten und wichtige Meilensteine in verantwortungsbewusstem Einsatz von neuen Technologien dem Rücken zu zukehren. Des weiteren sind die Mitgliedsstaaten bereits mit der Umsetzung des AI Acts und der Einrichtung nationaler Behörden weit fortgeschritten. Einige haben in den letzten Wochen ihre nationale Gesetzgebung auf den Weg gebracht, andere sind gerade dabei. Jede Änderung bezüglich des zeitlichen Ablaufs der Implementation könnte möglicherweise divergirende Regelungen oder Ausnahmeregelungen innerhalb der EU schaffen, was zu mehr Komplexität und Unsicherheit führen und das zentrale Ziel der Harmonisierung untergraben würde. Den AI Act, dieses erst 2024 verabschiedete Gesetz, jetzt zu öffnen, führt unterm Strich also zum Verlust von Planungssicherheit - für Unternehmen und Gesellschaft. Die beteiligten Stakeholder, ob es staatliche Akteur:innen, Unternehmen oder Bürger:innen benötigen zu diesem Zeitpunkt Stabilität und Konsistenz in der Umsetzung der Implementation. Ein koordinierter Austausch mit zuständigen nationalen Aufsichtsbehörden ist dafür zielführend. Das vorgeschlagene Digital Omnibus-Paket als Teil des Simplification Package droht dieser Rechtssicherheit jedoch entgegenzuwirken. Unter dem Deckmantel der Vereinfachung besteht die Gefahr, mühsam erkämpfte Rechte etwa in der DSGVO oder der ePrivacy-Verordnung sowie die institutionelle Verantwortung marktmächtiger Akteur:innen nach DMA, AI-Act und DSA zu unterminieren. AlgorithmWatch fordert daher eine konsequente Durchsetzung und kohärente Anwendung des bestehenden Rechts, statt Deregulierung und Rechtsunsicherheit. Auf dem Spiel steht nichts weniger als die Glaubwürdigkeit des Gesetzes, des Gesetzgebungsprozesses und der Kommission als Aufsichtsbehörde, da die Wirksamkeit und Stärke eines jeden Gesetzes in seiner konsequenten Durchsetzung und Umsetzung liegt.
Read full responseResponse to European Democracy Shield
26 May 2025
For full feedback see PDF Zukunft D (Future D) is a newly launched initiative by five civil society organizations in GermanyAlgorithmWatch, aula, Liquid Democracy, the Schwarzkopf Foundation Young Europe, and Wikimedia Germanywith the aim of strengthening democratic structures and competences in the digital sphere. As part of the German federal program Demokratie leben! (Live Democracy!), Zukunft D promotes democracy in a digitalized society by fostering knowledge sharing, developing policy proposals, nationwide networking, and the empowerment of underrepresented groups. The project focuses on increasing civic participation, enhancing digital literacy, and creating inclusive, secure, and accessible democratic infrastructures across the country. We welcome the European Commissions initiative to launch the European Democracy Shield as an important step in reinforcing democratic resilience across the European Union. In particular, we support the emphasis on ensuring citizens access to quality information, securing the integrity of elections, and strengthening public trust and participation in democratic processes. Zukunft D supports a strong and proactive approach to defending democracy against evolving threats. Research within our organizations has repeatedly shown that automated systems used by online platforms can systematically distort public discourse, promote polarization and lead to a diffusion of responsibility for decisions that deeply affect users. These dynamics cannot be adequately addressed without robust transparency, independent oversight, and meaningful public engagement. Recommendations: - Enforce DSA provisions on platform accountability, especially data access for independent researchers. - Curb tech monopolies. EU regulation should promote a more diverse and open landscape in AI. This includes breaking up monopolies, fostering a level-playing field, and ensuring contractual fairness along the AI value chain. - Promote transparency and access to information, including the legal right to open data. - Provide full transparency on all AI systems used in the public sector by ensuring strong accountability mechanism in the AI Act (e.g. via guidelines) and clear guidance for an effective fundamental rights impact assessment. - Ban the use of remote biometric identification systems, by addressing exemptions for state actors. - Prioritize open and free software in EU procurement to strengthen digital sovereignty and European independence. - Support programs that strengthen democratic skills from a young age and ensure that public institutions such as schools incorporate meaningful participation. - Empower non-profit civil society organizations that are vital for grassroots engagement and reduce legal burdens for civic engagement and volunteer work. - Develop and test novel social and technical tools, emphasizing public, open-source and interoperable platforms technologies and interoperable technologies for transparency, reusability, and collaboration. - Establish low-threshold, multilingual participation processes, that come to productive and binding agreements ready for implementation. - Promote digital literacy and skills for effective online participation - Expand the use of deliberative approaches like citizens' assemblies. - Prioritize open and free software in EU procurement to strengthen digital sovereignty and European independence.
