Researchers from Duke University, Penn State University and Intel Labs developed a piece of software called TaintDroid, which has been able to detect and report when applications transmit private information to remote servers.
Of the 30 applications tested with TaintDroid, 15 were found to be sending private information to advertising servers without informing users. That information included a phone’s SIM card number, GPS location and phone number.
And these weren’t just any Android applications. Researchers used the thirty most popular applications across 22 categories which required Internet permission, along with permission to access location, camera, or audio data.
Users put their faith in application developers and the fact is that, “Android’s coarse grained access control provides insufficient protection against third-party applications seeking to collect sensitive data,” wrote the researchers involved in this study.
A Google representative has responded to the report, saying, “In all computing devices, desktop or mobile, users necessarily entrust at least some of their information to the developer of the application. Android has taken steps to inform users of this trust relationship and to limit the amount of trust a user must grant to any given application developer. We also provide developers with best practices about how to handle user data.”
This doesn’t leave users with many options. Sure, you can try to put faith in application developers, but the fact is that until stricter security standards are put in place, many Android applications may continue sharing your data.

I think you mean "a sample size of 30 would give you a MARGIN OF ERROR of 18% at a confidence level of 95%," etc.
In fact, I'm certain.
every smartphone does this. including apples iphone
'A Google representative has responded to the report, saying, “In all computing devices, desktop or mobile, users necessarily entrust at least some of their information to the developer of the application."'
Pure BS! There is NO reason for an app like, say, Word to be communicating any of my data to anyone.
The problem lies more in the advertiser's lap. They require this information in order to supply your app with banners. Most of it is localisation data for geo targeted adverts. If you want free apps this is something that you have to expect as developers appreciate some income for their hard work.
I expect the situation is also the same on the iphone.
This is the price we pay for an open marketplace. I'm not condoning this action, just saying that consumers are responsible considering the central authority for apps is much more accepting of programs. Unfortunately, this is the downside of open source, people who abuse it.
While this might be true, there's a lot of vagueness. They don't tell you which apps and which categories these apps come from. Since it's 30 apps from 22 categories, that's about 1 per category, with some having 2. It doesn't mean is has to be the number one most popular app in that category. For all we know, they specifically targeted apps which may do this, and only published the results of this, making it seem like there's a much higher percent of apps doing it than in reality.
Regardless, when you go to the market and install these apps, they all ask for permission to do certain things. Most people don't read it, and just click okay. If you don't want your information out there, then don't use the apps, or at least, use a program that will protect it.
A lot of the popular Apps are free as well, the developer has to be making money on it somehow, or the app wouldn't be getting updated.
Man this is like the CFA lesson I took this week Hypothesis Testing lolz…. this made my "fundas" all clear
Just buy the iphone
Really Lou? Because Apple won't collect your info?
The problem is, these developers also write apps for the iPhone and share mostly the same info.
Recently, Apple came under fire for semi-secretly collecting and sharing GPS-information for location based advertising.
Did they fix it? No, the changed the EULA a little, which by the way already mentioned the issue for a few years.
(who reads them anyway?)
Let's face it, our privacy is not compatible with technology.
If you don't like it, don't use a computer or smartphone.
u forgot the sarcasm tags
<sarcasm> just buy the iphone </sarcasm>
fixed we can go about our day now.
I don't understand why they don't give the names of the applications. I'd like to see what I've been using in the top categories that are sending me information. This article seems to just be a paranoia driver. In other words, some apps tested bad with information, therefore you shouldn't trust any of them. How about giving us some insight as to how we can track and protect ourselves from these privacy invasions?
If they gave the names of the apps, then someone could prove that this isn't a good representation, or possibly an outright lie for attention.
This "news article" brought to you by the crApple marketing department and their bribed cronies in the media.
Read the article, "Researchers used the thirty most popular applications across 22 categories which required Internet permission, along with permission to access location, camera, or audio data." The most popular applications were doing this, no reason to believe that less popular apps are behaving any diffferently.
One day you might crave the refuge of that that walled garden.
Checking 30 apps out of the thousands once could install is hardly a reasonable spread. Test a couple hundred and see where the ratio is then.
Not that I doubt that the amount of apps sending personal data would be significant, but with such a small number of apps tested, this article is misleading and sensationalist at best.
With a population of approximately 100,000 apps, a sample size of 30 would give you a confidence interval of 18% at a confidence level of 95% (the typical confidence level used in polls). In other words, we can be 95% confident that the true ratio is between 32-68%. As far as I'm concerned, any result in that range is alarming and hardly sensationalist. However, their selection method is flawed because it isn't random. There could easily be a hidden factor affecting the ratio for popular apps. For example, developers could be under greater pressure to sell private data for popular apps because their data is more valuable. However, it's hardly a result to dismiss.