- F5's 2025 AI Application Strategy Report revealed that on average 25% of AI use AI, but only 2% of companies that are certified as extremely prepared for AI.
- While 77% of businesses are AI ready, they still face important security and governance hurdles.
- 71% of organizations use AI to increase security, while only 31% have deployed AI firewalls.
Seattle, Washington – F5 (NASDAQ:FFIV), the global leader in delivering and protecting all apps and APIs, today released its 2025 AI Application Strategy Report, revealing that only 2% of global organizations are ready to safely expand AI. The report summarises insights from 650 global IT leaders and additional research with 150 AI strategists, representing an organization with at least $200 million in annual revenues.
The report uncovers the harsh truths about today's AI readiness and the ability to adapt quickly enough to accommodate new innovations. The most notable findings from the report reveal that 77% of companies show moderate AI preparation, but most lack robust governance and cross-cloud security, putting them at risk. Meanwhile, 21% of companies fall into the low-reservation category, limiting their competitiveness as AI transforms industry.
F5 research reveals trends that show a rapid expansion of AI use by companies today. It is said that 70% of moderately prepared organizations have an actively used generation AI, and virtually everyone else is working on it. Additionally, 25% of apps use AI on average. Usually, ready organizations use AI at a much higher percentage and expect saturation of the entire portfolio. Low-understanding organizations use AI in less than a quarter of apps, typically in siloed or experimental settings. Currently, organizations that are reasonably prepared have AI in about a third of their applications.
This report provides a snapshot of the latest trends from companies working to embrace AI. Almost two-thirds of survey respondents (65%) use two or more paid models and at least one open source model. The average organization uses three models, and the use of multiple models correlates with deployment in multiple environments or locations. Most of the models currently in use are paid models such as GPT-4, but open source alternatives are also popular. The top open source models cited are the Rama variants of Meta, the Mistral AI variants, and Google's Gemma.
“When AI is at the heart of business strategy, preparation requires more than experimentation. It requires security, scalability, and alignment,” said John Madison, F5's Chief Products and Corporate Marketing Director. “This report highlights practical steps for organizations to operate AI with confidence. AI is already transforming security operations, but without mature governance and dedicated protection, businesses risk amplifying threats.”
Cybersecurity challenges in AI adoption
This report highlights critical cybersecurity issues as organizations expand AI capabilities and reveals trends regarding the enterprise's ability to tackle the complexity of protecting AI workloads. The key cybersecurity trends identified in the report include:
- Organizations see AI as a viable cybersecurity asset. 71% of all respondents already use AI to enhance security.
- There is a lack of AI-specific protection: Only 18% of moderately prepared organizations deploy AI firewalls, with 47% aiming to do so within a year.
- Weaknesses in data governance: Only 24% of organizations practice continuous data labeling, indicating a decrease in transparency and an increased risk of hostile attacks.
- Cross-cloud contradiction: Hybrid environments create governance gaps and expose workflows and data to vulnerability.
- Extended Attack Surface: The use of diverse AI models exacerbates risk without the proper control framework of open source tools.
Recommendations for improving AI preparation
This report introduces the AI Readiness index. It is a framework that measures six factors of operational maturity, including security and infrastructure adjustments. F5 outlines key actions for businesses to enhance AI scalability and security.
- Diversification of AI models: Use both paid and open source AI tools while improving governance to mitigate risk.
- Extend AI use across workflows: Move beyond pilots and embed AI in operations, analytics and security for enterprise-wide transformation.
- AI-specific security integration: Deploy protections such as AI firewalls and formalize data governance processes, including data labeling, to protect your workflows.
Organizations with a wealth of AI preparation can effectively scale, mitigate risk and leverage innovation strategically. Those without a mature framework face operational bottlenecks, compliance challenges and stifling growth. The AI Readiness Index serves as a roadmap for businesses to benchmark progress and implement actionable changes for safe scalability. Download the 2025 State of AI Application Strategy Report to benchmark preparation, identify gaps, and accelerate secure AI scaling.
Support Materials
