Enterprise security leaders are facing a new reality as artificial intelligence moves from experimentation to production, according to F5’s latest State of Application Strategy Report.
The report identifies three major forces transforming how organizations build, operate and secure digital experiences: the rise of distributed AI inference, the rapid evolution of AI-powered cyber threats, and the growing complexity of hybrid multicloud environments.
According to F5, AI inference—the process of deploying trained models to generate real-time responses, predictions and decisions—has emerged as the primary source of business value from artificial intelligence. The survey found that 77% of respondents now consider inference their dominant AI activity, surpassing model training and tuning.
Organizations are increasingly operating multiple AI models simultaneously, with inference workloads spread across data centres, public clouds and edge locations. Respondents reported running an average of seven AI models, creating new operational and security challenges.
“AI has moved from experimentation to production-scale operations that directly impact revenue and business performance,” said John Maddison, Chief Marketing Officer at F5.
The report highlights growing concerns around the complexity of multi-model AI environments, where prompts, responses and contextual data move between systems in real time. This creates new security risks and increases the need for visibility, governance and policy enforcement across distributed infrastructures.
At the same time, cybercriminals are adopting AI tools at a similar pace. While 98% of surveyed organizations said they are adapting applications for autonomous AI agents, 77% expect identity and access management challenges related to these systems.
F5 argues that AI is fundamentally reshaping the attack surface by introducing threats such as prompt injection, data poisoning and model inversion attacks, which often bypass traditional security controls.
The report also underscores the growing complexity of enterprise infrastructure. More than 93% of organizations now operate multicloud environments, while 86% run applications across a mix of on-premises, public cloud and co-location facilities.
On average, respondents manage five private data centres, five co-location sites and four public cloud providers, illustrating how distributed modern application environments have become.
To address these challenges, F5 advocates a platform-based approach that combines networking, application delivery and security into a single operational framework. The company says organizations need consistent policy enforcement, end-to-end visibility and integrated governance across all environments to manage enterprise-scale AI workloads effectively.
The findings reflect a broader shift underway across enterprises, where AI adoption is not only driving innovation but also forcing security and technology leaders to rethink how digital infrastructure is secured, managed and governed.
