CTA #3: Develop a Data Strategy for AI, ML & Automation

2019 Billington CyberSecurity Summit

On the cybersecurity front, the federal government is moving swiftly from a crouch-and-defend posture to a mode of active threat hunting and response, as discussed with 75 top experts at the 10th Annual Billington CyberSecurity Summit.

One key takeaway from the Summit: federal cybersecurity leaders must develop a data strategy for AI, ML, and Automation.

Panel 10th Annual Summit
“Harnessing AI/ML in Cybersecurity” Panel at 10th Annual Billington CyberSecurity Summit

The data required to understand and mitigate threats grows exponentially. Deploying AI will take cooperation among actors in several domains. Brad Medairy, Executive Vice President at Booz Allen Hamilton, states that we must “bridge the gap between government, the developers, Silicon Valley and the community to provide the data [government] needs to build and tune their algorithms.”

Swami Sivasubramanian, Vice President of Amazon Machine Learning
Swami Sivasubramanian, Vice President of Amazon Machine Learning

Swami Sivasubramanian, Vice President of Amazon Machine Learning added, “Get executive buy-in on your machine learning strategy really working well, and get your talent really skilled in machine learning.”

There is a clear case for artificial intelligence and machine learning in cybersecurity, but it leads to the need for a strong data strategy.

Dr. Lynne Parker, the Artificial Intelligence Assistant Director at the White House Office of Science and Technology Policy, emphasized the need to trust the cyber data you’re feeding your analytics and AI programs, while Air Force Lieutenant General Jack Shanahan, Director of the DoD Joint Artificial Intelligence Center, stressed the importance of a strategy for “data access, data quality, data content, data classification, and data formats and standards.”

Due to differences in the ways cyber service providers collect data, this can be challenging. Shanahan notes, “We have to curate our data, train against that data, integrate those models into our systems that were just never built to have AI in them.”

To get more information about these and other timely cybersecurity issues, read our full white paper: 15 Calls to Action in Federal Cybersecurity for 2020. We thank journalist Tom Temin for his work encapsulating our two-day Summit into these plans of action.

15 Calls to Action in Federal Cybersecurity