HomeRisk ManagementsAI Reduces Vulnerability Discovery Time from Nine Months to 90 Hours

AI Reduces Vulnerability Discovery Time from Nine Months to 90 Hours

Published on

spot_img

Forescout Technologies has recently contributed to the ongoing dialogue regarding the integration of artificial intelligence (AI) within security frameworks, emphasizing how the technology has significantly enhanced the speed of vulnerability discovery in various software ecosystems. In an age where cyber threats are becoming increasingly sophisticated, Forescout’s insights provide a timely perspective on how security teams can adapt and respond to emerging challenges.

A report titled Finding Critical Infrastructure Bugs with Off-the-Shelf AI offers a comprehensive exploration of how publicly available AI models can assist organizations, particularly those operating on limited budgets, in identifying vulnerabilities within both open-source software and proprietary binaries. This focus on accessibility underscores the growing recognition that effective cybersecurity does not solely rely on extensive resources, but can leverage existing technologies to improve outcomes.

The report highlights a noteworthy methodology employed by researchers at Forescout, who developed an AI-assisted vulnerability research harness specifically designed to perform rigorous analysis on 10 commonly used TCP/IP stacks. The results are striking: this innovative approach condensed the typical analysis time from approximately nine months to a mere 90 hours. During this accelerated timeframe, the team successfully identified 99 vulnerabilities, 66 of which were classified as zero-day flaws. Such findings not only demonstrate the capabilities of AI in identifying security weaknesses but also raise critical questions about the implications for both defenders and attackers in the cybersecurity landscape.

After conducting AI-driven analyses for about 10 hours, the researchers dedicated another 80 hours to validate the findings manually. Remarkably, the entire static analysis process across the ten stacks incurred a cost of around $935, which translates to an impressive $2.58 per vulnerability discovered. However, it is essential to note that the expense and time associated with binary analysis were significantly higher—about six times more costly and requiring double the duration compared to source code analysis.

Daniel dos Santos, Vice President of Research at Forescout, articulated that while AI does not replace the necessity for expert vulnerability researchers, it has undeniably accelerated the work significantly. He stated, "With access to an effective AI-assisted workflow, vulnerabilities in complex embedded software can be found much faster and at a fraction of the cost, by security researchers and attackers alike." This remark underscores a dual reality: while security professionals can utilize AI to enhance their capabilities, malicious actors can similarly exploit AI for their purposes.

Forescout’s research draws attention to the shifting economics surrounding vulnerability discovery. This acceleration is particularly crucial for critical infrastructure sectors, where aged and difficult-to-patch devices may remain operational for extended periods. The urgency of addressing vulnerabilities in such environments compels security teams to prioritize their strategies more effectively.

Nevertheless, Forescout cautions that AI is not a panacea for all security challenges. Research conducted by ReliaQuest indicates that while AI can make attacks cheaper and easier to execute, this technology can also lead to inefficiencies in some cases. Experts, including a representative from Proofpoint, have noted that AI-generated exploits can often be flawed and less effective than traditional methods. This perspective emphasizes the nuanced role AI plays in the cybersecurity domain—while it represents a powerful tool for vulnerability discovery, it can also complicate the threat landscape.

To navigate this new environment, Forescout has put forth a series of recommendations aimed at enhancing security measures amidst the proliferation of AI technologies. Key suggested strategies include:

  1. Risk Assessment and Prioritization: Organizations should assess and prioritize risks based on factors such as reachability, exploitability, and operational impact, rather than treating all vulnerabilities as equal threats.

  2. Minimizing Exposure: Reducing unnecessary exposure of connected devices by disabling certain services, including FTP, Telnet, IPv6, and web administration interfaces, can significantly enhance security posture.

  3. Device Isolation: Isolating devices that cannot be patched, particularly those that have reached their end of life, is critical in preventing vulnerabilities from being exploited.

  4. Accelerating Patching Timelines: Organizations should strive to expedite patching processes for known vulnerabilities, considering the rapid pace of vulnerability disclosures.

  5. Building Security into Procurement: Companies should incorporate security measures into their procurement processes, requiring suppliers to outline support lifetimes, disclose vulnerabilities, provide software bills of materials (SBOMs), and showcase ongoing security testing protocols.

In conclusion, Forescout’s insights highlight that AI can dramatically enhance the capabilities of security teams in vulnerability discovery, allowing them to respond swiftly to cyber threats. However, the dual-use nature of this technology necessitates astute strategies and caution from both defenders and adversaries alike. As the interplay between AI and cybersecurity continues to evolve, organizations must remain vigilant and proactive in adapting to these shifts to safeguard their infrastructures effectively.

Source link

Latest articles

DOQEX Expands Identity Management Support and Eliminates Extra SSO Licensing Fees

UK-based secure data exchange specialist DOQEX has made significant strides in enhancing its Single...

Cyber Briefing – November 16, 2026

Cybersecurity Briefing Highlights Significant Vulnerabilities and Incidents in the Tech Landscape In the ever-evolving landscape...

ICO Receives Over 300 Reports of Medical Snooping

The UK’s data protection authority, the Information Commissioner’s Office (ICO), has taken a proactive...

More like this

DOQEX Expands Identity Management Support and Eliminates Extra SSO Licensing Fees

UK-based secure data exchange specialist DOQEX has made significant strides in enhancing its Single...

Cyber Briefing – November 16, 2026

Cybersecurity Briefing Highlights Significant Vulnerabilities and Incidents in the Tech Landscape In the ever-evolving landscape...