Access to Public Information, Defamation / Reputation, Digital Rights, Privacy, Data Protection and Retention
Denegri v. Google Inc (Appellate Court)
On Appeal Mixed Outcome
Global Freedom of Expression is an academic initiative and therefore, we encourage you to share and republish excerpts of our content so long as they are not used for commercial purposes and you respect the following policy:
Attribution, copyright, and license information for media used by Global Freedom of Expression is available on our Credits page.
The General Court of the EU recognised that the European Commission’s Research Executive Agency (REA) had failed to properly assess a number of access to information requests filed by digital rights activist and a member of the European Parliament Patrick Breyer. Mr. Breyer requested for access to documents related to the EU-funded iBorderCtrl, a research project aiming to develop a “video lie detector” based on “artificial intelligence” to be used on people travelling to the EU and to detect whether people were lying when answering questions. The REA had denied access to the information on grounds that the disclosure would have undermined the protection of the commercial interests of the consortium of companies developing the technology for the project. The General Court ruled that the harm to the commercial interests of the consortium members outweighed the public interest in having such information disclosed as the requested documents related to the early stages of the research project. The Court arguably ruled that the public interest in the disclosure of information existed only once the innovations and research had been completed.
On April 19th 2016, the European Research Executive Agency (REA) kicked off a research project with a consortium of companies to develop an EU-funded project called “iBorderCtrl: Intelligent Portable System” as part of the Horizon 2020 framework in a 36-month period. The aim of the project was to test new technologies in controlled border management scenarios. In particular, it related to the development and testing of emotion recognition technologies, including a video lie detector to assess whether people lied to border agents. As part of the funding process and project implementation, several documents related to the development phase were received by REA from the consortium members.
On November 5th 2018, Patrick Breyer, pursuing Regulation 1049/2001 regarding public access to European Parliament, Council and Commission documents, requested the European Commission to access documents related to the grant agreements of the iBorder Ctrl and the following development of the iBorderCtrl project. The request was transmitted to REA.
On November 23rd 2018, REA replied to Patrick Breyer that full access was granted to one document, partial access to another one but it was refused for those documents under two of the exemptions listed in Article 4 of the Regulation, namely the protection of personal data and the commercial interests of the consortium members.
On November 26th 2018, Patrick Breyer replied to the European Commission accepting the release of documents without disclosing any personal information. On January 17th 2019, REA refused to grant him access to documents in order to protect the commercial interests of the members of the iBorderCtrl consortium following Article 4 of Regulation 1049/2001 and Article 3 of Regulation 1290/2013 laying down the rules for participation and dissemination in “Horizon 2020 – the Framework Programme for Research and Innovation (2014-2020)”.
On March 15th 2019 , Patrick Breyer appealed against the European Commission’s denial to access information before the General Court of the EU complaining against a violation of Articles 4 “exceptions”, Article 7 “processing of initial applications” and Article 8 “Processing of confirmatory applications” of Regulation 1049/2001.
The main issue before the Court was whether the Research Executive Agency (REA) had properly assessed the information requests submitted by the applicant as they related to the processing of the requests under Articles 7 and 8 of Regulation 1049/2001 and exceptions under Article 4 of Regulation 1049/2001.
Violation of Articles 7 and 8 of Regulation 1049/2001
The applicant complained that the REA violated Articles 7 and 8 as both the initial and review requests regarded documents produced during the iBorderCtrl project. The same set of documents were requested in the initial and review requests.
The applicant also complained that REA did not examine in full the access to information request as it omitted the part about documents authorising the iBorderCtrl project. He argued that the review request specifically mentioned that it was following his initial request and there was no explicit mention that he was reducing the scope of his initial request. He was still requesting access to the grant agreements.
The Court agreed with the applicant and held there was no indication for REA to presume the applicant was withdrawing the request to access documents related to the grant agreements in his confirmation application. On the contrary, he explicitly said that the request was following up on his initial request. This was also confirmed by the fact the requester had agreed that documents that included personal information would have been released with that information deleted. The Court then declared REA’s reply invalid for the section related to the request to access authorisation documents of the iBorderCtrl project.
Violation of Article 4 of Regulation 1049/2001
This ground included two main arguments for appeal against the REA’s denial to access documents: first, the lack of harm to the commercial interests of the consortium members of the iBorderCtrl and second, the existence of a public interest overriding the harm that would ensure the release of the documents mentioned in the applicant’s access to information request.
