The Bug Bounty Model: 21 Years & CountingA look back on the beginnings of crowdsourced vulnerability assessment and how its robust history is paving the way for the future.
When Netscape launched the first bug bounty program 21 years ago, it redefined the way companies approach system vulnerabilities. Today, there is widespread adoption of crowdsourced security programs across mainstream companies with more than 600 publicly disclosed programs and counting.
I’ve worked on a number of these bug bounty programs over the years, and served as director of penetration testing for HP Fortify. The changes have happened so fast, it’s easy to lose sight of how far we’ve come since the very first program was introduced in 1995. As we approach the new year, let’s take a look at the robust history that set the foundation for the modern bug bounty program.
The First Bug Bounty
Netscape Technical Support Engineer Jarrett Ridlonghafer designed and launched the first bug bounty program to discover vulnerabilities in Netscape’s beta version Navigator 2.0 Internet Browser. The company offered cash rewards to hackers who found bugs in the software.
Although this was a major advancement for the security industry, the model wouldn’t catch on for another seven years. By 2002, IDefense launched its own bug bounty program and in 2004, Mozilla created a program that is still running today. These early programs paved the way for the modern bug bounty and for the emergence of managed programs and bug bounties as a service.
Breaking the Mold
In 2010 and 2011, Google and Facebook took notice of crowdsourced security, adding them to their business models, which increased their popularity and incentivized more researchers to join the bug bounty community. In March 2011, Facebook paid a 22-year-old security researcher $15,000 for a bug discovered. By 2015, Facebook had paid more than $4.3 million to researchers globally.
Bug bounty programs were beginning to increase in popularity, yet many organizations still perceived them to be too risky. This perception was tied to the belief that a bug bounty gives hackers free reign of critical code. But the reality is much more controlled than that, because, whether you invite hackers in or not, as long as applications are connected to the Web, they’re vulnerable. Tapping into the intelligence of thousands of security researchers helps identify these vulnerabilities before the bad guys do and lowers the risk of being vulnerable.
Bug Bounties as a Service
In recent years, the growing need for bug bounty programs and the challenges and costs associated with managing them internally drove the creation of third-party platforms or bug bounties as a service. This opened new pathways for a growing hacker community and furthered adoption by other market sectors such as healthcare, financial services, automotive, and the Internet of Things.
For companies, third-party platforms offer the opportunity to create personalized programs by connecting organizations with trusted partners and a community of diverse security researchers. For researchers, the third-party platform verifies their results, handles arbitration issues with the company, and makes it easier for individuals to get paid and move onto testing for more bugs. Third-party platforms also drive the creation of a thriving community where researchers connect, educate, and inspire one another in an environment that allows people with a variety of backgrounds to share their knowledge and expertise.
Crowdsourced vulnerability assessment has evolved to include more than just public programs. As I mentioned earlier, a common misconception about the bug bounty model is that all programs are public. In reality, the majority of all programs launched are invite-only. Private, ongoing, and on-demand programs are incredibly common and give companies a way to facilitate testing on harder-to-access applications, or focus testing on a small subset of an attack surface to meet organizational testing needs.
Private programs allow organizations of all sizes (like Western Union, Okta, and Aruba Networks) to validate the security work they’re doing internally, and leverage a curated crowd of talent to scale up their team and improve response time before going public.
Crowdsourced security programs have taken on many different forms and will continue to play a major role in securing applications, especially as companies face increased pressure to release updates and keep their customers’ data secure. From the increase of vulnerabilities in healthcare devices, IoT and the automotive industry, these programs can bring advancements to industries across the board. With the willingness and constant interest from intelligent engineers, bug bounty programs will continue to thrive.
Jason is the head of trust and security at Bugcrowd. Jason works with clients and security researchers to create high value, sustainable, and impactful bug bounty programs. He also works with Bugcrowd to improve the security industry's relations with researchers. Jason's ... View Full Bio