Google's Privilege Claim: A Cautionary Tale

more+
less-

Winnie the Pooh feared the Heffalumps and Woozles that he believed inhabited the Hundred Acre Wood, although he never saw one. There are things that lurk in the woods of e-discovery that lawyers can’t see either…like the Auto-Save. Although unseen, some unfortunate lawyers for corporate giant Google recently felt its bite.

Google and The Auto-Save

In August 2010, Oracle America filed a patent infringement action against Google alleging that Google’s Android smart phone platform infringed certain patents related to Oracle’s Java-based smart phone platform. During the course of discovery, Google produced over 3.7 million electronic documents totaling over 19 million pages to Oracle.

Within this massive amount of produced electronic records were eight draft versions of an email authored by a Google engineer after attending a strategy meeting called by Google’s general counsel regarding Oracle’s claims. The final version of this email included the headings “Attorney Work Product” and “Google Confidential,” and was sent to Google’s senior in-house counsel and a corporate officer and was copied to another Google engineer and the author himself. This final version was withheld from production and appeared on Google’s privilege log. Unfortunately for Google, the eight draft versions did not contain the headings and the recipient names were not yet added. This allowed the drafts to escape detection prior to production.

Where did these drafts come from? The record revealed that during the five minutes it took to draft and send the email, Google’s Gmail email system automatically saved eight “snapshots” of the email and put the copies into the author’s draft email folder. No action was required by the author. It was all done by the Auto-Save.

When Google learned that it had inadvertently produced draft versions of the email to Oracle, it requested that Oracle return all copies. Oracle complied, but filed a motion to compel production of the draft and final versions of the email. Oracle successfully convinced the district court that the email was not protected by any privilege, and the court ordered the production of all versions of the email. Google sought a writ of mandamus to have the district court’s ruling overturned, but the Court of Appeals for the Federal Circuit denied the writ. Even though it was ultimately determined that the Google engineer’s email was a non-privileged business communication, there are several significant lessons to be learned from Google’s encounter with the Auto-Save.

Should You Fear the Auto-Save? Ask IT!

As the Google case suggests, in the world of e-discovery, it has become increasingly important to talk to the IT Department before any data is identified, collected, processed, reviewed and produced. Google’s counsel should have interviewed an IT Department representative to learn what features or systems were in place that could impact the discovery process. A series of questions could have been asked to determine how Google’s Gmail system worked. Does the system auto-save drafts?  What happens to the drafts once the final email is sent? Where are the drafts stored? Answers to these questions would have alerted counsel to the possible presence of draft versions of sensitive and potentially privileged emails.

Moreover, these types of questions should be asked for other office productivity software as well. Word processing, spreadsheet, and presentation software also may auto-save drafts or create backup versions of files during editing. Knowing if these features exist and how they operate can provide valuable information that can be used during the e-discovery process. For example, knowing that auto-saved drafts or backups of a privileged or otherwise confidential document may exist, lawyers can then use targeted searches to ferret out the drafts.

It Says It’s Privileged. That’s Enough, Right?

Wrong. It doesn’t matter what labels are placed on the document. A grocery list doesn’t become privileged because you write “Attorney-Client Communication” on it. As the appellate court made clear in the Google case, the contents and context of the communication are determinative when assessing privilege. The court found that the email at issue was a response to a question from Google’s management that addressed business rather than legal matters regardless of how it was labeled internally. That does not mean to suggest that placing privilege labels on documents isn’t helpful. It certainly can be when it comes to identifying possible privileged communications. But again, the privilege analysis cannot end with the label.

But I Sent it to a Lawyer!

The fact that a lawyer is a recipient doesn’t by itself protect the communication from disclosure. Like a privilege label, the fact that a communication was sent to a lawyer standing alone does not render the communication privileged. If it did, every written or electronic communication would include a lawyer’s name, and the truth seeking process would be severely crippled. The appellate court went beyond the “who” and focused on the “what” to determine that the email concerned business not legal matters.

B. Jay Yelton III - 269.383.5819

Kenneth J. Treece - 269.393.5810