Preliminary considerations related to the application of Regulation 1290/2013
Parties first disputed which Regulation was applicable to examine this argument- whether it was Regulation 1049/2001 on access to documents of EU institutions or Regulation 1290/2013 on the rules for participation and dissemination in “Horizon 2020 – the Framework Programme for Research and Innovation (2014-2020).” REA argued Regulation 1290/2013 is lex specialis and therefore should prevail over provisions of general application which are in Regulation 1049/2001. Therefore, according to REA, contracts signed in the framework of Horizon 2020 such as the ones related to the iBorderCtrl project containing personal data should fall under Article 3 of Regulation 1290/2013 and therefore, they could not be either released or disseminated. The applicant argued that Regulation 1049/2001 and the exceptions regime laid down in Article 4 as well as the principle of maximum disclosure should be considered complementary and be compatible with the interpretation of Regulation 1290/2013.
The Court considered that Regulation 1049/2001 applies to all documents held by REA and that the principle of maximum disclosure implied that all documents should be accessible to the public except for very limited circumstances strictly defined in Article 4 of the Regulation. The exemption regime is based on a balance between competing interests, namely the interest in the release of the information concerned and the interest that would be harmed by such release. When assessing an access to information request, a balancing test should be drawn to see what interest is at stake in the release of the information concerned. Hence, the EU body receiving an access to information request should assess whether any of such protected interests exist and whether the release of the information would bring serious harm to one of those protected by Article 4. The Court had previously held that EU bodies could make their assessment on the basis of presumptions applicable to certain categories of documents without analysing the particular document concerned.
In the case at hand, REA invoked the commercial interest exemption under Article 4 of Regulation 1049/2001 and Article 3 of Regulation 1290/2013 that protected any data, knowledge and information part of an action as confidential information as in the case of the iBorderCtrl project for 36 months since the beginning of the action. This time was not over when the applicant filed his information request. The rationale behind Article 3 was to protect documents produced during the action to be disseminated to third parties in order to ensure the action succeeds. The Court reminded REA that its interpretation of Article 3 presented during the proceedings implied a presumption that any document produced as part of an action could not be released but this interpretation was not mentioned in its reply to the access to information request. Hence, this argument was rejected by the Court.
It went on by considering, on the contrary, that the interpretation of Article 3 of Regulation 1290/2013 suggested by the applicant would imply that the principle of maximum disclosure of Regulation 1049/2001 would always prevail over the confidentiality of documents under the scope of Regulation 1290/2013. This must be rejected as well. The Court recalled that the interpretation of the two regulations should be compatible and none can prevail over the other.
The Court held that the REA should be mindful of the confidentiality of documents produced by consortium members which have been qualified as confidential and held by REA as part of an action of Horizon 2020 when receiving an access to information request about such documents. If such documents are protected from disclosure under Article 3 of Regulation 1290/2013, it did not imply a straightforward application of the exception under Article 4 of Regulation 1049/2001 for REA to withhold that information. REA should examine if despite the confidential nature of the documents the exemption could be applied fully or partially and whether there is some information that could be nonetheless released without any harm to the protected interests of the consortium members. The Court reminded EU bodies that when receiving an access to information request under Regulation 1049/2001, they should strike a balance between the public’s right to know and the interests of the parties at stake such as the interests of the consortium members of the iBorderCtrl project in the present case.
The Court concluded its preliminary findings by stating that REA correctly examined if any information included in the documents concerned was protected by professional secrecy under Article 399 of the Treaty on the Functioning of the EU (TFEU) and therefore, they could not be released.
The lack of harm to the commercial interests of the consortium members
The applicant argued that the documents he had requested to access could be released, fully or partially, without any harm to the commercial interests of the companies concerned.
REA argued that the documents concerned included confidential information of the consortium members about intellectual property, ongoing research, know-how, methodology, techniques and strategies whose release could have harmed their commercial interests. These considerations were compatible both with Regulation 1049/2001 on access to EU documents and Regulation 1290/2013 which was protecting the confidentiality of information related to the iBorderCtrl project.
The Court first considered that Article 15.3 TFEU enshrined that “any citizen of the Union, and any natural or legal person residing or having its registered office in a Member State, shall have a right of access to documents of the Union institutions, bodies, offices and agencies, whatever their medium.” The same right is also protected by Article 42 of the Charter of Fundamental Rights of the EU. Regulation 1049/2011 gives effect to such right by providing access to information held by EU institutions. The principle of maximum disclosure is foundational but can be limited in order to protect either public or private interests. These interests are listed in the Regulation in Article 4 which prevented the EU institutions from releasing such information. Since these are exceptions, they must be interpreted strictly. This means that the decision of withholding the requested information did not have to fall under the exception of Article 4 only but the EU body must justify why the release of the information would concretely and effectively harm the interest protected by Article 4.
As for the commercial interests, the Court had previously held that all the information related to a company and its commercial relations should not be considered commercial per se and do not fall under the exception of Article 4 of Regulation 1049/2001. The EU institution should prove that in the documents the requester was asking access to there is information, such as commercial strategies, relationships, methodologies or any data related to their expertise that, if released, would harm the commercial interests. Whether partial access can be granted should be examined in light of the principle of proportionality. This meant that information that would harm the commercial interests could not be released but if there are parts of the document whose release would not cause any harm, they should be partially released.
