2026 AI Predictions - Part 3

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. Part 3 covers barriers and challenges for AI.

THE AI BUBBLE

Navigating the AI Bubble Requires Discipline and Customer Focus: 2026 will test every infrastructure company's ability to stay grounded amid relentless AI hype. The cycle is self-reinforcing: capital pressures vendors to tell AI stories, vendors amplify the narrative, and the bubble grows. Marketing has raced several cycles ahead of where the technology actually delivers value. The hardest challenge for infrastructure startups isn't technical — it's maintaining strategic discipline when everyone around you is chasing the same shiny objects. The path forward requires customer-driven development: listening to the actual needs of enterprises as they transform their businesses for AI, rather than building features that sound good in pitch decks. The companies that win in 2026 won't be those with the most AI buzzwords in their positioning — they'll be the ones solving real problems with reliable, cost-effective infrastructure. When the hype cycle inevitably corrects, the survivors will be those who built for genuine customer needs rather than investor narratives.
Sijie Guo
CEO, StreamNative

THE TRUE COST OF AI

The Free Ride for AI Will End: AI's true cost will catch up with all of us. Currently, models are priced significantly below what it costs to run them, with companies burning investor capital to obtain market share. This can't go on forever. Just as happened with the cloud, vendors will eventually have to start charging based on usage. When that happens, teams will understand what they're paying for, budgets will be optimized, and a lot of smaller AI providers will either go away or be acquired.
Tanner Burson
CTO, Prismatic

LACK OF ROI

True AI Adoption (and ROI) Hinges on Reducing Friction: In 2026, I expect companies will turn their attention away from AI capabilities towards AI implementation. Models will continue to get smarter, but that won't solve the ROI problem organizations are facing with their AI deployments. The reality is that most tech requires significant upfront investment with little payoff. People don't want to change how they operate or spend days building a custom AI workflow just to save a few minutes. Even if it's beneficial in the long run, it feels like a bad trade. In 2026, the vendors that get implementation time closer to zero will see higher adoption rates.
Jake Stauch
CEO and Co-Founder, Serval

GenAI ROI Debate Will Not Go Away … But Hold Your Nerve: The second half of 2025 saw a backlash against claims of AI's impact. Studies and commentators reported that enterprises were seeing little, if any, ROI from generative AI implementations. So, as we enter 2026, we push through a veil of disappointment, if not disenchantment, with AI. This is a natural reaction to the pumped-up hype and noise. Loud, exaggerated claims won't accelerate adoption in 2026. Yet 2026 will be a transition year. The AI projects that matter are those delivering transformative results, major enterprise programs where AI is woven into processes to improve service outcomes and operational efficiency. These initiatives don't wrap up in a few months; they require 12–18-month timelines for full-scale rollout and adoption. They're not about quick fixes promoted by some vendors but about ensuring foundational work around data, processes, and change management is thoroughly done and dusted.
Peter van der Putten
Director, AI Lab, Pega and Assistant Professor of AI, Leiden University

After billions wasted on ChatGPT wrappers and vaporware, CFOs are demanding real ROI — and most generative AI projects can't deliver. The honeymoon phase where "AI innovation" justified any budget is over, replaced by brutal questions about cost per query, accuracy rates and measurable business outcomes. Companies that can't show concrete savings, revenue growth or productivity gains within six to 12 months will see their AI initiatives shelved, or their vendors replaced.
Manisha Khanna
Senior Product Manager, AI & Generative AI, SAS

AI SPENDING FREEZE

CFO Oversight Amps Up the Pressure on AI Investments: We are officially exiting the experimental phase of AI and entering an "era of extreme accountability" for tech investments. By the end of 2026, the single most critical metric for C-level executives won't be innovation or speed, but operating costs. With millions already approved for AI projects, investors and CFOs will demand a freeze on additional spending, forcing IT leaders to justify their budgets solely by how well they maximize existing resources. The question from the board is no longer "What can AI do?" but "How will this deliver ROI?"
Ryan Manning
Chief Product Officer, BMC Helix

The Great AI ROI Reckoning: 2026 is when the music stops. CFOs are done writing blank checks for "AI innovation" that can't be tied to actual business results. We're already seeing enterprises start to pump the brakes on a significant percentage of their planned AI spending because leadership finally asked the obvious question: "What are we actually getting for this?" And most teams have no good answer from a year of PoCs that never made it into production. The handful of use cases that actually move numbers will survive. Revenue up, costs down, cycle time reduced … real KPIs that matter. Everything else gets killed. No more pilot purgatory, no more "let's experiment and see," no more demos that wow executives but never ship. If you can't show business impact in three to six months, you're done. The companies winning in late 2026 are the ones who got religious about measurement early and weren't afraid to kill their darlings.
Jarrod Vawdrey
Field Chief Data Scientist, Domino Data Lab

