Agentic AI Adoption Soars, Tech Job Growth Stalls, Study Shows

The slowdown has also hit wages. Between 2016 and 2023, IT salaries grew roughly 50%. Over the past two years, wage growth has dropped to just 3% per year. 

This uncertainty underscores the need to evolve skill sets toward higher-value work, such as designing governance frameworks, orchestrating complex workflows, and building the infrastructure that enables agentic AI. For enterprises, it reinforces the urgency of preparing systems, APIs, and governance models to handle autonomous operations at scale.

Are enterprises prepared for agentic AI adoption?

90% of survey respondents with visibility into their organization’s plans say their companies are actively adopting AI agents. This isn’t a cautious “wait and see” approach; organizations are putting strategies in place now. More than half (52%) even report having a clearly defined roadmap for adoption.

The timelines for enterprise-wide implementation are aggressive: 79% believe their organizations will reach full-scale adoption within three years. 

Much like cloud migration a decade ago, enterprises are recognizing that the question isn’t whether they will embrace agentic AI, but how quickly they can do it without being left behind.

What’s driving enterprise adoption of agentic AI?

The motivations behind this rapid adoption are both pragmatic and strategic. Reducing costs and increasing efficiency emerged as the single biggest driver, cited by 23% of survey respondents. Close behind were automating IT operations (21%) and improving development and API management (20%). 

When asked which teams were driving adoption, respondents pointed most often to software development (34%). Developers are at the heart of building and scaling applications, and agentic AI offers clear productivity gains by automating code testing, deployment, and integration tasks.

Other functions making big strides include security and compliance (22%), infrastructure and cloud teams (19%), and data science and analytics (16%). The fact that security and compliance appear so high on this list is worth noting: it shows that adoption isn’t just about speeding up delivery pipelines, but also about strengthening guardrails and enabling organizations to manage risk more effectively.

Which agentic AI use cases are delivering impact?

So where are organizations actually seeing value today? The report highlights several use cases that are moving beyond hype into measurable business impact.

What challenges do enterprises face when adopting agentic AI?

Despite strong momentum, enterprises aren’t adopting agentic AI without friction. Integration complexity with existing systems emerged as the most significant barrier, cited by 31% of respondents. Retrofitting new agentic systems into fragmented legacy infrastructure is proving challenging.

A critical but often underappreciated aspect of this integration challenge is the state of APIs inside most organizations. Agentic workflows depend on agents being able to discover, access, and orchestrate APIs. But if APIs are scattered across systems without a central point of management, discovery becomes nearly impossible. Without a unified platform to bring all APIs together, the promise of agents executing end-to-end workflows breaks down.

Security and compliance concerns were close behind at 28%. As AI agents gain access to sensitive systems and data, organizations are rightfully cautious. Compliance with privacy regulations, auditability, and ensuring proper authentication are top of mind.

Notably, very few organizations doubted the business value of agentic AI. Only 12% of respondents said unclear ROI was their biggest concern, suggesting that most enterprises already see a clear path to value if they can overcome integration and governance challenges.

Why is AI governance essential for enterprise agentic AI adoption?

To tackle these hurdles, enterprises are leaning heavily on governance. Nearly three-quarters (72%) of surveyed organizations have an AI governance solution in place. Among those, over half rely on an AI gateway, a specialized middleware layer that manages, secures, and monitors interactions between AI systems and enterprise infrastructure.

An AI gateway provides policy enforcement, access control, and observability, ensuring that agentic AI can be deployed at scale without undermining compliance or introducing new risks. For many organizations, it’s the missing link that allows them to balance speed and control.

Ownership of governance responsibilities is still evolving. In many cases, platform teams (41%) are leading the charge, given their experience managing APIs and infrastructure. Others are creating dedicated AI teams (30%), while some are taking a decentralized approach, asking developers to follow guidelines without centralized oversight.

How can businesses prepare for the agentic AI era?

The message is clear: agentic AI is on a fast track to becoming a core enterprise function. Adoption is already mainstream, driven by practical goals like cost reduction and efficiency, and reinforced by real-world use cases such as customer support and automation.

At the same time, barriers remain particularly around integration, security, and compliance. Organizations are responding by embracing governance, with solutions like the Kong AI Gateway emerging as critical enablers for scaling responsibly.

Perhaps most importantly, agentic AI is reshaping the workforce in ways that are still playing out. While many see it as a supportive tool, labor market trends suggest that disruption is underway and may accelerate as systems become more autonomous.

The enterprises that succeed in this new era will be those that not only adopt AI quickly but also prepare their infrastructure, governance, and workforce for what comes next.

For a deeper dive into adoption trends, workforce impacts, and the strategies enterprises are using to scale responsibly, check out the Agentic AI in the Enterprise report.