In the case at hand, the contested decision stated that the information related to the inside knowledge of the consortium and its intellectual property. The release of such information would harm the competitive position of the consortium in the market and hence could undermine their commercial interests including the intellectual property which would have created an advantage for their competitors. This advantage would have consisted in the ability to anticipate the consortium members’ strategy and their weaknesses when participating in future tenders. Secondly, they could also copy and use their intellectual property to develop products and apply for patents. Thirdly, the release of the information would harm the chance for the consortium members to get funds from investors in a very competitive market. Fourthly, the commercial information in the documents could potentially damage the reputation of the consortium members. For all these reasons, REA decided that full access to the documents could not be granted but only a partial one.
The applicant argued that REA invoked the interest of the “consortium” that is not a single entity but is composed of different members including universities and scientific institutes that should not have commercial interests. The Court rejected this argument. It noted that the results of the project developed by the consortium members contain commercial information that, if disclosed, could be used by their competitors as it is either copyrighted or commercial, like patents.
After analysing why each of the documents concerned included commercial information whose disclosure would have harmed the commercial interests of the consortium members of the iBorderCtrl project, the Court concluded that REA had correctly applied the harm test under Article 4 of Regulation 1049/2001.
The existence of a public interest overriding the harm to the commercial interests of the consortium members
The applicant argued that there were multiple public interests that overrode the commercial interests of the consortium members and therefore, REA should have released the documents fully. Firstly, the project was funded by EU funds and the public should always be able to access its results. Secondly, there was a public interest in the dissemination of such results in order to have a public debate and provide accountability for those results. Thirdly, the project concerned the application of biometric technologies that are notably controversial for their ethical impact on human rights. Fourthly, media has shown a lot of interest in the use of biometrics for border control and they have produced many reports on the topic. Lastly, there is a democratic and political interest to disseminate such results from the research phase where its incompatibility with the protection of human rights could be highlighted from the early stages without wasting public funds on the development of a technology that would be considered unlawful only when it’s ready to be used. The applicant said that all these considerations would make the commercial interests of the consortium members sound “irrelevant” or in any case have less weight.
The Court held that it is the requester’s burden to argue why there should be an overriding public interest by laying down all the concrete circumstances that would justify the release of the information concerned. The Court did not reject the applicant’s argument about the existence of a public interest but noted that the EU legislator introduced specific provisions to limit the dissemination of the results from a project funded in the framework of Horizon 2020. Hence, Regulation 1290/2013 and the grant agreement struck a balance between the public interest, the interest of the scientific community and the media to impart that information, and the interest of the consortium members to have their intellectual property rights protected. In this regard, this Regulation contained specific provisions bringing obligations to disseminate project results and REA could use information that was not specifically protected for media and dissemination activities. The Court held that the public interest was already correctly considered in the provisions of the Regulation.
As for the alleged unlawful impact of the technology on the protection of human rights, the Court reminded that all projects funded by the Horizon 2020 programme had to respect human rights and be compatible with the Charter of Fundamental Rights of the EU. The Court added the project was a research project whose aim was to test new technologies and not to apply them. The public interest existed only in one of the following steps where this technology will be applied and implemented “on the ground”.
The Court agreed with the applicant’s argument that the public had a right to know and to participate in an informed and democratic debate on the deployment of biometric technologies in the border control particularly if they are public funded. However, the same could not be said for the research phase which is the early stage of the development of such technologies.
The Court further noted that the public interest to transparency does not have the same weight when access to documents related to the administrative activity of the body concerned is requested. The applicant failed to show that the principle of transparency was so relevant in the present case that it overcame the commercial interests of the consortium members and therefore, the court held that the REA had correctly considered that there was no overriding public interest and the information had to be fully released.
The Court concluded that the applicant did not demonstrate the existence of the public interest outweighing the harm to the commercial interests of the consortium members and therefore, the requested documents could not be released under Article 4 of Regulation 1049/2001 on access to documents of the EU institutions.
Decision Direction indicates whether the decision expands or contracts expression based on an analysis of the case.
The General Court of the EU established that a number of access to information requests denied to Patrick Breyer by the REA were not sufficiently justified by the REA. While, on the one hand, the Court recognised that there was a public interest in the democratic oversight of the development of surveillance and control technologies such as iBorderCtrl, on the other hand, it suggested that such democratic oversight should begin only after the research was concluded. The Court has failed to acknowledge the importance of ensuring that transparency is in place at the outset of any such projects that are publicly funded. The Court has arguably ruled that the public interest exists only once the innovations and research had been completed.
Global Perspective demonstrates how the court’s decision was influenced by standards from one or many regions.
Case significance refers to how influential the case is and how its significance changes over time.
Let us know if you notice errors or if the case analysis needs revision.