Attorney – Client in the cloud

Over the past few months I’ve been asked several times about the status of attorney-client privilege when attorneys use cloud technology.  It is an interesting question and there are a couple concepts that need to be explained about A-C.  So buckle up, this is a long one.

First (and very broadly speaking), A-C is lost when disclosed to a third party intentionally or inadvertently.  So an attorney and client discussing a case in a busy coffee shop could potentially lose A-C since a third party could overhear the communications.  I know the attorneys reading this post will likely come out of their chairs with exceptions, but I’m trying to paint a high level picture.

This loss of A-C does not mean that attorneys have to hide everything in locked safes buried in concrete.  The comments on the model rules of professional conduct state:

“…unauthorized access to, or the inadvertent or unauthorized disclosure of, information relating to the representation of a client does not constitute a violation of paragraph (c) if the lawyer has made reasonable efforts to prevent the access or disclosure. ” (emphasis added).

So, this brings us to Harleysville Insurance Company v. Holding Funeral Home, No. 1:15cv00057, memorandum op. (WD Va. Feb. 9, 2017).  In Harleysville, an investigator for the parent company of Harleysville uploaded surveillance video of the underlying event to a file sharing site.  He then emailed a link to the video to another party.  The same investigator later placed the case file in the share.

The share was not password protected or otherwise protected.  In fact anyone with the link or anyone who found the share could see the information.  Remember the language of the model rule above?  The Virginia court echoed this language in their opinion stating that inadvertent disclosure can be caused “by failing to implement sufficient precautions to maintain its confidentiality.” (emphasis added)  The court continued “With regard to the reasonableness of the precautions taken to prevent the disclosure, the court has no evidence before it that any precautions were taken to prevent this disclosure.” (emphasis added).

The court concluded that A-C had been waived by posting the information to a publically available website.

As I’ve counseled clients in the past, whether A-C will survive in the world of cloud usage depends on the steps taken to prevent disclosure.  Encryption, access control, and logging are your friends.

Breach notification laws: Better privacy or the 10th circle?

Today (April 19th) New Mexico became the 48th state to enact a data breach notification law.  Only Alabama and South Dakota do not have a notification law on the books.

On the one hand this is good news for the privacy on New Mexicans.  They are now ensured they will have notice of a breach of their personally identifying information.  They will have the opportunity to mitigate the damage resulting from such a personal exposure.

For security and privacy folks, though, there is a different perspective.  We now have 48 distinct regulations to track.  If I have a client that does business across the country, I have to ensure I am able to help them comply with 48 different (and sometimes contradictory) regulations.  As an example, assume I have a client doing business in Texas, Oklahoma, and Colorado:

  • In Texas and Oklahoma consider a drivers license an element of personally identifying information; Colorado does not.
  • Colorado requires notification to the credit reporting agencies (CRA) if more than 1000 records are breached.  CRA reporting is required in Texas if more than 10,000 records are breached.  Oklahoma does not require CRA reporting at all.
  • Oklahoma allows for electronic communications if the cost of written communication exceeds $50,000; in Texas and Colorado it is only allowable if the costs exceed $250,000.

Add the other 45 states to the mix and the mapping becomes complex.  I won’t comment on whether there is a “better” rule, but the hodgepodge of requirements makes it more difficult for everyone.   This is the type of conflict that is ripe for a federal rule to unify requirements.  Unfortunately the attempts to do so over the past few years have failed to garner much attention.

BTW:  For the curious, there are at least 89 different counties with breach or privacy laws.  A breach at a multi-national corporation can be very complex.

Data classification: Attempting to solve for x without knowing a, b, and c.

Computerweekly.com has an interesting post on federal government security classifications and cloud provisioning.  The TLDR is that many federal agencies are paying too much, because they are classifying information incorrectly and vendors are happy to upsell protections.

In my experience in state government, the problem is very different.  To begin with few agencies have strong data classification policies.  In Texas, the Department of Information Resources published a data classification template that agencies can use to develop a classification scheme.  Personally, I think the template (and associate white paper) is a marvelous piece of work :).  It is unclear how many actually used the template, though.

The problem isn’t limited to Texas.  Based on discussion with the CISOs from other states, data classification is a difficult problem for many.

The issue raised is how to determine the appropriate protections for data when classification programs don’t exist.  The result is agencies will either over protect public data or under protect sensitive data.  Several states have a de facto policy of requiring all data to be hosted in the continental United States (conus).  While this is appealing, it also drives up prices for cloud services.

Many of the regulations that affect states (most recently CJIS) have dropped the conus requirements.  Requiring conus storage for “public” data is probably not the best use of taxpayer money.  Without a strong data classification program, though, it is hard to make informed decisions.