Agentic AI will cause a public breach in 2026 – this is the latest prediction from Forrester senior analyst Paddy Harrington, who warned of the cybersecurity risks accompanying the rise of AI agents.

“Without the right guardrails, systems of autonomous AI agents may sacrifice accuracy for speed of delivery, especially when interacting directly with customers,” he warned in a blog post.

While some companies may still blame individual employees for these breaches, Harrington says they will actually be the result of “a cascade of failures.”

Forrester says that organisations must “develop agentic applications with minimum viable security” to avoid this fallout from agentic AI cyber risks, specifically by following the AEDIS framework.

AEGIS, Agentic AI Guardrails for Information Security, is a six-domain framework aiming develop proper governance, security, and management protocols for agentic AI.

These domains include governance, risk and compliance; identity and access management; data security and privacy; application security; threat management; and zero trust architecture.

To implement these principles, Forrester recommends four steps.

Starting with governance, organisations should focus on leveraging minimal tech for maximum impact, establishing governance at the very start of AI adoption and agentic integration.

Next, organisation should focus on building foundational data security, treating agents as their own identity class.

Threat management and advancing DevSecOps should follow, including securing the agent lifecycle, and detecting flaws such as hallucinations.

Finally, organisations should ensure that zero trust principles are implemented across the technology stack and the entire AI application tool.

Recommended reading

• Report: 28% Global Cyber-attack Increase in Q1 2024
• 92% of Enterprise Devices Unprepared for AI Security Challenges
• NCSC Releases Cyber Incident Response Guidance for CEOs
• Over 1 in 10 Business Leaders Don’t Know if They’ve Been Hacked

The stark prediction, and advice, comes as more and more AI security risks emerge. Cyber-attacks targeting critical infrastructure and personal data increase as national state threats mount across the globe.

AI is being used as a tool by malicious actors, but its rapid application is also leaving increasing vulnerabilities in organisations hastily integrating the emerging tech.

Forrester also predicted other security risks emerging in 2026 as a result of rising global tensions, such as restrictions are critical telecoms infrastructure in relation to the Salt Typhoon cyberespionage campaign.

However, Harrington also predicted that quantum security spending is set to exceed 5% of overall IT security budgets, meaning that organisations are still seeking out novel technologies to advance their security defences amidst growing threats.