Since I just got back from an interview on the subject of the tort liability law, I thought I should run through this Baidu case (h/t @davesgonechina), which is sort of the first one decided under the new law, even though technically it just became effective today.1 It seems like Pacific Epoch had the English-language scoop on this story, so let’s start with what they had:
Shanghai Jing’an District People’s Court has ruled against Baidu (Nasdaq:BIDU) Thursday in the defamation case of a young woman who charged the company with offering private photos and information in its search results and other products, China Youth Daily reported July 1. The court ordered the domestic search giant to pay a RMB 22,000 remuneration and display an open apology on its homepage for three consecutive days, according to the report.
OK, first off, the damages are not a big deal to a company like Baidu, nor I think is the apology, considering that search engines are engaged in litigation on an ongoing basis.
The important issue here is liability. What exactly went on in this case, why was a search engine found liable, and is this result likely to be replicated by other courts?
In reading the news reports (here is a link to one of the longer articles – Chinese only), it is not clear to me whether the information posted online, and indexed by Baidu, was considered to be a violation of the right to privacy or reputation. Seeing as the former was only formally recognized under the new Tort Liability Law, and that the information posted online included some indecent photographs, I’m thinking that we’re dealing with a reputation right in this case.2
Regardless, the court found that Baidu’s indexing (i.e. providing search results) of the pages that included this information was a violation. Some of the news out there includes quite a lot of information about the type of content that was posted about this woman. From a legal perspective, I don’t really care about the content, I care about the issue of liability for third-party content. For the purposes of this discussion, let’s stipulate that the online information was definitely a violation of this woman’s reputation right.
The question is whether Baidu should be liable for indexing these pages, and for allowing this information to be posted on Baidu Baike (sort of a Wikipedia type platform). In the case of Baike, the defamatory information was being hosted on Baidu’s servers.
For several years now, Chinese courts have trended in the direction of allowing Net platforms, and search engines, a certain amount of flexibility when it comes to taking down information that is an infringement of individual rights or intellectual property rights. Understanding that some of these sites are deluged with traffic, and whose systems are to a major extent automatic (i.e. information is gathered by bots and not people), it is unreasonable to hold these firms responsible for all of their data, all of the time. We’re already talking about devoting a large number of staff to deal with these kinds of issues, and there has to be a limit.
The legal construct that China has borrowed from U.S. copyright law is the “safe harbor” provision. As long as a site maintains an adequate “notice and takedown” system, whereby IP owners (for example) can notify the site that infringing material has been uploaded, then as long as the site owner takes the material down in a reasonable amount of time, they will not be liable.
I consider this to be a fairly settled concept under China IP law. It has not, however, been entirely clear that this provision will be applied across the board to all types of online torts, such as defamation. I cannot say, therefore, that this is an obvious poor decision by the Jing’An court.
On the other hand, if the facts show that Baidu took reasonable measures to take down the defamatory material given the notice it received, then this judgment looks suspect and may be reversed by the Intermediate Court on appeal.
Did Baidu receive adequate notice? Again, the facts in the news are quite incomplete. There is mention of some kind of publication by the plaintiff in a local newspaper of an “Attorney statement” signed by “YH ??”. Well, this is hardly adequate notice in my opinion, and certainly not something that Baidu should have necessarily known about.
Another issue involves Baidu’s responsibility with respect to links to third party content. Apparently in this case, links were removed at some point, but new material was constantly being updated and then automatically indexed as search results. Again, I wish I had more information, but if the above is accurate, then again we have a problem.
Just what exactly is a search engine supposed to do if such information keeps being re-posted? Apparently the judge performed an in-court search on Baidu during the case and found links to some of this content that were active; yeah, that looks bad, but it also kind of smacks of a stunt that could have ultimately been unfairly prejudicial to Baidu, particularly if the judge had inadequate knowledge of the underlying technology. If the links were to fresh pages/posts, then I’m still sympathetic to the search engine.
Finally, the judge apparently thought that a lot of this could have been fixed via keyword filtering, since a lot of the photos and other material were posted using the terms “???” and the plaintiff’s name. Again, this seems quite problematic from an operational standpoint. Does the court really think it’s reasonable that a search engine should institute keyword blocking every time something like this happens? That could add up to a lot of blocked words very quickly! Moreover, depending on the keyword, a lot of non-infringing material might be blocked at the same time.
To close this out, I want to first reiterate that I have limited knowledge of the facts here. However, based on what I do know, I think Baidu should go all out on appeal. While the material in question certainly seems defamatory, and the law does say that operators like Baidu may be held liable for hosting and linking to such material, there are limits. This is a great case for Baidu to argue that the same limits that have been instituted in IP cases (i.e. safe harbor for sites that have adequate notice and takedown systems) should be followed in cases of infringement to the rights of privacy and reputation. Same issues of reasonability, so we should expect the same legal results.
- Specifically, the law did not apply, but the judge did reference the government’s legislative intent in the judgment.[↩]
- The specific differences between the right of reputation and the new right of privacy is one area of the new law that will require clarification.[↩]