Analyst(s): Fernando Montenegro
Publication Date: April 7, 2025
Organizations are rapidly adopting agentic AI technologies that promise significant business productivity gains through autonomous software agents. This emerging “digital labor” comes with complex security challenges spanning governance, software development, identity management, etc. Futurum explores how organizations are navigating this multi-layered security landscape.
Key Points:
- Agentic AI combines large language models with supportive tools to create autonomous systems that make decisions and take actions with minimal human intervention.
- Nearly 90% of organizations expect agentic AI to play a strategic role, but 78% identify trust and governance as key barriers to adoption.
- Securing agentic AI requires addressing challenges across multiple domains: baseline software security, LLM-specific protections, and agentic-specific concerns.
Overview:
The rise of interest in agentic AI – autonomous software capable of goal-driven behavior – presents organizations with transformative opportunities and significant security challenges. Unlike traditional automation, modern agentic AI aims to have advanced reasoning, planning, and coordination capabilities that can function as “digital labor” across business operations.
Securing agentic AI requires a multi-layered approach. At the foundation, agents are software instances requiring traditional security protections. This includes proper identity management for non-human identities, comprehensive data security controls, risk management frameworks, secure application development practices, and robust security operations monitoring.
The second layer addresses security concerns related to large language models that power these agents. Organizations must implement protections against prompt injection vulnerabilities, sensitive information disclosure, and data/model poisoning using frameworks such as OWASP’s LLM security list and MITRE’s ATLAS, SANS Critical AI Guidelines, and others.
The third layer focuses on agentic-specific security considerations: protecting reasoning/planning capabilities, securing agent memory, maintaining accurate identity for audit purposes, controlling access to remote tools, managing multi-agent interactions, and incorporating appropriate human oversight.
Organizations typically begin with smaller experimental projects before moving to production. This approach requires security measures that don’t impede innovation while ensuring proper safeguards. Successful implementation requires close collaboration between security teams and agentic development teams to bridge knowledge gaps.
Industry observers should watch several key developments:
- How enterprise agentic AI platforms evolve their trust and governance capabilities
- How established security vendors position themselves in the agentic AI security landscape
- The role of startups in developing specialized agentic AI security solutions
- The ongoing evolution of regulatory frameworks affecting agentic AI deployments.
The full report is available via subscription to Futurum Intelligence’s Cybersecurity IQ service—click here for inquiry and access.
Futurum clients can read more in the Cybersecurity Intelligence Portal. Nonclients can learn more here: Cybersecurity Practice.
About the Futurum Cybersecurity Practice
The Futurum Cybersecurity Practice provides actionable, objective insights for market leaders and their teams so they can respond to emerging opportunities and innovate. Public access to our coverage can be seen here. Follow news and updates from the Futurum Practice on LinkedIn and X. Visit the Futurum Newsroom for more information and insights.
Disclosure: While preparing this work, the author used Anthropic Claude Sonnet to summarize the original report. After using this service, the author reviewed and edited the content as needed. The author takes full responsibility for the publication’s content.
Author Information
Fernando Montenegro serves as the Vice President & Practice Lead for Cybersecurity at The Futurum Group. In this role, he leads the development and execution of the Cybersecurity research agenda, working closely with the team to drive the practice's growth. His research focuses on addressing critical topics in modern cybersecurity. These include the multifaceted role of AI in cybersecurity, strategies for managing an ever-expanding attack surface, and the evolution of cybersecurity architectures toward more platform-oriented solutions.
Before joining The Futurum Group, Fernando held senior industry analyst roles at Omdia, S&P Global, and 451 Research. His career also includes diverse roles in customer support, security, IT operations, professional services, and sales engineering. He has worked with pioneering Internet Service Providers, established security vendors, and startups across North and South America.
Fernando holds a Bachelor’s degree in Computer Science from Universidade Federal do Rio Grande do Sul in Brazil and various industry certifications. Although he is originally from Brazil, he has been based in Toronto, Canada, for many years.
