Skip to main content
Mallory
← Back to News

AI-Driven Threats and Defenses in Modern Cybersecurity

4 sources · Updated 5mo ago · First seen Sep 29, 2025

AI-Driven Threats and Defenses in Modern Cybersecurity

Cybersecurity leaders are facing a rapidly evolving threat landscape shaped by the integration of artificial intelligence (AI) into both attack and defense strategies. According to a report from VikingCloud, 71% of surveyed security leaders have observed an increase in the frequency of cyberattacks over the past year, with 61% noting that the impact of these incidents has also grown. Nation-state actors are increasingly targeting a broader range of industries, including retail, healthcare, and hospitality, often leveraging AI to scale their operations and exploit third-party software vulnerabilities. The use of generative and agentic AI by attackers has enabled more sophisticated phishing, social engineering, and ransomware campaigns, with deepfakes and voice-based fraud becoming more prevalent. Security leaders report that AI is now commonly used to automate the creation of polished, linguistically sophisticated phishing emails, making them harder to detect and more convincing to recipients. Attackers are also targeting AI assistants embedded in workplace tools, using malicious prompts hidden in emails to manipulate these systems into leaking sensitive information or executing unauthorized actions. Research from Columbia University and the University of Chicago, as well as Barracuda Research, has documented how attackers exploit weaknesses in AI-driven security tools and tamper with systems that use retrieval-augmented generation (RAG), leading to unreliable outputs and poor decision-making. Despite these challenges, organizations are increasingly adopting AI to bolster their defenses, automating threat detection, identity management, and patching to free up security teams for more strategic tasks. However, a significant skills gap persists, as highlighted by the AI Workforce Consortium, which found that AI technical skills are now required across 50 job roles in ICT, yet the market lacks enough professionals with these capabilities. The demand for expertise in areas such as LLM architecture, prompt engineering, and generative AI is outpacing supply, putting additional pressure on security teams. Security leaders are also concerned about reduced government oversight, with many citing cuts to agencies like CISA and NSA as factors that have increased organizational vulnerability. The interplay between AI-driven attacks and defenses is creating new security, compliance, and governance challenges that organizations must address to safely deploy advanced AI systems. As both attackers and defenders race to leverage AI, the need for skilled professionals and robust governance frameworks has never been greater. The evolving threat landscape underscores the importance of continuous learning and adaptation for cybersecurity teams. Organizations must remain vigilant against AI-powered threats while investing in workforce development to close the skills gap. The integration of AI into cybersecurity is not only transforming technical requirements but also reshaping risk management and strategic decision-making at the highest levels.

Do you run this?See affected versions and whether adversaries are exploiting it against assets like yours.
Check your exposure
Event timeline

How this story unfolded

Events from the most recent confirmed update back to the earliest known activity.

4 events · Jun 10, 2024 – Sep 29, 2025
  1. Sep 29, 20251y ago

    Defenders recommend AI-specific controls for enterprise security tools

    The article recommended AI-aware detection, memory validation, isolation, least-privilege access, zero-trust verification, and employee awareness training to mitigate risks from AI assistants and agentic systems.

  2. Sep 29, 20251y ago

    Security experts detail prompt injection and RAG poisoning risks

    The reporting outlined attacker experimentation with prompt injection, retrieval-augmented generation poisoning, and abuse of AI-enabled security workflows to manipulate AI assistants and downstream actions.

  3. Sep 29, 20251y ago

    Research highlights growing abuse of generative AI in cyberattacks

    By 2025, research from Columbia University, the University of Chicago, and Barracuda Research described how attackers were using generative AI to improve spam, phishing, and malicious code, while also targeting AI systems as new attack surfaces.

  4. Jun 10, 20242y ago

    Security teams shift strategies to address AI-powered threats

    A Help Net Security report said cybersecurity professionals were changing their defensive strategies in response to the rise of AI-powered threats. The development reflects earlier industry adaptation to AI-enabled attacks before the later 2025 research-focused reporting.

    Cybersecurity pros change strategies to combat AI-powered threats - Help Net Security
Do you run this?See affected versions and whether adversaries are exploiting it against assets like yours.
Check your exposure
Mallory platform

See the full picture, correlated to your attack surface.

This page covers what’s public. Mallory adds the parts that aren’t: which of your assets are affected, which threat actors are using it right now, and what to deploy next.

Exposure mappingMap indicators from this story to your assets and identify affected systems in minutes.
Threat actor evidenceEvery observed campaign, victim, and pivot linked to actors named in this story.
Associated malwareMalware, exploits, and IOCs connected to the activity described here.
Detection signaturesYARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Scheduled alertsGet matching new stories delivered to your team as they break, not the next morning.
AI threadsAsk questions about this story and take action on the answers.