Enterprise spending on application security is set to approach US$13 billion globally by 2031 as organisations respond to the security risks associated with artificial intelligence-assisted software development.
Juniper Research forecasts that enterprise AppSec spending will rise by more than 16% over the next five years from US$11 billion in 2026. Juniper said the increase would be driven partly by the need to secure larger and faster-moving volumes of code produced with AI tools.
The report identifies “vibe coding” — the use of AI to help generate software code — as a growing challenge for application security teams. While AI-assisted development can accelerate software creation, Juniper warned that it can also reintroduce basic vulnerabilities and overlooked dependencies into applications.
Traditional AppSec tools may struggle to keep pace with the volume of code generated by AI. The research argues that the market risks remaining focused on a scan-and-report model, in which vulnerabilities are identified after development activity has already taken place, rather than being prevented earlier in the software lifecycle.
“Vibe coding will shift AppSec from securing a relatively controlled development pipeline, to continuously securing a much larger and faster-moving volume of AI-generated code,” said Thomas Wilson, research analyst at Juniper. He posits that automation and security-by-design would become increasingly important as enterprises expanded their use of AI coding tools.
Juniper’s research recommends that organisations treat AI-assisted development as a catalyst for higher AppSec investment rather than restricting the use of coding tools. It said enterprises should adopt AI-aware platforms that can detect vulnerabilities earlier, assess insecure coding practices and identify weaknesses in third-party and open-source dependencies.
The company’s 2026 Competitor Leaderboard identified Veracode, Checkmarx and Black Duck as the three leading AppSec vendors. Juniper said the three companies share several characteristics, including broad application-security portfolios, automated vulnerability detection and coverage across the software development lifecycle.
The wider leaderboard evaluated 15 vendors, including Contrast Security, Cycode, Endor Labs, GitLab, HCLTech, Mend.io, OpenText, Palo Alto Networks, Qualys, Semgrep, Sonatype and Snyk. Juniper assessed the companies according to their capacity and capability, placing them in categories including established leader, leading challenger, and disruptor and challenger.
The findings underline a broader change in software risk management. As AI becomes embedded in development workflows, security checks must increasingly operate continuously across planning, coding, testing, deployment and maintenance.
The challenge for enterprises will be to maintain development speed while ensuring that AI-generated code meets security, compliance and quality requirements before it reaches production.
