2026 AI Predictions - Part 1

APMdigest's Predictions Series concludes with 2026 AI Predictions — industry experts offer predictions on how AI and related technologies will evolve and impact business in 2026. This is not a comprehensive list of predictions. There are so many aspects to AI and related predictions that APMdigest simply could not cover them all. For example, this series does not delve into AI governance and security, AI's impact on consumers, or the AI market and investments. The main categories covered by the AI Predictions series include solutions and strategies to support AI, challenges and barriers to AI, negative impacts of AI, and AI's impacts on roles. Part 1 covers the Big Picture.

AI: CORE ENGINE FOR IT

In 2026, AI will become the core engine for IT — driving detection, prioritization, and response.
Even with progress, skepticism is healthy. In fact, I count myself among AI skeptics — not because I doubt its value, and not because I am new to the technology. I began applying machine learning for analysis at US Cyber Command in 2018. My skepticism comes from experience with teams or products that expect AI to do everything but ignore what it actually does well.

The art of the possible has leapt forward this year, but most people missed it amid the noise. Frontier-class models, paired with MCP servers that connect them to live tools and data, have moved from promise to proof. And because MCP has emerged as an open standard, organizations can extend models with far greater flexibility than in past proprietary ecosystems. These systems now deliver precise, real-time recommendations for identifying risks, ranking priorities, and accelerating remediation.

Applied strategically, AI can highlight patterns, surface deviations, and flag early indicators of risk while also shrinking patching and remediation windows and taking on routine work. The strongest programs will define where AI should look for abnormal behavior, set boundaries for how those signals are interpreted, and use confidence scoring to decide when an alert is trustworthy or needs human review.

2026 will not be defined only by vendor capabilities. It will be the year organizations extend their own platforms, build new features, and tap directly into frontier models and MCP servers to create capabilities tailored to their environments. Success will come from applying AI with intention — combining automation with human judgment to improve time to patch, reduce manual effort, and strengthen clarity, speed, and resilience without giving up oversight.
Jason Kikta
CTO, Automox

ON-DEMAND WEBINAR: Visibility is Velocity - Bridging Insight and Action in ITOps

THE YEAR OF THE INTELLIGENT ENTERPRISE

Agentic AI makes sense, and it sounds exciting on paper. Bringing it to life has been much harder. Demos have been the best way to see which agents are real and which ones are still on the drawing board. Proliferating success stories gives us confidence that AI can deliver real business value. We now know what good agentic AI looks like. In 2026, we expect these bright spots to scale and become the norm. Leaders have shown us what's possible and how to get it. Laggards have discovered the common pitfalls we want to avoid.  What's emerging is a set of benchmarks that will guide companies on where to invest, how to invest, and what results to expect. New AI prototyping techniques give business teams a better way to express capability requirements that will keep them competitive in the AI race. New capabilities that specify where agents lead, where humans oversee, and how collaboration happens are being designed now and will take flight in 2026. Agentic AI is still evolving, but organizations are learning how to make it work at scale. With new benchmarks, smarter oversight, and clear accountability, 2026 could be the year of the intelligent enterprise.
Dan Priest
US Chief AI Officer, PwC

AI AUTONOMY

AI Will Shift from Being an Assistant to Being an Autonomous Agentic Operator: AI made its mark in the form of passive copilots that are ready when we need help. But the next evolution of AI will take the form of autonomous agents that can execute multi-step tasks, trigger workflows, approve transactions, and collaborate with people in an increasingly sophisticated, complex way.
Paul Moxon
SVP Data Architecture and Chief Evangelist, Denodo

AI moves from copilots to co-operators: We'll see enterprises evolve from experimenting with copilots that assist individuals to deploying "co-operators." AI systems that act across platforms, completing multi-step workflows autonomously. The challenge will shift from building models to governing and integrating them responsibly.
Ha Hoang
CIO, Commvault

