Another key aspect of the security and privacy label project is that the information is also encoded to be machine readable. This way, even if different countries or industries develop their own assessment tools, there's still a way to compare and process all the data. The researchers point out that data from the labels could make it easier to search for products by their privacy and security features, creating the potential for these to be mainstream product considerations rather than niche points that are difficult for consumers to research. Ecommerce websites could even offer filters for privacy and security features like they already do for things like price, weight, or screen size. In this way, consumers could make intentional choices about the products they buy, with digital safety as one of the factors.
The researchers say that they've had a lot of private-sector and congressional interest in their label. But so far they've only been able to make example labels based on imaginary products or mock up labels for real products based on public data. The researchers are looking for a manufacturer to pilot the labels in a more serious way, with honest information about the products.
There is real momentum toward doing these types of tests. Finland, Singapore, and the United Kingdom are all working on national IoT label programs focused on security. And while some IoT security bills have floated around the US Congress, the National Telecommunications and Information Administration within the Department of Commerce is actively working on a similar type of project for software. The idea is to develop a software "bill of materials" that would help the industry keep track of all the different open source and third-party components that go into one single software program or platform.
"Standardization I think will help, just like the ingredients label on food educates people about how much sugar or sodium they're consuming," says Chris Wysopal, chief technology officer of the software auditing firm Veracode. "Standardizing a software bill of materials would make it more clear to a consumer what they’re getting."
The researchers are realistic that for their work to have a long-term impact there would either need to be widespread voluntary adoption of the label by manufacturers or a government mandate to do so. But they say that's why they've designed the label with room for manufacturers to explain their choices to consumers.
"There may be a really good reason that your thermostat has a microphone, but if the company doesn’t tell you, then you’re shocked," says Lorrie Cranor, director of Carnegie Mellon's usable privacy and security lab. "If they tell you about the microphone up front and explain why that is, then you might say 'Oh, OK, that makes sense.'"
Conventional wisdom says that consumers won't typically pay a premium for privacy and security features. The researchers had preliminary findings, though, that an easy-to-read label might help people better understand potential risks and make them more willing to pay more for strong guarantees. It will take more investigation to expand on that finding, and the easiest way to do extensive testing would be for companies to start adopting security and privacy labels on their IoT products. You likely won't be seeing IoT privacy labels on store shelves anytime soon. But the stakes are high enough that something certainly needs to change.