Agentic AI Becomes Essential: Why Adoption Is Accelerating and What Comes Next

The cautious optimism business leaders held towards AI agents has evolved into more widespread enthusiasm. In our last survey from April 2025, just over half (51%) of companies had deployed AI agents in their organization . Six months later, 75% of companies are deploying more than one agent, according to PagerDuty's latest research .

This 24-point increase is further evidence of something we've witnessed over the past year: As the range of applications for AI agents expands, business leaders increasingly see them as a mission-critical layer of their operations stack.

In this article, we'll break down our findings and discuss two use cases driving this change. We'll also look at how the latest numbers speak to the risks and responsibilities that come with agentic AI's increased operational maturity, and how organizations can position themselves to succeed.

New use cases accelerate adoption

Across departments, organizations have begun harnessing the power of AI agents to build and act with autonomy. Leaders are exploring new use cases while ramping up existing applications with increased operational maturity.

Here are two areas where our survey shows AI agents evolving into critical parts of operations.

Coding

Since the release of the first GPT models, development work has been the proving ground for AI adoption. Our survey found that 84% of companies already use AI to write, review, and suggest code. The appeal is clear: Generative coding helps teams move faster, reduce errors, enforce consistent practices, and free engineers to focus on higher-value design and problem-solving.

Now, AI agents are accelerating development and improving engineers' skills. Early experiments with code completion are giving way to more ambitious use cases, from automated testing and debugging, to real-time security checks and even architecture-level recommendations. As this continues, we anticipate AI agents to become trusted collaborators across the entire software development lifecycle.

Incident response

Incident response teams have seen significant benefits from using generative AI to surface relevant context faster: pulling logs, past incidents, and dependency data into one view, and generating tailored updates for customers and internal stakeholders.

AI agents take this further by closing the loop between detection and resolution. Instead of only assisting humans, agents can auto-remediate well-understood issues, trigger failovers, and coordinate workflows across systems.

As agentic AI matures, there's no better indicator of leadership's growing confidence than their willingness to deploy it in an emergency. Our research found that 81% of executives now trust AI agents to act on their behalf during a crisis.

We expect the next step to be full integration into operations. AI agents have the potential to act as frontline responders, detecting anomalies, initiating containment, and escalating only when human judgment is essential. By reducing noise, shortening resolution times, and orchestrating cross-team workflows, AI agents will become a critical layer of resilience for modern enterprises.

Output quality and good governance build trust

Our survey found that 77% of respondents have more confidence in AI-generated outputs than a year ago. Nearly half of respondents point to improved output quality as the main reason. Other factors include gaining a better understanding of how AI works and seeing successful AI use in other teams or at other companies.

This growing confidence also comes from deliberate investments in governance, oversight, and guardrails that help organizations adopt agentic AI responsibly.

Growth creates new responsibilities

Even as automation expands and trust grows, leaders are recognizing that overreliance on AI agents can introduce new risks. This reflects what typically happens when a new technology is deployed: reliability, governance, and resilience move to the top of the agenda.

Treating agentic AI like core infrastructure means ensuring that it is reliable and resilient. In our survey, 85% of respondents expressed the need for stronger procedures to detect AI errors and failures.

Although there is still work to be done, nearly all (96%) respondents are at least somewhat confident their company can detect and mitigate AI failures before they affect operations.

Guardrails build trust

Investments in governance and safeguards reinforce trust in AI agents. In France, where trust in AI agents rose the most in our survey, more than half of leaders said better oversight and control measures drove that shift.

Much of this responsibility depends on vendor relationships. As organizations rely more heavily on external providers for core AI infrastructure, they need confidence in the vendor's reliability, transparency, and responsiveness.

The flywheel effect

With rising trust has come a greater willingness to deploy agentic tools. Three-quarters of organizations now use more than one AI agent; a quarter report running five or more.

The result is a flywheel effect: Increased usage further builds trust among leadership, which opens the door to more potential applications.

Managing complexity takes people power

As AI agents become embedded in daily operations, organizations face the challenge of ensuring they have the people and skills internally to manage the added complexity. Today, 76% of organizations running multiple agents believe AI-driven complexity will soon outpace the number of people available to oversee it.

Resilience depends on the intersection between people and technology. Teams need staff who can monitor systems, validate outputs, and intervene when things go wrong. Building and retaining expertise will be essential for organizations that want to harness the benefits of AI agents without letting complexity overwhelm them.

How does your AI maturity stack up?

AI agents are forming the new foundation of business operations. The companies that embrace them today are building the next layer of their competitive advantage. Faster product cycles, sharper customer insights, and more resilient operations are already within reach for early adopters.

As with any core technology, organizations need to implement AI with care to benefit. That means investing in governance, retaining and training the people who can manage complexity, and building vendor relationships that ensure reliability and transparency.

Leaders who approach AI agents strategically will build a durable foundation for growth. Those who don't risk falling behind as competitors scale with a new layer of technology already in place.

To see what that future can look like, download the full PagerDuty report and explore our AI agents .