Agentic Workflows to Autonomous Agentic AI Systems: Most of what's being sold as "agentic AI" in 2025 is just glorified workflow automation with better prompts. Sure, it's smarter than the old RPA bots, but calling it autonomous is generous. Today's systems are workflows with LLMs integrated - predetermined paths, needing constant human checkpoints, and break the moment something unexpected happens. 2026 is when we actually cross the threshold into true autonomous agents. Systems that genuinely make decisions, adapt to changing conditions without predefined rules, and communicate bi-directionally with other agents and humans to negotiate outcomes. Your supply chain agent doesn't just execute a workflow when inventory drops—it assesses the situation, evaluates multiple solutions, coordinates with procurement and finance agents, and chooses the best path forward without a script. That's the shift: from "follow these steps intelligently" to "here's the goal, figure it out." Most companies aren't remotely ready for this.
Jarrod Vawdrey
Field Chief Data Scientist, Domino Data Lab

PROACTIVE AI

Proactive AI Agents Will Replace Reactive Systems: The buzz around agentic AI has dominated technology conversations throughout 2025, but these systems remain fundamentally reactive, in that they wait for triggers, prompts and human commands. In 2026, we'll witness the emergence of truly proactive AI agents that act autonomously based on context and environmental signals. Consider the difference: today's AI agent requires a prompt to research industry news and draft LinkedIn posts. Tomorrow's proactive agent monitors your calendar, analyzes trending topics in your industry, and suggests content without being asked. This shift transforms AI from sophisticated tools requiring constant human direction into what could genuinely be considered "digital employees" operating with real autonomy. This evolution requires more than incremental improvements to large language models (LLMs). It demands new architectures capable of continuous environmental monitoring, contextual decision-making, and action without explicit triggers. For enterprises, it means rethinking governance models, approval workflows, and trust boundaries. The question shifts from "what can we ask AI to do?" to "what should we allow AI to do on its own?"
Efrain Ruh
Regional CTO, Digitate

The rise of proactive AI. If 2025 was the year AI learned to store memory, 2026 will be the year it thinks proactively. We're witnessing a market-wide shift toward AI systems that can anticipate needs and act without human prompting. Following the debut of long-term memory features from OpenAI, Anthropic, Microsoft, and Google in 2025, the next wave of innovation will focus on proactive, autonomous workflows with AI that not only responds but also initiates, collaborates, and makes decisions. Emerging "anticipatory AI" will redefine both enterprise operations and individual productivity, marking a pivotal step toward true machine collaboration. 
Sarah Hoffman
Director of AI Thought Leadership, AlphaSense

PRODUCTION AI

I believe 2026 will be the year AI stops experimenting and starts executing, at a scale we haven't yet seen. After a cycle of pilots and proofs of concept, businesses are now ready to scale and they're betting big on agentic AI to do it.
Sebastian Enderlein
Chief Technology Officer, DeepL

Production AI will remain narrow but operationally focused: In 2026, enterprise institutions, especially in financial services, will shift from experimentation to a small number of focused production AI use cases, prioritizing efficiency and operational streamlining rather than broad transformation. Most firms will run only a handful of AI systems in live environments, but these will be tightly integrated into workflows where automation yields measurable process improvements. The year will be marked less by scale and more by disciplined, high-value deployment.
Robert Cooke
CEO, 3forge

2026 will be the year of the "AI reality check" as enterprises move from pilot programs to production deployments, discovering that successful AI isn't about the most advanced models but about solving the unglamorous data preparation challenges. Companies will shift their AI investments from moonshot projects to practical applications that address specific operational pain points such as automating partner onboarding or streamlining compliance processes, where ROI can be measured in weeks, not years. The winners won't be those with the biggest AI budgets, but those who've mastered the art of making AI work reliably with real-world, messy data.
Deepak Singh
Chief Innovation Officer, Adeptia

INFERENCING

Keep an eye out for innovation in the inferencing side of the house: We'll see this from a number of silicon manufacturers. Inferencing for the edge is going to come in a huge number of aspects —  from the personal computing devices to those that run branch or remote offices. Silicon and software capabilities for this are evolving at a staggering race and the winner(s) are not clear yet.
Juan Orlandini
CTO, North America, Insight Enterprises

In 2026, the real breakthrough in AI won't come from billion-dollar training runs. Instead, it will come from running hundreds of smaller, specialized models reliably and securely at scale. Inference is where AI gets real. It's not just a model in a lab — it's answering questions, guiding decisions, and powering intelligent services across every industry. That's a fundamentally cloud native challenge, and it's platform engineers who build the open source infrastructure that makes enterprise AI practical and scalable.
Jonathan Bryce
Executive Director, Cloud Native Computing Foundation