Read full responseResponse to Delegated Regulation on data access provided for in the Digital Services Act
9 Dec 2024
AlgorithmWatch is a non-governmental, non-profit organization based in Berlin and Zurich. We fight for a world where algorithms and Artificial Intelligence (AI) do not weaken justice, human rights, democracy and sustainability, but strengthen them. Attached is our response to this consultation, building on our wider work on the DSA and its implementation, as well as our ongoing efforts to access and use data for public interest research including practical experience of an Article 40(4) style request, under the DSA Election Guidelines. Our response is organised in three parts: a background section on the DSA & Research in Practice, building on our experiences; specific proposals on the Draft; and an Annex with input from other researchers about potential data types for inclusion in Recital 12. As background, we strongly welcome efforts to reverse increasing limitations by many VLOPs and VLOSEs on access to data, despite the importance of said data to public interest research. From our experience, it is important to balance (i) secure and streamlined processes for straightforward requests and (ii) opportunities for mediated collaboration and negotiation for more bespoke needs. We believe the Delegated Act, as it is written, accounts well for these twin requirements, though requires some clarification and specification. We particularly welcome that (i) no further specifications on research requirements have been added beyond Article 34 and 35 of the DSA and (ii) vetting seems to be based on security requirements for particular types of projects and data, rather than blanket vetting for types of organisations. This gives flexibility for CSOs and public-interest journalism to also be involved in research projects which are appropriate to their level of technical capabilities. This will be important to increase the diversity of research projects, which is necessary to properly investigate the diversity of risks and best practises relevant to the DSA. We propose some amendments to ensure this is clear, explicit, and held to in implementation. We also welcome the drafts use of expert judgement from DSCs, supported by external expertise, rather than standardised checklists; though we also propose changes to ensure oversight and balanced application of DSCs judgements, and to avoid bottlenecks (particularly in Ireland). We also propose ways by which DSCs, and others, can ensure that research supported by Article 40(4) contributes to (i) initial insights from other transparency reporting, particularly into organisational practices of VLOPs and VLOSEs and (ii) locating inaccuracies in information provided, either under Article 40 or other transparency requirements. Finally, we propose changes regarding applicants protection of potential overexposure of their research data, and abilities to appeal decisions. We see this draft as a good step in the implementation of the DSA, and welcome opportunities to continue this work. Such regulations will always require learning together from experience and improving accordingly, for all actors involved. This needs to be done in an appropriately iterative, transparent, and structured fashion to ensure the most effective, efficient, and accountable rollout of this important regulation.
Read full responseMeeting with Alexandra Geese (Member of the European Parliament) and European Digital Rights and
25 Oct 2023 · Greens/EFA event "Reclaiming our Digital Future"
Response to Performance of independent audits provided for in the Digital Services Act
2 Jun 2023
AlgorithmWatch and AI Forensics welcome the opportunity to provide feedback to the European Commissions work on a Delegated Regulation, which further specifies principles and procedures for independent, second-party audits within the Digital Services Act (DSA) framework. Our joint feedback (in the attached document) follows from our organisations respective firsthand experiences conducting independent, third-party audits of platforms algorithmic systems. Our feedback focuses on three points: 1) Independence and expertise of second-party auditors, 2) Recognition of independent, third-party auditors within the auditing framework, and 3) Publicly disclosed audit reports.
Read full responseResponse to Delegated Regulation on data access provided for in the Digital Services Act
26 May 2023
AlgorithmWatch welcomes the opportunity to provide written evidence on facilitating data access for researchers under Article 40 of the Digital Services Act. The recommendations detailed in the attached document were developed through AlgorithmWatch's dedicated research and policy work on the DSA, and includes insights from the Governing Platforms Project and our extensive firsthand experience conducting third-party algorithmic audits of major platforms using data donations. We strongly believe that platforms past and present actions vis-à-vis public interest research make clear that they cannot be relied upon to voluntarily share data with independent researchers in a consistent or meaningful way, which has been a major impediment to holding platforms accountable for the risks their algorithmic systems pose individuals and society. A strong access framework is therefore necessary to ensure adequate and enforceable standards for data sharing in the public interestboth in terms of platforms data transparency and access infrastructures as well as in the vetting of researchers and research proposals. Our submission centers on six main areas to help inform a strengthened data access framework: 1) empowering a broad base of vetted researchers, 2) governance documentation as data, 3) vetting exemption requests, 4) defining and ensuring reliable access to publicly available data, 5) protecting independent research from platform abuse, and 6) independent advisory mechanisms to support data access applications & vetting.
Read full responseMeeting with Alexandra Geese (Member of the European Parliament)
25 May 2023 · CPDP: THE GLOBAL HARMS OF POWERING ARTIFICIAL INTELLIGENCE – TOWARDS A SUSTAINABLE FUTURE OF DATA USE AND GOVERNANCE
Meeting with Alexandra Geese (Member of the European Parliament)
1 Oct 2022 · Event: Bits und Bäume
Meeting with Brando Benifei (Member of the European Parliament, Rapporteur)
9 Feb 2022 · Discussion on the AIA
Meeting with Sergey Lagodinsky (Member of the European Parliament, Shadow rapporteur)
8 Feb 2022 · Artificial Intelligence Act
Meeting with David Cormand (Member of the European Parliament)
7 Feb 2022 · IA
Meeting with Kim Van Sparrentak (Member of the European Parliament)
12 Oct 2021 · Digital Services Act