Have you heard a vendor tell you about their old product, which now protects the Internet of Things? No, it isn’t a pull-up bar, it’s an Iron Bar Crossfit (TM) Dominator!
You should be mentally prepared for the Official RSA Conference IoT Onslaught (TM). But when a vendor asks how you are protecting IoT, there’s really only one appropriate response:
“I do not think that means what you think it means.”
Not that there aren’t risks for Internet-connected devices. But we warned you this would hit the hype bandwagon, way back in 2013’s Securosis Guide to RSAC:
We are only at the earliest edge of the Internet of Things, a term applied to all the myriad of devices that infuse our lives with oft-unnoticed Internet connectivity. This wonʼt be a big deal this year, nor for a few years, but from a security standpoint we are talking about a collection of wireless, Internet-enabled devices that employees wonʼt even think about bringing everywhere. Most of these wonʼt have any material security concerns for enterprise IT. Seriously, who cares if someone can sniff out how many steps your employees take in a day (maybe your insurance underwriter). But some of these things, especially the ones with web servers or access to data, are likely to become a much bigger problem.
We’ve reached the point where IoT is the most under- or mis-defined term in common usage – among not just the media, but also IT people and random members of the public. Just as “cloud” spent a few years as “the Internet”, IoT will spend a few years as “anything you connect to the Internet”.
If we dig into the definitional deformation you will see on the show floor, IoT seems to be falling into two distinct classes of product: (a) commercial/industrial things that used to be part of the Industrial Control world like PLCs, HVAC controls, access management systems, building controls, occupancy sensors, etc.; and (b) products for the consumer market – either from established players (D-Link, Belkin, etc.) or complete unknowns who got their start on Kickstarter or Indiegogo.
There are real issues here, especially in areas like process control systems that predate “IoT” by about 50 years, but little evidence that most of these products are actually ready to address the issues, except for the ones which have long targeted those segments. As for the consumer side, like fitness bands? Security is risk management, and that is so low on their priority list that it is about as valuable as a detoxifying foot pad. We aren’t dismissing all consumer product risks, but worry about your web apps before your light bulbs.
At RSAC this year we will see ‘IoT-washing’ in the same way that we have seen ‘cloud-washing’ over the last few years – lots of mature technology being rebranded as IoT. What we won’t see is any meaningful response to consumer IoT infiltration in the business. This lack of meaningful response nicely illustrates the other kinds of change we still need in the field: security people who can think about and understand IPv6, LoPAN, BLE, non-standard ISM radios, and proprietary protocols. Sci-Fi writers have told us what IoT is going to look like – everything connected, all the time – so now we’d better get the learning done so we can be ready for the change that is already underway, and make meaningful risk decisions, not based on fear-mongering.