INFERENCE OPTIMIZATION

2026 will be the year of inference optimization. AI training has been getting most of the attention but inference is where the money is made. As enterprises shift from building massive models to running them efficiently, inference-optimized hardware and software will dominate R&D. Expect new chips designed solely for inference and tighter collaboration between compute and network layers to extract more performance from existing infrastructure. Simultaneously, efficiency is becoming the real AI arms race. High compute does not always equal high ROI. Organizations will chase smarter utilization to cover bigger budgets. Expect a focus on energy-aware compute strategies, dynamic resource scheduling, and ML-driven automation that maximizes ROI without inflating power consumption or hardware dependency.
Jim Piazza
Chief AI Officer, Ensono

MEASURABLE ROI

Forrester just predicted that enterprises will defer 25% of planned AI spend to 2027. I believe that to be true; in fact, I think it will be closer to 40%. We spent so much of 2024 and 2025 blowing up AI as "the thing," and 2026 will be the year it has to prove itself. We will see a significant focus on results. Did AI make employees more productive? Did AI models replace all that human capital that companies predicted? Was it even useful? These are the questions we will be asking.
Eric Ethridge
Senior Technical Account Manager, DoiT

AI's real impact will come after the hype fades. 2026 will be the year when organizations will stop buying AI tools just because they are labeled with "AI." With higher expectations and tighter budgets, companies will start asking for measurable value: how often the tools are used, whether they improve workflows and whether they make employee collaboration more effective. AI will still transform the workplace, but in a more practical, focused way. The difference is that success will be measured not by how much AI an organization has, but by how much it helps people work better.
Oliver Van Camp
Director of Meeting Room Experiences, Barco ClickShare

Throughout 2025, enterprises have struggled to capture real business value from AI. While we've seen impressive demo after impressive demo, most of these initiatives have fallen short of producing measurable ROI at scale. That's because thus far, much of the focus has rested on generic intelligence and LLMs, which are great at "knowing a lot about a lot," but fail to hold expertise in anything. In 2026, that changes. Companies will move from experimenting with general-purpose AI to deploying business applications centered on specific AI intelligence. Tangible ROI isn't found in generic chatbots; it's found in tailor-made systems that understand the intricacies of each business and every workflow. It's time to leave the AI stagecraft behind and enter into a new phase where AI applications act as true strategic collaborators and revenue drivers.
Kevin Thompson
CEO, Tricentis

IT leaders are under more pressure than ever to innovate fast and prove ROI, and that pressure isn't easing in 2026. In fact, 94% of organizations are investing in AI, but few can measure its impact. Visibility gaps remain, with a majority of IT leaders admitting that business units purchase more SaaS and cloud apps than they are aware of. As AI spend continues to rise, 2026 will focus on redefining success and optimizing costs.
Mike Jerich
President, Flexera

AI ROI and Impact - A Roadmap to Success: Budgets are rising, confidence is high, and the experiences of 2025 have brought organizations far up the AI learning curve. As the focus shifts from experiment to execution, practitioners face intensifying pressure to deliver AI performance. A Microsoft survey revealed that nearly 80% of executives say their boards and investors are calling for a clear AI strategy, demonstrable results, elevated productivity, and measurable ROI. The route to AI payoff will shape investment priorities in agentic programs in 2026 and beyond, focusing on leveraging multi-agent systems, adopting reliable and unattended agents able to work autonomously within established guardrails, and investing in orchestration, and ensuring humans are in the loop to make decisions that drive impact.
Raghu Malpani
CTO, UiPath

The AI bubble won't burst — it'll mature. We've seen a steady increase in mentions of AI investments and ROI across company transcripts, suggesting companies that may have started experimenting with AI in 2025 expect to see tangible business impact in the months ahead. In 2026, we won't have the dramatic "pop" that so many fear is around the corner; instead, we'll see a maturation of the broader technology market as companies shift from general models to domain-specific AI with a greater focus on measurable and trusted impact. The goalposts of "success" are always changing, and companies that win in this next phase of AI will prioritize purposeful deployment, employee training, cultural acceptance, and strong governance.
Sarah Hoffman
Director of AI Thought Leadership, AlphaSense