By the end of 2026, 40% of enterprise AI projects initiated in 2024-2025 will be defunded for failing to demonstrate ROI, forcing a brutal reckoning in AI budgets. Organizations that experimented with AI in the last two years will begin to demand measurable returns in 2026. The "let's try AI" era ends as investors and CFOs require clear ROI timelines and sharpen their focus on margin impact. Labs pitching vague productivity gains will lose to competitors demonstrating specific cost reductions or revenue increases. AI budgets shift from innovation teams to line-of-business owners who need to justify every dollar. Projects that can't prove ROI within a defined timeframe will be at risk, forcing organizations to build economic models into their deployment plans.
Shimon Ben-David
CTO, WEKA

AI SYSTEM FAILURES

By the end of 2026, we will see at least three Fortune 500 CEOs lose their roles explicitly due to AI system failures that their organizations cannot explain, reproduce or defend post-incident. Unlike past outages tied to infra or human error, these failures will stem from opaque AI decision paths, missing training lineage and irreproducible agent behavior. Boards and regulators will treat "we don't know why the model did that" as an unacceptable operational answer, especially in regulated or revenue-critical environments. The root issue will not be that AI failed, but that organizations lacked systems to reason about AI behavior at runtime and after the fact.
Sameer Agarwal
CTO, Deductive AI

LOSS OF CONFIDENCE

Agentic AI Won't Go Beyond Basic IT Tasks in 2026: In 2026, there's going to be a massive gap between what vendors market and what IT leaders actually allow AI agents to do. Vendors are talking about agents that add memory to a virtual machine or remediate infrastructure autonomously, but experienced employees handle those kinds of tasks for good reason. Rushing into complex, autonomous remediation will create expensive failures that make stakeholders lose confidence and set AI implementations back by months.
Phil Christianson
Chief Product Officer, Xurrent

LACK OF TRUST

Agentic AI Isn't Ready to Run the Show, and Caution Will Prevail: Even as agentic AI dominates headlines and many companies begin to leverage its basic use cases, businesses will continue to exercise caution in 2026 — leading to a slow rate of widespread adoption and increased human oversight. While the technology remains a common source of excitement across many different industries, it's important to remember that agentic AI has only existed meaningfully for about two years—hardly enough time to establish the trust needed to turn mission-critical tasks over to autonomous AI agents. As organizations grapple with trust, governance, and security in 2026, agentic AI adoption will continue to mature, but it won't achieve the sky-high adoption rates some have predicted.
Matt Kunkel
CEO and Co-Founder, LogicGate

AI FATIGUE

"Checkbox AI" Fatigue Sets In: After two years of forcing AI into every feature for marketing appeal, businesses will pull back in 2026. Adding AI simply to "tick the box" often results in longer development cycles, slower QA, and lower-quality products. The next phase of AI maturity will prioritize meaningful, use-case-specific integration, deploying AI only where it adds measurable value or unlocks new capabilities. 
Rob Mason
CTO, Applause

INSUFFICIENT AI-READY DATA

According to IDC, most AI pilots from the past year never made it into production. Here's the real problem: companies are sitting on mountains of valuable data, but have no way to make it AI-ready without creating new risks. 88% of AI pilots fail due to insufficient AI-ready data, and by 2027, 40% of AI projects will be canceled due to escalating costs. The deeper issue? AI hallucinations. When AI gives you different answers each time or can't explain where it got its information from, that's unacceptable in production environments. Leaders can't defend decisions they can't trust or verify. Today, data teams spend more time cleaning and managing permissions than delivering AI applications that drive business value. In 2026, the AI winners won't be the companies with the biggest models. Success will be defined by who cracked the code on giving AI systems governed, contextual access to enterprise data without hallucinations. That's how data becomes a competitive advantage instead of digital dust. 
Matt Belkin
President and GM, Data + Analytics, insightsoftware

DATA ARCHITECTURE

Over the next year, companies will discover that their biggest AI barrier is data architecture. Business data lives across countless disconnected systems, and this distributed information represents the critical context that AI needs to be truly effective, yet often remains largely inaccessible. AI is only as intelligent as the context it can access, and today's fragmented data landscape is the primary barrier preventing companies from unlocking AI's full potential. Organizations that prioritize building unified data and context frameworks will gain faster AI deployments, reduced security risk, and the ability to leverage organizational knowledge across their entire technology stack. Companies that solve for context architecture first will see their AI investments compound in value. Meanwhile, those taking ad hoc approaches will find diminishing returns, as context fragmentation slows AI adoption and innovation.
Michelle Gill
Sr. Director of Engineering, GitLab

