Artificial Intelligence: A Powerful Tool Against Cyberattacks

The full text is below and can be viewed as a PDF here.

Artificial Intelligence: A Powerful Tool Against Cyberattacks

The growth of Artificial Intelligence (AI) has led some to worry that it will lead to human extinction or similar global catastrophe. But if the world is truly to avoid catastrophe and become a better place, AI must play a central role in that transformation, be it through finding more rapid cures for diseases, optimizing energy production and consumption, better and more quickly detecting fires, and many other benefits.

While the doomsayers may feel we are better off returning to a simpler pre-AI world, the world will not stop pursuing this technology. Indeed, the accelerated use of AI can play a pivotal role in stopping the worst of technology today – cyberattacks that can wipe out data or cause systems to operate in dangerous, horrific ways.

The December 17 bipartisan U.S. House of Representatives report on AI directly addressed this matter.

“As sophisticated attackers seek to use AI offensively to exploit complex systems, security teams must use AI defensively to improve cybersecurity resilience. The interest in using AI to improve cybersecurity is underscored by projections that the AI cybersecurity market will reach $60.6 billion by 2028.

“AI already has a growing role in identifying, mitigating, and responding to threat actors and cybersecurity incidents. This is largely due to AI’s ability to rapidly process large datasets, detect subtle patterns, and adapt to new threats, resulting in ‘a powerful level of efficiency and continuous learning that complements human capabilities’ and acts as a force multiplier.”

Indeed, cyberattacks are widespread.

In June 2024, the U.S. Government Accountability Office (GAO) said, “Cyberattacks are on the rise and pose risks to the nation’s critical infrastructure—which relies on computer systems to function. These attacks can compromise sensitive data and disrupt important operations.”

The R Street Cybersecurity-Artificial Intelligence Working Group has identified five ways in which AI is helping to answer the cybersecurity challenges: “threat detection and incident response, vulnerability management and remediation, red-teaming, enhanced security analysis and human workforce efficiency, and data privacy and security.” Each of these five areas relies upon AI to help address cybersecurity threats.

AI can identify complex patterns, handle large amounts of data, engage in predictions based on analyzing data, allow for automation, promote efficient data review, operate in real time, and help prioritize risk. It acts as a massive force multiplier, aiding humans in identifying, understanding, and acting on cyberattacks in a manner that was previously impossible at scale.

A key feature of AI is its ability to learn over time. As it does so, it takes lessons from previous threats, adapts to new ones, and prepares for upcoming threats. Among the advantages of AI in fighting cyberattacks is that it is anticipatory. This allows AI to act nimbly and robustly, outpacing the abilities of a single person or even teams.

The Department of Homeland Security (DHS) is working to leverage AI to enhance the nation’s security. In an October 17, 2024 publication, it says it has:

“a project underway to research advanced methods for enabling real-time management of cyber threats to critical infrastructure. Another project is increasing the resilience of software analysis tools by helping to identify and mitigate possible weaknesses in Machine Language-based reverse engineering tools, as part of an overarching strategy to assess and mitigate risks of adversarial attacks on AI-based systems.”

This supports the goals of the Cybersecurity and Infrastructure Security Agency (CISA) and provides valuable tools to protect a wide array of American assets that could be compromised or destroyed. 

This increased focus on coordination and personnel to improve the federal government’s ability to harness AI to protect government and government-adjacent assets will pay dividends as AI becomes an indispensable tool in America’s cyber arsenal.

Conclusion

Bad actors are targeting states, businesses, and private sector institutions. In response, AI has effectively been deployed to identify, repel, and learn from these attacks. These techniques are being refined as technologists understand AI threats to cybersecurity, and AI’s innate robustness and ability to learn makes deploying AI in cyber defense increasingly effective. The government of the United States is leaning forward in using and understanding AI as a key tool in the fight against cyberattacks.

Additional Resources

For further reading about AI and cyberattacks, consider the following foundational documents guiding the United States in its efforts to use AI to prepare for, mitigate against, and fight cyberattacks. 

2024 DHS Artificial Intelligence Roadmap. An outline of the Department of Homeland Security’s efforts to use AI to protect America. 

CISA — Cybersecurity Best Practices. Covers how CISA will use AI to improve cybersecurity, emphasizing protecting critical infrastructure.

Three Ways AI Can Be a Game-Changer for Your Security Operations Center. A 2024 report on how AI has emerged as a powerful tool for cybersecurity professionals dealing with threats. 

About the Author: Paul Steidler is a Senior Fellow with the Lexington Institute, a public policy think tank based in Arlington, Virginia.