Agentic AI will Routinely Deliver Nine-Figure Savings, Addressing CFO Mandates in 2026: By 2026, agentic AI will successfully reset enterprise IT economics, moving beyond the initial plateau of productivity gains to deliver game-changing financial results and meet urgent business needs. The technology is here, and the imperative will be for meaningful financial returns: CFOs will mandate business cases rooted in the actions performed, costs avoided, and the value delivered to the company. Global companies will routinely report nine-figure returns to boards and investors. In IT, projects will reduce non-headcount IT spend in addition to amplifying the impact of the workforce. Companies will see dramatic increases in agility and reduce the risk of change, empowering IT organizations to confidently pursue another high-value project each year, transforming their backlog of ideas and shelved projects into real financial returns and cementing ServiceOps as an engine for business results.
Ryan Manning
Chief Project Officer, BMC Helix

AI MOVES BEYOND THE HYPE

During the last years, we have experienced the fastest change of technical possibilities we will see for a very long time. Only 30 years back, a mobile phone was kind of a rarity, which only the very technical early adopters had. Today, AI is taking over our repetitive work, rearranging our schedules, or just helping us get information faster than a classical search engine. Looking forward, this speed of change will not stop. While I do see us at a peak of the hype as of now, the hype will decrease and go into a more regular usage. The last year was about trying out, challenging the models, and, very importantly, also failing. We now need to take what we learned and start setting up reliable and reusable AI frameworks and usage patterns. This is the challenge we need to solve next year and ahead.
Iris Adae
VP of Data & Analytics, KNIME

The honeymoon phase with AI is ending. As the perceived "magic" fades, organizations will begin to approach AI adoption with the same rigor they apply to traditional IT platforms. Scrutiny will increase — not just on the technology itself, but on the vendor's narrative, governance model, and risk posture. Expect clear exit strategies and off-ramps, strong evaluation frameworks, and contingency planning to become standard practice. AI failures will be treated like any other system outage, with drills and failover plans embedded into IT operations. Break-glass moments won't be rare — they'll be anticipated. In short, AI will start to move from being a shiny disruptor to a managed, accountable part of the enterprise stack.
Dominick Profico
CTO, Bridgenext

Go to: 2026 AI Predictions - Part 2, covering solutions and strategies to support AI

Related Links

Hot Topics

2026 AI Predictions - Part 1

APMdigest's Predictions Series concludes with 2026 AI Predictions — industry experts offer predictions on how AI and related technologies will evolve and impact business in 2026. This is not a comprehensive list of predictions. There are so many aspects to AI and related predictions that APMdigest simply could not cover them all. For example, this series does not delve into AI governance and security, AI's impact on consumers, or the AI market and investments. The main categories covered by the AI Predictions series include solutions and strategies to support AI, challenges and barriers to AI, negative impacts of AI, and AI's impacts on roles. Part 1 covers the Big Picture.

AI: CORE ENGINE FOR IT

In 2026, AI will become the core engine for IT — driving detection, prioritization, and response.
Even with progress, skepticism is healthy. In fact, I count myself among AI skeptics — not because I doubt its value, and not because I am new to the technology. I began applying machine learning for analysis at US Cyber Command in 2018. My skepticism comes from experience with teams or products that expect AI to do everything but ignore what it actually does well.

The art of the possible has leapt forward this year, but most people missed it amid the noise. Frontier-class models, paired with MCP servers that connect them to live tools and data, have moved from promise to proof. And because MCP has emerged as an open standard, organizations can extend models with far greater flexibility than in past proprietary ecosystems. These systems now deliver precise, real-time recommendations for identifying risks, ranking priorities, and accelerating remediation.

Applied strategically, AI can highlight patterns, surface deviations, and flag early indicators of risk while also shrinking patching and remediation windows and taking on routine work. The strongest programs will define where AI should look for abnormal behavior, set boundaries for how those signals are interpreted, and use confidence scoring to decide when an alert is trustworthy or needs human review.

