A Recap of the “Cybersecurity at the Edge” Breakfast Sponsored by AWS and GDIT, 2/29

A recent Billington Cybersecurity breakfast event on February 29 showcased two expert panels with government speakers from the NSA, DIA and DHS, a former high ranking CIA official and industry experts from AWS and GDIT. They discussed why people should care about what happens at the edge of computer networks and ways in which this space is changing as real 5G continues to roll out and as the public sector and industry both expand their reach into and use of space.

Edge computing allows devices in geographically dispersed locations to process data at the “edge” of a network, either by the device itself, or using the assistance of a local server. Edge computers could include Internet of Things (IoT) devices—such as an Internet connected refrigerator or a city camera system—web connected ICS or SCADA devices that run industrial or critical infrastructure machinery, or devices used by the military to conduct operations on the ground, at sea, in the air, and increasingly in space. The goal of edge computing is to improve the response times of these devices while garnering more insights from the data being captured by the devices.

Connecting more things and building more infrastructure to support it, creates new surface areas for bad actors to leverage to exploit these edge devices and use this new infrastructure to target the connected networks. The panels discussed a number of issues surrounding where both the public and private sectors are in regard to edge computing security. On the one hand, there are a number of concerns with the level of security found in some publicly-available IoT devices currently deployed edge devices and the security that they have baked into them. A $150 home router or an IP connected toaster or light bulb will likely not include advanced security capabilities that will protect it from a capable adversary. In addition, edge computing has also created a marketplace of local content providers in which multiple users depend for trusted workplaces; workplaces that have varying degrees of real security and access policies.

On the other hand, the new generation of edge devices are leveraging new baked-in capabilities—such as new admin-controlled update processes—that will better protect the entire ecosystem as older devices get replaced. Some of the more exciting new capabilities the panels discussed included:

1. The increasing role of artificial intelligence (AI) and how it is being used to improve edge computing security. For example, neural networks are already being used to better monitor and assess how edge devices are being used and this learning process is better enabling automated processes to protect these smarter edge devices on the fly.
2. More edge computing software developers are leveraging memory safe languages—such as RUST—and combining them with formal methods—such as running logic proofs against traditional code—to better understand and fix vulnerabilities in real time.
3. The White House is actively publishing policy and guidance on technologies that impact edge compute such as its strategic 6G principles announcement in February 2024.

The panels also acknowledged the much-improved dialogue between the government and the private sector as it relates to ongoing technical and security developments. Government speakers acknowledged the differences in creating policy from the speed of technology generation and highlighted the continued need for the commercial world to actively participate in moving the policy forward with smarter outcomes. Some great examples of this ongoing collaboration included:

1. NIST’s use of secure messaging channels to engage with the private sector to keep their standards updated such as with their software development framework.
2. Greater transparency in CISA and other government agencies’ policy, regulatory, and guidance processes.
3. Lastly, this Administration’s proactive approach to pushing all of the Federal Government to embrace Zero-Trust and the use of the cloud is driving a fundamental change in cybersecurity overall.

Of course, the panels agreed that there is still much work to be done. A couple of these areas discussed included:

1. The continued focus on a data-centric approach to better understand what needs to be protected and how.
2. Continued efforts to improve the increasing connectivity between operational and business networks especially leveraging AI and formal methods.
3. Continued focus on working with allied governments to share best practices and help protect edge devices downstream.
4. Continuing to find ways to address legacy systems and the way they are protected from adversaries.

Speakers

Neal Ziring, Technical Director, Cybersecurity Directorate, NSA
Stephen Kensinger, Technical Director, Cyber & Security Division, DIA
Iranga Kahangama, Assistant Secretary for Cyber, Infrastructure, Risk and Resilience, DHS
Andrew Boyd, Former Director, Center for Cyber Intelligence, CIA (retired)
Mark Ryland, Director, Amazon Security
Jennifer Krischer, Vice President for Defense Intelligence, GDIT
Matt Hayden, Vice President for Cyber and Emerging Threats, GDIT
Ash Thankey, Director, National Security, AWS