news CyberScoop

Unit 42 warns AI has shifted balance of power from defenders to attackers

Palo Alto Networks’ Unit 42 says frontier AI is creating a major cybersecurity shift that favors attackers. Senior vice president Sam Rubin described it as a “generational shift,” arguing that machine-speed AI attacks are outpacing defenses built for slower, human-driven threats. Unit 42 believes both commercially available agentic AI tools and more advanced frontier models are enabling attackers to find and exploit weaknesses much faster than before. The team says the threat is already emerging in real incidents. In one active investigation, an attacker reportedly used an agentic framework to exploit 50 applications and other enterprise weaknesses in under 10 hours—work Rubin estimates would have taken at least 10 days before AI. According to vice president Sherrod DeGrippo, attackers are using AI across the full attack chain, including malware creation, delegation, social engineering, and ransomware negotiations. She said the landscape is changing in four key ways: AI is multiplying attacker capabilities, identity is becoming the main compromise point, attackers are targeting foundational libraries and software supply chains, and nation-state groups are learning more about enterprise vulnerabilities. Unit 42 warned that organizations are not adequately prepared for these faster, more automated attacks and that how they adapt now will be critical to their future security.

Build this article’s timeline

Extract dated events, key actors, and open questions using the same timeline analysis available in Signal.

Log in to build

Timeline generation is available to signed-in community members.

Five Ws 🤖 3 AI prefills · 2 still unanswered
Who
Sam Rubin, Sherrod DeGrippo, Unit 42, and Palo Alto Networks
🤖 AI prefill
What
Palo Alto Networks’ Unit 42 says frontier AI is creating a major cybersecurity shift that favors attackers.
🤖 AI prefill
When When did this happen, or what is the timeframe?
Where Where did this occur?
Why
Centers on frontier AI model capabilities, agentic AI models, and machine-speed attacks
🤖 AI prefill

Login to answer and earn XP for each gap you fill.

AI Cybersecurity Shift frontier AI, cybersecurity, attackers, defenders, machine-speed Agentic Attack Automation agentic AI, attack chain, exploit, automation, speed Malware And Social Engineering malware, social engineering, ransomware, negotiations, delegation Identity And Supply Chain Risks identity, compromise vector, software supply chain, foundational libraries, enterprise systems Defender Readiness Gap prepared, detect, respond, defenses, organizations
frontier AI model capabilities98% agentic AI models95% machine-speed attacks93% cybersecurity90% attack chain89% threat landscape87% malware development at scale86% social engineering84% ransomware negotiations82% software supply chains81%
42
Manipulation Score: Medium How much this article uses emotional framing vs. straight reporting
Loaded Language 2825%
Absolutist 3415%
Threat/Panic 6225%
Us vs Them 1815%
Engagement Bait 3020%
Manipulation score look right?

Factual Claims (10)

medium technology
Unit 42 has seen enough internal frontier AI testing and malicious in-the-wild AI use to be genuinely concerned.
medium technology
Frontier AI model capabilities represent a generational shift in cybersecurity.
medium technology
A prior balance between security and exposure has been broken by frontier AI capabilities.
medium technology
Readily available agentic AI models and frontier AI capabilities have shifted power from defenders to attackers.
medium technology
Organizations are ill-equipped to detect and respond quickly to machine-speed attacks.
medium technology
Unit 42 estimated in April that attacker access to these capabilities would occur within a year.
medium technology
Five months later, early waves of this threat are being seen in the wild.
high technology
An attacker used an agentic framework to exploit 50 applications and other weaknesses across an enterprise in less than 10 hours.
medium technology
AI enabled the attacker to do in 10 hours what would have taken at least 10 days before AI.
medium technology
Attackers are already using AI across the entire attack chain, including malware development, social engineering, and ransomware negotiations.

Comments

Login to view and post comments.

🏠Portal 📰Links ❓Q&A 📅Events 💼Jobs