2026 will not be defined only by vendor capabilities. It will be the year organizations extend their own platforms, build new features, and tap directly into frontier models and MCP servers to create capabilities tailored to their environments. Success will come from applying AI with intention — combining automation with human judgment to improve time to patch, reduce manual effort, and strengthen clarity, speed, and resilience without giving up oversight.
Jason Kikta
CTO, Automox

ON-DEMAND WEBINAR: Visibility is Velocity - Bridging Insight and Action in ITOps

THE YEAR OF THE INTELLIGENT ENTERPRISE

Agentic AI makes sense, and it sounds exciting on paper. Bringing it to life has been much harder. Demos have been the best way to see which agents are real and which ones are still on the drawing board. Proliferating success stories gives us confidence that AI can deliver real business value. We now know what good agentic AI looks like. In 2026, we expect these bright spots to scale and become the norm. Leaders have shown us what's possible and how to get it. Laggards have discovered the common pitfalls we want to avoid.  What's emerging is a set of benchmarks that will guide companies on where to invest, how to invest, and what results to expect. New AI prototyping techniques give business teams a better way to express capability requirements that will keep them competitive in the AI race. New capabilities that specify where agents lead, where humans oversee, and how collaboration happens are being designed now and will take flight in 2026. Agentic AI is still evolving, but organizations are learning how to make it work at scale. With new benchmarks, smarter oversight, and clear accountability, 2026 could be the year of the intelligent enterprise.
Dan Priest
US Chief AI Officer, PwC

AI AUTONOMY

AI Will Shift from Being an Assistant to Being an Autonomous Agentic Operator: AI made its mark in the form of passive copilots that are ready when we need help. But the next evolution of AI will take the form of autonomous agents that can execute multi-step tasks, trigger workflows, approve transactions, and collaborate with people in an increasingly sophisticated, complex way.
Paul Moxon
SVP Data Architecture and Chief Evangelist, Denodo

AI moves from copilots to co-operators: We'll see enterprises evolve from experimenting with copilots that assist individuals to deploying "co-operators." AI systems that act across platforms, completing multi-step workflows autonomously. The challenge will shift from building models to governing and integrating them responsibly.
Ha Hoang
CIO, Commvault

Agentic Workflows to Autonomous Agentic AI Systems: Most of what's being sold as "agentic AI" in 2025 is just glorified workflow automation with better prompts. Sure, it's smarter than the old RPA bots, but calling it autonomous is generous. Today's systems are workflows with LLMs integrated - predetermined paths, needing constant human checkpoints, and break the moment something unexpected happens. 2026 is when we actually cross the threshold into true autonomous agents. Systems that genuinely make decisions, adapt to changing conditions without predefined rules, and communicate bi-directionally with other agents and humans to negotiate outcomes. Your supply chain agent doesn't just execute a workflow when inventory drops—it assesses the situation, evaluates multiple solutions, coordinates with procurement and finance agents, and chooses the best path forward without a script. That's the shift: from "follow these steps intelligently" to "here's the goal, figure it out." Most companies aren't remotely ready for this.
Jarrod Vawdrey
Field Chief Data Scientist, Domino Data Lab

PROACTIVE AI

Proactive AI Agents Will Replace Reactive Systems: The buzz around agentic AI has dominated technology conversations throughout 2025, but these systems remain fundamentally reactive, in that they wait for triggers, prompts and human commands. In 2026, we'll witness the emergence of truly proactive AI agents that act autonomously based on context and environmental signals. Consider the difference: today's AI agent requires a prompt to research industry news and draft LinkedIn posts. Tomorrow's proactive agent monitors your calendar, analyzes trending topics in your industry, and suggests content without being asked. This shift transforms AI from sophisticated tools requiring constant human direction into what could genuinely be considered "digital employees" operating with real autonomy. This evolution requires more than incremental improvements to large language models (LLMs). It demands new architectures capable of continuous environmental monitoring, contextual decision-making, and action without explicit triggers. For enterprises, it means rethinking governance models, approval workflows, and trust boundaries. The question shifts from "what can we ask AI to do?" to "what should we allow AI to do on its own?"
Efrain Ruh
Regional CTO, Digitate

The rise of proactive AI. If 2025 was the year AI learned to store memory, 2026 will be the year it thinks proactively. We're witnessing a market-wide shift toward AI systems that can anticipate needs and act without human prompting. Following the debut of long-term memory features from OpenAI, Anthropic, Microsoft, and Google in 2025, the next wave of innovation will focus on proactive, autonomous workflows with AI that not only responds but also initiates, collaborates, and makes decisions. Emerging "anticipatory AI" will redefine both enterprise operations and individual productivity, marking a pivotal step toward true machine collaboration. 
Sarah Hoffman
Director of AI Thought Leadership, AlphaSense

