Investigating the State of Vulnerability Disclosure in Consumer IoT Products

 

In August 2018, we were asked by the IoT Security Foundation to look at companies across the world producing consumer focused Internet of Things products and see what the situation is for security researchers when they try to contact these businesses.

 

Security researchers often have problems when it comes to speaking to companies about their findings, but we wanted to gather some real data about the current market situation because no-one had done this before. In this process, we also tried to record what types of mechanism were in place – i.e. did the company follow best practice for vulnerability disclosure by having a webpage that researchers could report through? Was there an email address to contact the company and was there public key available to use to encrypt submitted reports? Did the company operate any kind of ‘bug bounty’ scheme?

IoT devices in the IoT Security Village at DEF CON#26

The IoT Security Foundation published our findings (pdf) today, including a full list of the companies we looked at. The data is also available on request from the Foundation in a machine-readable format (with some additional fields we didn’t include in the report).

 

Some high-level findings from the report include the following:

  • over 90% of consumer IoT product companies out of 331 companies researched, have no way for a security researcher to be able to contact them easily to report a vulnerability.
  • Of those companies which had a disclosure policy:
    • 41.9% with disclosure policies gave no indication of the expected disclosure timeline.
    • 0.9% of the companies operated with a hard deadline of 90 days for fixes to reported issues.
    • 46.9% of policies also had a bug bounty programme. Two of these programmes were however by invitation only, so were not open for general contribution.
    • 78.1% of companies with policies supplied researchers with a public key for encryption to protect their communications and report details.
    • 18.8% of companies with policies utilised a proxy disclosure service (1.8% of total companies examined).
  • 7.6% of the overall companies publicised a public PGP key for researchers to use to encrypt, protecting their communications and disclosure report details.
  • 0.9% of companies had forms for reporting vulnerabilities or contact points, but no published vulnerability disclosure policy.

 

Our CEO, David Rogers said: “The data doesn’t lie – connected product companies are woefully bad, when it comes to allowing security researchers to report issues to them. It is further evidence of the poor situation for product security in the Internet of Things. There is no need for this, there are recommendations and an international standard available for companies to adopt. There needs to be a shift of mind-set to take security seriously at the Boardroom level of connected product companies and for them to realise that regulators are starting to take action against the existing lax attitude towards product security.”

 

John Moor, the MD of the IoT Security Foundation said: “We conducted this research to better understand the contemporary status of vulnerability disclosure policy in practice,” says John Moor, Managing Director, IoTSF. “It’s part of our mission to raise awareness and help improve the situation and we hope that by highlighting this subject area, and identifying companies in the report, we can make positive progress in the future. For any company making connected products, it is fundamental to understand the importance of disclosure policy and leverage the research community to help make safer connected products.”

 

It is clear that things need to change and fast. Guidance on how to implement Coordinated Vulnerability Disclosure is available from the IoT Security Foundation (pdf).