CNET también está disponible en español.

Ir a español

Don't show this again


MyPrivacyAudit wants to keep your tweets from leaking

A new service checks for cases in which you may have leaked personal information over your Twitter feed -- but it doesn't seem to catch everything.

This is the privacy audit of a Twitter account that had recently posted a photo of a driver's license and part of a credit card.
This is the privacy audit of a Twitter account that had recently posted a photo of a driver's license and part of a credit card.

Hopefully, we all know that we should be careful what we tweet, especially after hearing stories about people who mention being on vacation and come back to find their home burglarized. But slips can happen in the fast-paced Twitterverse.

To see if you may have been careless with your tweets, you can use new service at and find out exactly where you went wrong. Though my quick testing of the service suggests that while it may be better than nothing, its privacy checks are far from infallible.

From the Web site:

Initially, the focus of MyPrivacyAudit is understanding the privacy concerns associated with using Twitter. This tool can identify personal information that is connected to a tweet but not necessarily contained in the tweet itself. With this information, account holders will be better able to manage the overall security of their Twitter account.

I entered my Twitter ID and it analyzed all my tweets and concluded that my feed was not risky except for the fact that I don't use Protected Tweets, which allow only approved people to see the tweets. But that would defeat the purpose of using Twitter to reach a large audience, and MyPrivacyAudit acknowledges that.

The service also assesses how much the tweets disclose related to location, such as if geolocation is turned off and if location is exposed when using check-in services or posting photos. And it looks for specific types of sensitive information, such as credit or debit card pictures, travel information, vacation or home location. To see which tweets are raising red flags with the service, you need to create an account and log in.

I then posted a test tweet referring to my home address being "1234 Main Street, Fargo, North Dakota." But when I went back to MyPrivacyAudit, the results didn't change. Maybe it could tell I was lying.

I then tested the service on some random accounts whose tweets with photos of credit cards had been retweeted by the @NeedADebitCard account. The service didn't flag a user for "credit card photo" leak even though she had posted a photograph of her card cut into pieces. But two other users did get flagged -- one who had posted a blurry photo and another who posted a close-up photo of her drivers license partially covering her credit card.

The Web site mentions that the service will be available for other social media sites in the future.