THE ECHO CHAMBER

A return to revisit the basics to solving the Echo Chamber Problem in AI: The "AI echo chamber" problem happens when humans start to experiment on hypotheses to prove (or disprove) them, and write up these expectations in detail. The final results of these experiments may not always be cataloged in as much detail, or may be hard to correlate with the experiments. When LLMs start picking up this intermediate data, they will get more influenced by the hypothesis (as opposed to the actual results), which lead to poor or erroneous recommendations — defeating the core strength of LLMs. In 2026, I expect the tech industry to revisit these basics as it prioritizes solving the problem of "AI echo chambers." We will see more investment in, and a deeper examination of, the inner workings of a variety of topics, including input data quality, embellishing inputs with relevant data, and how vector store architectures can support GenAI applications that need rapid iterations and scaling
Karthik Ranganathan
CEO and Co-Founder, Yugabyte

CHANGE MANAGEMENT

Change management will become the primary barrier to AI adoption, not technology — organizations claiming "tech-first" strategies will continue operating "people-first."
Ritu Dubey
Market Head, Digitate

Go to: 2026 AI Predictions - Part 4, covering negative impacts of AI

Related Links

Hot Topics

2026 AI Predictions - Part 3

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. Part 3 covers barriers and challenges for AI.

THE AI BUBBLE

Navigating the AI Bubble Requires Discipline and Customer Focus: 2026 will test every infrastructure company's ability to stay grounded amid relentless AI hype. The cycle is self-reinforcing: capital pressures vendors to tell AI stories, vendors amplify the narrative, and the bubble grows. Marketing has raced several cycles ahead of where the technology actually delivers value. The hardest challenge for infrastructure startups isn't technical — it's maintaining strategic discipline when everyone around you is chasing the same shiny objects. The path forward requires customer-driven development: listening to the actual needs of enterprises as they transform their businesses for AI, rather than building features that sound good in pitch decks. The companies that win in 2026 won't be those with the most AI buzzwords in their positioning — they'll be the ones solving real problems with reliable, cost-effective infrastructure. When the hype cycle inevitably corrects, the survivors will be those who built for genuine customer needs rather than investor narratives.
Sijie Guo
CEO, StreamNative

THE TRUE COST OF AI

The Free Ride for AI Will End: AI's true cost will catch up with all of us. Currently, models are priced significantly below what it costs to run them, with companies burning investor capital to obtain market share. This can't go on forever. Just as happened with the cloud, vendors will eventually have to start charging based on usage. When that happens, teams will understand what they're paying for, budgets will be optimized, and a lot of smaller AI providers will either go away or be acquired.
Tanner Burson
CTO, Prismatic

LACK OF ROI

True AI Adoption (and ROI) Hinges on Reducing Friction: In 2026, I expect companies will turn their attention away from AI capabilities towards AI implementation. Models will continue to get smarter, but that won't solve the ROI problem organizations are facing with their AI deployments. The reality is that most tech requires significant upfront investment with little payoff. People don't want to change how they operate or spend days building a custom AI workflow just to save a few minutes. Even if it's beneficial in the long run, it feels like a bad trade. In 2026, the vendors that get implementation time closer to zero will see higher adoption rates.
Jake Stauch
CEO and Co-Founder, Serval

GenAI ROI Debate Will Not Go Away … But Hold Your Nerve: The second half of 2025 saw a backlash against claims of AI's impact. Studies and commentators reported that enterprises were seeing little, if any, ROI from generative AI implementations. So, as we enter 2026, we push through a veil of disappointment, if not disenchantment, with AI. This is a natural reaction to the pumped-up hype and noise. Loud, exaggerated claims won't accelerate adoption in 2026. Yet 2026 will be a transition year. The AI projects that matter are those delivering transformative results, major enterprise programs where AI is woven into processes to improve service outcomes and operational efficiency. These initiatives don't wrap up in a few months; they require 12–18-month timelines for full-scale rollout and adoption. They're not about quick fixes promoted by some vendors but about ensuring foundational work around data, processes, and change management is thoroughly done and dusted.
Peter van der Putten
Director, AI Lab, Pega and Assistant Professor of AI, Leiden University

After billions wasted on ChatGPT wrappers and vaporware, CFOs are demanding real ROI — and most generative AI projects can't deliver. The honeymoon phase where "AI innovation" justified any budget is over, replaced by brutal questions about cost per query, accuracy rates and measurable business outcomes. Companies that can't show concrete savings, revenue growth or productivity gains within six to 12 months will see their AI initiatives shelved, or their vendors replaced.
Manisha Khanna
Senior Product Manager, AI & Generative AI, SAS

