According to a recent LinkedIn post from Cyberhaven, data from Cyberhaven Labs suggests enterprise adoption of endpoint-based AI agents has risen 276% over the past year, outpacing growth in GenAI SaaS tools by more than threefold. The post also indicates endpoint coding assistants reportedly expanded from 20% to 50% adoption in 2025.
Claim 30% Off TipRanks
- Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
- Discover top-performing stock ideas and upgrade to a portfolio of market leaders with Smart Investor Picks
The company’s LinkedIn post characterizes this shift as AI moving from content generation to executing work by accessing data, using tools, and taking actions on endpoints. In this context, Cyberhaven highlights a new offering branded as Agentic AI Security, positioned to provide visibility, observability, and real-time controls for AI agents operating on endpoints.
The post further argues that many security programs still concentrate on user prompts rather than the behavior of AI agents themselves, pointing to what the company describes as a protection gap. If adoption trends and associated security risks continue, Cyberhaven’s focus on endpoint-based AI security could support demand for its platform and potentially strengthen its competitive standing in enterprise data security.
For investors, the emphasis on endpoints as a “new control plane for AI” suggests a strategic bet on where security budgets may shift as AI agents become more embedded in day-to-day workflows. Successful commercialization of Agentic AI Security and proof of traction with large enterprises would be key factors in translating this product positioning into revenue growth and enhanced market relevance.