Enterprises are expanding their use of automation in IT, where AI is changing the landscape with trends such as agentic workflows, automated decision-making and AI-driven pipelines.
Automation has been a transformative force in the IT department for decades, mirroring the impact it has had throughout organizational operations.
Now, evolving artificial intelligence capabilities -- and agentic AI in particular -- are unleashing the next wave of automation within organizations. These capabilities will transform workflows in every department, including IT.
The pace of IT spending in this space reflects this new wave of automation. According to the 2026 "Global State of IT Automation Report" from Stonebranch, an IT orchestration and automation software provider, nearly two-thirds (64%) of surveyed IT professionals are investing in cloud automation, and 50% are investing in workload automation/service orchestration and automation platforms. Investments in infrastructure automation (49%), DevOps automation (49%) and workflow automation (45%) round out the top five areas of IT automation investment, Stonebranch research found.
As is the case for automation deployments throughout the organization, multiple factors are driving the increasing use of automation in IT, according to IT leaders. CIOs are implementing automation in IT to increase the speed at which their departments can operate, enhancing their ability to keep pace with business needs. They're also implementing more automation to reduce costs, do more with existing head count and improve the quality of the services and products they provide.
Here, technology leaders and analysts share six trends in IT automation for 2026.
1. Expanding use of automation
Thanks to technology advances, and advances in AI in particular, CIOs can automate more tasks within their organizations -- including in their IT function, said Krishna Prasad, CIO and chief strategy officer at UST, a global digital transformation consultancy.
"We can deliver more automation with today's tools," he explained. "We can now justify the use of automation [in more places] because it has become faster and easier and thus cheaper to automate."
In other words, automation investments can deliver returns in more areas -- including niche tasks that in past years would not be worth automating because of the work and costs involved, he said.
2. AI is automating decisions
Organizations are increasingly using AI to automate decision-making. That's happening in IT operations, too.
"We can deploy more decision-making algorithms and use some level of AI in the mix to interpret data that you couldn't [interpret in the past]," Prasad said.
For example, IT teams are implementing AI that can continuously monitor network traffic, predict bottlenecks and independently reallocate resources to avert problems. Similarly, IT departments are using AI to automate provisioning, configuration and deployment of applications and infrastructure, thereby removing the decisions around those activities and boosting the accuracy, consistency and efficiency of that decision-making in the process.
3. Agentic AI becomes mainstream
According to "The 2026 State of AI Agents Report" from Anthropic, 57% of organizations are now using agents to handle multistage workflows, and 16% have progressed to cross-functional or end-to-end processes spanning multiple teams or business functions.
IT is among the areas within organizations where agents are being deployed, said Joe Raudabaugh, a principal in the digital and analytics practice of Kearney, a global strategy and management consulting firm.
Common examples of agents at work in IT include autonomous incident remediation, where agents detect anomalies and execute remediations, and self-optimizing infrastructure, where agents adjust resources in real time.
In fact, the Anthropic report noted that among the organizations deploying agentic AI, the functional area seeing the biggest impact is software development (57%), followed by customer service (55%), marketing and sales (46%), and supply chain, logistics and operations (44%).
Raudabaugh said as CIOs -- and their C-suite colleagues -- become more confident in the abilities of AI agents and in their skills deploying them, they're implementing agents to automate increasingly complex workflows.
The Anthropic report confirmed as much, finding that 81% of organizations said they are looking beyond simple task automation, with plans for "more complex" AI initiatives in 2026. "Looking at what 'more complex' actually means: 39% expect to develop agents that handle multi-step processes, while 29% plan to deploy agents for cross-functional projects that span multiple teams or departments," the report stated.
4. Automation to create increasing amounts of code
IT departments have used automation in their code generation and deployment operations for many years. However, automation here has advanced well beyond the use of simple scripts to handle repetitive manual tasks -- an early example of automation in coding -- to today's fully integrated AI-driven pipelines and, in the most modern departments, the use of AI agents to write, test, secure and push code autonomously.
According to the 2026 "State of Code Developer Survey" from tech company Sonar, AI-assisted code accounts for 42% of committed code today and is expected to reach 65% by 2027.
Researchers with professional services firm PwC noted in a 2026 report on autonomous software delivery that "Software development is pivoting from handcrafted systems-of-record to agentic, goal-driven applications that translate intent directly into autonomous action."
5. Automation makes everyone in the enterprise capable of coding
Organizations first deployed low-code and no-code platforms to empower workers outside of IT to create code. And while workers used these automation tools to help them transform their work, they're now turning to vibe coding. Vibe coding is more than AI-assisted coding; it automates code creation by enabling users to tell the tool in plain language what they want to accomplish. Users can describe the desired outcome and let AI handle the architecture, API connections and technical execution.
This has democratized coding, Raudabaugh said, adding that it changes development from a mostly or completely IT task to one anybody and everybody can tackle. Moreover, he noted, people use vibe coding to automate pieces of their regular work responsibilities, thereby further extending the use of automation in the organization.
6. Data remains an obstacle for automation ambitions
Executive ambitions to automate more tasks, processes and workflows are high, but they continue to face challenges in achieving their goals due to data issues, said Craig Le Clair, vice president and principal analyst at Forrester Research.
Most organizations have scattered and siloed data, and they're underestimating the work they have to do to get it ready for [advanced agentic AI uses].
Craig Le ClairVice president and principal analyst, Forrester Research
"Most organizations have scattered and siloed data, and they're underestimating the work they have to do to get it ready for [advanced agentic AI uses]," he added.
The Anthropic report noted that 42% of surveyed professionals said they either faced or expect to encounter data access and data quality issues when deploying AI agents. Experts said data access and data quality issues exist in all functional areas in many organizations, including IT.
The future of IT automation
Although automation scripts, robotic process automation and machine learning will each continue to have a role within IT and other functional areas, automation using agentic AI is "the big opportunity" on the horizon, Le Clair said.
Forrester researchers said they expect agents will be used to automate everything from customer service to back-office workflows, such as those in IT operations, and the work of physical robots.
Forrester classifies agents into three types: solver agents, or purpose-built AI agents that are highly specialized and task-specific; worker agents, or AI specialists that orchestrate across workflows, act autonomously and also collaborate directly with human workers; and executive agents, or orchestrators -- the highest tier of agentic AI -- which coordinate and manage lower-level agents to ensure their actions align with business objectives and that they operate within guardrails.
"In a truly agentic system, you would say, 'Here's my goal,' and it would invent new tools and processes to do that," Le Clair said, noting that the system would have extensive authority and autonomy to take action, such as the authority to replace or eliminate steps within a workflow as part of its self-optimization.
Executive agents aren't in production yet, even though the technology exists to develop and deploy them, Le Clair said. But they're coming as soon as organizations have the necessary control planes and governance in place -- and once workers develop trust in them.
The Stonebranch survey found that IT professionals are looking forward to deploying such technology in their space: "Respondents showed strong interest in automating both sides of the business: front-office platforms like CRMs and ERPs, and back-office systems like ITSMs, databases, and cloud infrastructure. This demand signals a bigger shift: teams don't just want more automation, they want connected workflows where customer requests, cases, and orders trigger governed execution throughout the enterprise."
Raudabaugh predicted that as IT organizations deploy more intelligent automation, IT operations will be one area where systems act proactively. He said in such an environment, instead of IT using automation to enable self-service password resets, the system would recognize when someone needs to reset their password, alert them to that need and work with them through the process.
Mary K. Pratt is an award-winning freelance journalist specializing in enterprise IT, cybersecurity strategy and data management.