PRODUCTION AI

I believe 2026 will be the year AI stops experimenting and starts executing, at a scale we haven't yet seen. After a cycle of pilots and proofs of concept, businesses are now ready to scale and they're betting big on agentic AI to do it.
Sebastian Enderlein
Chief Technology Officer, DeepL

Production AI will remain narrow but operationally focused: In 2026, enterprise institutions, especially in financial services, will shift from experimentation to a small number of focused production AI use cases, prioritizing efficiency and operational streamlining rather than broad transformation. Most firms will run only a handful of AI systems in live environments, but these will be tightly integrated into workflows where automation yields measurable process improvements. The year will be marked less by scale and more by disciplined, high-value deployment.
Robert Cooke
CEO, 3forge

2026 will be the year of the "AI reality check" as enterprises move from pilot programs to production deployments, discovering that successful AI isn't about the most advanced models but about solving the unglamorous data preparation challenges. Companies will shift their AI investments from moonshot projects to practical applications that address specific operational pain points such as automating partner onboarding or streamlining compliance processes, where ROI can be measured in weeks, not years. The winners won't be those with the biggest AI budgets, but those who've mastered the art of making AI work reliably with real-world, messy data.
Deepak Singh
Chief Innovation Officer, Adeptia

INFERENCING

Keep an eye out for innovation in the inferencing side of the house: We'll see this from a number of silicon manufacturers. Inferencing for the edge is going to come in a huge number of aspects —  from the personal computing devices to those that run branch or remote offices. Silicon and software capabilities for this are evolving at a staggering race and the winner(s) are not clear yet.
Juan Orlandini
CTO, North America, Insight Enterprises

In 2026, the real breakthrough in AI won't come from billion-dollar training runs. Instead, it will come from running hundreds of smaller, specialized models reliably and securely at scale. Inference is where AI gets real. It's not just a model in a lab — it's answering questions, guiding decisions, and powering intelligent services across every industry. That's a fundamentally cloud native challenge, and it's platform engineers who build the open source infrastructure that makes enterprise AI practical and scalable.
Jonathan Bryce
Executive Director, Cloud Native Computing Foundation

INFERENCE OPTIMIZATION

2026 will be the year of inference optimization. AI training has been getting most of the attention but inference is where the money is made. As enterprises shift from building massive models to running them efficiently, inference-optimized hardware and software will dominate R&D. Expect new chips designed solely for inference and tighter collaboration between compute and network layers to extract more performance from existing infrastructure. Simultaneously, efficiency is becoming the real AI arms race. High compute does not always equal high ROI. Organizations will chase smarter utilization to cover bigger budgets. Expect a focus on energy-aware compute strategies, dynamic resource scheduling, and ML-driven automation that maximizes ROI without inflating power consumption or hardware dependency.
Jim Piazza
Chief AI Officer, Ensono

MEASURABLE ROI

Forrester just predicted that enterprises will defer 25% of planned AI spend to 2027. I believe that to be true; in fact, I think it will be closer to 40%. We spent so much of 2024 and 2025 blowing up AI as "the thing," and 2026 will be the year it has to prove itself. We will see a significant focus on results. Did AI make employees more productive? Did AI models replace all that human capital that companies predicted? Was it even useful? These are the questions we will be asking.
Eric Ethridge
Senior Technical Account Manager, DoiT

AI's real impact will come after the hype fades. 2026 will be the year when organizations will stop buying AI tools just because they are labeled with "AI." With higher expectations and tighter budgets, companies will start asking for measurable value: how often the tools are used, whether they improve workflows and whether they make employee collaboration more effective. AI will still transform the workplace, but in a more practical, focused way. The difference is that success will be measured not by how much AI an organization has, but by how much it helps people work better.
Oliver Van Camp
Director of Meeting Room Experiences, Barco ClickShare

Throughout 2025, enterprises have struggled to capture real business value from AI. While we've seen impressive demo after impressive demo, most of these initiatives have fallen short of producing measurable ROI at scale. That's because thus far, much of the focus has rested on generic intelligence and LLMs, which are great at "knowing a lot about a lot," but fail to hold expertise in anything. In 2026, that changes. Companies will move from experimenting with general-purpose AI to deploying business applications centered on specific AI intelligence. Tangible ROI isn't found in generic chatbots; it's found in tailor-made systems that understand the intricacies of each business and every workflow. It's time to leave the AI stagecraft behind and enter into a new phase where AI applications act as true strategic collaborators and revenue drivers.
Kevin Thompson
CEO, Tricentis

