The Californian media giant is not having a good run European Courts. In between Mosley’s two victories Vidal-Hall and others also succeeded in resisting Google’s jurisdiction challenge to the progress of their privacy/DPA claim arising from the search engine’s use of their personal search data. Time for a rethink of their litigation strategy you would have thought.
Whilst the Hamburg District Court did not accept all of Mosley’s case (the least intrusive of the thumbnail images identified in the claim were not considered to engage privacy rights, and other images were not proven to have been published on google.de) the decision emphatically supports the argument he is pursuing across jurisdictions that Google should not provide access to images that are patently unlawful. Translated extracts of the judgment are below:
Regarding the nature of the images the panel of judges made the following comments:
- “These images do not have a reporting value to be taken into consideration, and a legitimate interest in information to the public cannot be seen”
- “The images violate the specially protected privacy of the plaintiff… to a very considerable extent, as they provide information about specific sexual acts. In that regard, they are likely to lead to the stigmatisation of the plaintiff”
- “It should be noted that a particularly high degree of intervention regularly emanates from the pictorial representation of the sexual behaviour of a person, since the viewer actually has the scene “before his eyes””.
Google’s liability as an “accomplice”:
- “The defendant is liable for cessation of the dissemination of the likenesses lit. c) – h). It can be the subject of a claim as an accomplice because it has not taken the possible and reasonable steps in accordance with the indications of the plaintiff to prevent further breaches of rights.”
- “An accomplice is a person who – without being a perpetrator or participant – in any way willfully and causally contributes to the infringement of the legal interest protected (BGH judgement of 15.08, 2013, I ZR 80/12, Juris para. 30 with further references.). Even if the defendant stores its “own” thumbnails on its servers in order in the context of the thumbnail view to efficiently display search results to the user, it only releases the original images that have been provided by a third party to requests from users. In essence, due to the technical process of crawling, indexing and displaying search results this involves the reproduction of third party content. Indeed, the typical user of the search engine recognises this because he knows that the defendant offers through the image search service only links between a search query input and the “content of the Internet”. It can therefore only be subject to claim as an accomplice. However, it contributes with the search function offered by willingly and causally to the distribution of likenesses which – as shown – seriously breach the general personality right of the plaintiff.”
- “The infringement in the form of interference in the intimate sphere of the plaintiff was recognisable for the defendant without further investigative steps. It required as described – if at all – no extensive, difficult trade-off between the general personality right of the plaintiff under Art. 1 para. 1, 2 para. 1 GG, Art. 8 ECHR and the freedom of communication and opinion of the defendant under Art. 5 para. 1 GG, Art. 10 of the ECHR.
- “the interests of the plaintiff in the protection of his intimate sphere against such a serious interference outweigh the interests of the defendant, given the absence of a legitimate public interest in the disputed images.”
- “the search engine of the defendant is not just of a merely technical, automatic and passive nature. The defendant not only makes the technical means available, but searches the Internet using software, stores the retrieved image files not least for reasons of efficiency in the context of display on its own servers and “connects” them with requests from users. Thus, it is not limited to the provision of information for access by third parties It creates the link between the contents – in this case images – which have been posted on the Internet and the search terms entered by a user.”
- “The photos at issue here contain a substantial and easily perceptible interference with the personality rights of the plaintiff, because they violate his intimate sphere. The images in lit. c) to h) are unlawful in every possible context. An acceptable distribution of these images – if there is no agreement of the plaintiff – is inconceivable, given the circumstances of this case, and a comprehensive assessment with the respective contextual embedding of the images is just not appropriate. The already shown balancing between the interests of the plaintiff and an information interest of the public can be made in this case without consideration of a further substantive context and results in a clear and easily perceptible predominance of the interests of the person affected.”
The steps Google should take regarding the images:
- “For these reasons, even the offer to “notice and take down” provided by the defendant failed to meet its obligation, because the present proceedings are also oriented to the fact that the relevant affected person must prove to the defendant the allegedly infringing facts in each individual case. In the present case, because of the above, and the recent efforts of the plaintiff this is only an inadequate tool for monitoring and verification of the results, because the duty to monitor and control would provisionally remain with the plaintiff.”
- “Given the gravity of the infringement and of his efforts so far, he is not required to take action against all the major media companies – possibly in the world – distributing these images on their own sites. This is because it must be considered that the image search engine of the defendant regularly searches the entire Internet and can thus always find pages and include them in its index that contain appropriate images, even if they are not publications of known media companies. However, the interference in the protected intimate sphere is no less relevant to the plaintiff in this case, if the image is displayed in the search results with the URL of a lesser known website. It is also necessary to consider the position of the defendant, and the size and reputation of the search engine operated by. The fact that these images can also be distributed on other search engines in the event of blocking by the defendant, and particularly on social networks, does not exempt the defendant from the obligation incumbent on it.”
- “The requirements for a secondary burden of proof are met. The plaintiff has discharged his burden of proof. He has convincingly argued with regard to the secondary burden of proof to the defendant for the use of software for image recognition and the erasure or blocking of content.”
On Google’s technical capacity to take these steps
- “The defendant has failed to substantiate the opportunities and the scope of image recognition software already deployed or on its efforts to develop such a software. The defendant has confined himself to claiming that there is no such software or that the existing software was unsuitable for filter actions and explaining general issues of scalability, latency, reasonableness and overblocking.”
- “The defendant has argued that software programmes like PhotoDNA, itWatch and Content-ID and image recognition software that works with so-called robust hash values, are not able to meet the application of the plaintiff, since they enable a technical review only to a small extent. It has also commented on the expert opinions submitted by the plaintiff to the effect that it assumes a hash value of data identical images and therefore also does not enable effective action. Also the function “search related images” offered by it is not a sufficient instrument of control. In this regard, however, the defendant has only commented on programmes that are already being used and whose configuration according to the defendant is insufficient for the applications of the plaintiff. The defendant, on the other hand, has not designated the measures it has taken to be able to prevent further similar infringements of the plaintiff – maybe not immediately, but with an eye to the future. It does not even argue that it has made efforts to develop appropriate software or to further develop existing software or that it has in this respect reached its limits.”
- “The questions whether the use of filtering software in this particular case would be unreasonable for the defendant, for economic or other aspects, or whether the communication process would be hampered to an unacceptable degree by such a measure, may be left open in these proceedings for lack of a substantiated submission on filter measures. Accordingly, it may also be left open whether a victory for the plaintiff would lead to a “wave” of actions that could no longer be handled by the defendant and could establish the danger of overblocking. Similarly, however, it should also be taken into account that this case involved a certain special case of an intimate sphere infringement by the proliferation of images.”
- “On other possible measures with respect to its responsibility for the obligation to take measures to prevent similar infringements, the defendant has also not offered arguments apart from the instrument “notice and take down”, which is insufficient for the present serious infringement.”
Dominic Crossley is a Partner in the Defamation & Reputation Management team at Collyer Bristow LLP. He acted for Max Mosley in his original claim against News Group Newspapers and advises on privacy and defamation.