EU GDPR enforcement

The Digital Omnibus Would Trade Workers' Best Evidence Tool for Modest Savings, and Narrower Fixes Exist

An ETUI study says the GDPR omnibus could let Uber-style employers refuse access requests as improper and defend automated deactivations more easily.

Digital Omnibus: Savings vs Worker Safeguards People of Internet Research · EU €5B Claimed administrative savings Commission estimate of savings by … 100+ Civil society organisations opposed Groups calling the proposal a cove… 4 Past workplace cases tested ETUI tested the amendments against… peopleofinternet.com
Digital Omnibus: Savings vs Worker Saf… People of Internet Research · EU €5B Claimed administrative s… 100+ Civil society organisations op… 4 Past workplace cases tested peopleofinternet.com

Key Takeaways

The strongest case for the Commission's Digital Omnibus is real. Data subject access requests do get abused. Controllers, especially small ones, face high volumes of requests, and some are tactical or meant to harass. Europe's compliance burden also falls hardest on firms without large legal teams. The Commission presents the package as a simplification and competitiveness measure, and estimates up to €5 billion in administrative savings by 2029 (Commission press release). A pro-innovation publication should take that seriously.

The question is whether the specific drafting delivers those savings without removing a tool that workers use to expose unlawful practices. A study for the European Trade Union Institute (ETUI) by Aida Ponce Del Castillo and Thomas le Bonniec, reported by Netzpolitik on 17 November 2026, argues that it does not.

What the ETUI study tested

The Commission proposed the omnibus without a full impact assessment. The ETUI researchers therefore tested the amendments against four past workplace cases. Their central finding concerns Article 15 access requests. The proposal could restrict requests made for purposes beyond data protection, such as gathering evidence for an employment dispute. Employers could then argue that a worker's request has a different purpose. Workers might have to make narrower requests, challenge refusals or hire lawyers to get the same information.

The platform-work cases show what is at stake. Six Uber drivers used access requests to obtain data on matching, pricing and ratings, and a similar case was brought against Ola. According to the study, the Amsterdam Court of Appeal upheld the drivers' claims in 2023 and ordered disclosure. The authors argue that similar outcomes would be harder to reach under the amendments. Ponce Del Castillo put it this way: "A worker may be right about unlawful surveillance and still be unable to prove it."

The second finding concerns automated decisions under Article 22. In a related Uber case, four drivers challenged account deactivations for suspected fraud. For three of them, the court found that Uber had not shown meaningful human intervention. The study says the proposed change would alter the necessity test. The mere possibility of a human alternative would no longer show that automated processing was unnecessary, and the authors say that would strengthen a platform's defence of fraud-flag deactivations.

The access-request drafting is the weak point

The EU's own data protection regulators have raised the same concern about purpose. In their joint opinion of 11 February 2026, the EDPB and EDPS said they support giving controllers legal clarity when they face abuse of rights. But they said that using access rights for purposes other than protecting personal data "should not be an element defining what an abuse is."

That position matches existing case law. In FT v DW (C-307/22, 26 October 2023), a patient requested records to pursue liability against their dentist. As summarised by Clyde & Co, the Court of Justice held that a request cannot be rejected only because the person pursues purposes unrelated to data protection. A controller can still refuse genuinely abusive requests under Article 12(5). The Court also noted that a controller holding a large quantity of data can ask the requester to specify what they want. The law therefore already has two tools for burdensome requests: the abuse exception and the specification request. A purpose test adds little against real abuse. Its main effect is to hand employers a standard objection to any request from a worker in a dispute.

Why this is a competitiveness issue too

Regulatory simplification only pays off if it cuts costs that do not buy accountability. Access requests are one of the few enforcement mechanisms that cost the state nothing. Workers and individuals can do the work themselves, which reduces the load on under-resourced regulators. If a test makes requests easier to refuse, disputes move to litigation and to regulators, and those costs are harder to count than the compliance savings the Commission advertises.

The automated-decision change carries a similar risk. Platforms that run fraud systems at scale have a legitimate interest in acting quickly. But a rule that makes automated deactivations easier to defend, with a worker's income at stake, rewards opaque systems over well-documented human review. Clear review standards would serve both sides better than a weaker necessity test.

The timing adds to the problem. The AI part of the package was adopted in July 2026 and delayed the AI Act's high-risk employment requirements to December 2027. If the GDPR changes follow, workers will have a longer period with fewer tools for challenging algorithmic management. More than 100 civil society organisations have called the proposal "an attempt to covertly dismantle Europe's strongest protections against digital threats." That language is polemical, but the ETUI's case-based test is harder to dismiss.

A narrower fix

The GDPR part is still being debated in the European Parliament, so there is room to change it. Three amendments would keep the legitimate aim and drop the harm:

The Commission should also complete the missing analysis. A targeted impact assessment of the access-request and Article 22 changes, with a count of the actual abusive requests they would stop, would let Parliament weigh the savings against the costs. Without it, a competitiveness measure risks reducing accountability while saving little.

Sources & Citations

  1. Netzpolitik: Proposed GDPR changes and workplace data abuses
  2. EDPB/EDPS: Digital Omnibus joint opinion news release
  3. European Commission press release IP/25/2718 on the Digital Omnibus
  4. Clyde & Co: ECJ C-307/22 on access requests