IT leaders are under more pressure than ever to innovate fast and prove ROI, and that pressure isn't easing in 2026. In fact, 94% of organizations are investing in AI, but few can measure its impact. Visibility gaps remain, with a majority of IT leaders admitting that business units purchase more SaaS and cloud apps than they are aware of. As AI spend continues to rise, 2026 will focus on redefining success and optimizing costs.
Mike Jerich
President, Flexera

AI ROI and Impact - A Roadmap to Success: Budgets are rising, confidence is high, and the experiences of 2025 have brought organizations far up the AI learning curve. As the focus shifts from experiment to execution, practitioners face intensifying pressure to deliver AI performance. A Microsoft survey revealed that nearly 80% of executives say their boards and investors are calling for a clear AI strategy, demonstrable results, elevated productivity, and measurable ROI. The route to AI payoff will shape investment priorities in agentic programs in 2026 and beyond, focusing on leveraging multi-agent systems, adopting reliable and unattended agents able to work autonomously within established guardrails, and investing in orchestration, and ensuring humans are in the loop to make decisions that drive impact.
Raghu Malpani
CTO, UiPath

The AI bubble won't burst — it'll mature. We've seen a steady increase in mentions of AI investments and ROI across company transcripts, suggesting companies that may have started experimenting with AI in 2025 expect to see tangible business impact in the months ahead. In 2026, we won't have the dramatic "pop" that so many fear is around the corner; instead, we'll see a maturation of the broader technology market as companies shift from general models to domain-specific AI with a greater focus on measurable and trusted impact. The goalposts of "success" are always changing, and companies that win in this next phase of AI will prioritize purposeful deployment, employee training, cultural acceptance, and strong governance.
Sarah Hoffman
Director of AI Thought Leadership, AlphaSense

Agentic AI will Routinely Deliver Nine-Figure Savings, Addressing CFO Mandates in 2026: By 2026, agentic AI will successfully reset enterprise IT economics, moving beyond the initial plateau of productivity gains to deliver game-changing financial results and meet urgent business needs. The technology is here, and the imperative will be for meaningful financial returns: CFOs will mandate business cases rooted in the actions performed, costs avoided, and the value delivered to the company. Global companies will routinely report nine-figure returns to boards and investors. In IT, projects will reduce non-headcount IT spend in addition to amplifying the impact of the workforce. Companies will see dramatic increases in agility and reduce the risk of change, empowering IT organizations to confidently pursue another high-value project each year, transforming their backlog of ideas and shelved projects into real financial returns and cementing ServiceOps as an engine for business results.
Ryan Manning
Chief Project Officer, BMC Helix

AI MOVES BEYOND THE HYPE

During the last years, we have experienced the fastest change of technical possibilities we will see for a very long time. Only 30 years back, a mobile phone was kind of a rarity, which only the very technical early adopters had. Today, AI is taking over our repetitive work, rearranging our schedules, or just helping us get information faster than a classical search engine. Looking forward, this speed of change will not stop. While I do see us at a peak of the hype as of now, the hype will decrease and go into a more regular usage. The last year was about trying out, challenging the models, and, very importantly, also failing. We now need to take what we learned and start setting up reliable and reusable AI frameworks and usage patterns. This is the challenge we need to solve next year and ahead.
Iris Adae
VP of Data & Analytics, KNIME

The honeymoon phase with AI is ending. As the perceived "magic" fades, organizations will begin to approach AI adoption with the same rigor they apply to traditional IT platforms. Scrutiny will increase — not just on the technology itself, but on the vendor's narrative, governance model, and risk posture. Expect clear exit strategies and off-ramps, strong evaluation frameworks, and contingency planning to become standard practice. AI failures will be treated like any other system outage, with drills and failover plans embedded into IT operations. Break-glass moments won't be rare — they'll be anticipated. In short, AI will start to move from being a shiny disruptor to a managed, accountable part of the enterprise stack.
Dominick Profico
CTO, Bridgenext

Go to: 2026 AI Predictions - Part 2, covering solutions and strategies to support AI

Related Links

Hot Topics