AI SPENDING FREEZE

CFO Oversight Amps Up the Pressure on AI Investments: We are officially exiting the experimental phase of AI and entering an "era of extreme accountability" for tech investments. By the end of 2026, the single most critical metric for C-level executives won't be innovation or speed, but operating costs. With millions already approved for AI projects, investors and CFOs will demand a freeze on additional spending, forcing IT leaders to justify their budgets solely by how well they maximize existing resources. The question from the board is no longer "What can AI do?" but "How will this deliver ROI?"
Ryan Manning
Chief Product Officer, BMC Helix

The Great AI ROI Reckoning: 2026 is when the music stops. CFOs are done writing blank checks for "AI innovation" that can't be tied to actual business results. We're already seeing enterprises start to pump the brakes on a significant percentage of their planned AI spending because leadership finally asked the obvious question: "What are we actually getting for this?" And most teams have no good answer from a year of PoCs that never made it into production. The handful of use cases that actually move numbers will survive. Revenue up, costs down, cycle time reduced … real KPIs that matter. Everything else gets killed. No more pilot purgatory, no more "let's experiment and see," no more demos that wow executives but never ship. If you can't show business impact in three to six months, you're done. The companies winning in late 2026 are the ones who got religious about measurement early and weren't afraid to kill their darlings.
Jarrod Vawdrey
Field Chief Data Scientist, Domino Data Lab

By the end of 2026, 40% of enterprise AI projects initiated in 2024-2025 will be defunded for failing to demonstrate ROI, forcing a brutal reckoning in AI budgets. Organizations that experimented with AI in the last two years will begin to demand measurable returns in 2026. The "let's try AI" era ends as investors and CFOs require clear ROI timelines and sharpen their focus on margin impact. Labs pitching vague productivity gains will lose to competitors demonstrating specific cost reductions or revenue increases. AI budgets shift from innovation teams to line-of-business owners who need to justify every dollar. Projects that can't prove ROI within a defined timeframe will be at risk, forcing organizations to build economic models into their deployment plans.
Shimon Ben-David
CTO, WEKA

AI SYSTEM FAILURES

By the end of 2026, we will see at least three Fortune 500 CEOs lose their roles explicitly due to AI system failures that their organizations cannot explain, reproduce or defend post-incident. Unlike past outages tied to infra or human error, these failures will stem from opaque AI decision paths, missing training lineage and irreproducible agent behavior. Boards and regulators will treat "we don't know why the model did that" as an unacceptable operational answer, especially in regulated or revenue-critical environments. The root issue will not be that AI failed, but that organizations lacked systems to reason about AI behavior at runtime and after the fact.
Sameer Agarwal
CTO, Deductive AI

LOSS OF CONFIDENCE

Agentic AI Won't Go Beyond Basic IT Tasks in 2026: In 2026, there's going to be a massive gap between what vendors market and what IT leaders actually allow AI agents to do. Vendors are talking about agents that add memory to a virtual machine or remediate infrastructure autonomously, but experienced employees handle those kinds of tasks for good reason. Rushing into complex, autonomous remediation will create expensive failures that make stakeholders lose confidence and set AI implementations back by months.
Phil Christianson
Chief Product Officer, Xurrent

LACK OF TRUST

Agentic AI Isn't Ready to Run the Show, and Caution Will Prevail: Even as agentic AI dominates headlines and many companies begin to leverage its basic use cases, businesses will continue to exercise caution in 2026 — leading to a slow rate of widespread adoption and increased human oversight. While the technology remains a common source of excitement across many different industries, it's important to remember that agentic AI has only existed meaningfully for about two years—hardly enough time to establish the trust needed to turn mission-critical tasks over to autonomous AI agents. As organizations grapple with trust, governance, and security in 2026, agentic AI adoption will continue to mature, but it won't achieve the sky-high adoption rates some have predicted.
Matt Kunkel
CEO and Co-Founder, LogicGate

AI FATIGUE

"Checkbox AI" Fatigue Sets In: After two years of forcing AI into every feature for marketing appeal, businesses will pull back in 2026. Adding AI simply to "tick the box" often results in longer development cycles, slower QA, and lower-quality products. The next phase of AI maturity will prioritize meaningful, use-case-specific integration, deploying AI only where it adds measurable value or unlocks new capabilities. 
Rob Mason
CTO, Applause

INSUFFICIENT AI-READY DATA

According to IDC, most AI pilots from the past year never made it into production. Here's the real problem: companies are sitting on mountains of valuable data, but have no way to make it AI-ready without creating new risks. 88% of AI pilots fail due to insufficient AI-ready data, and by 2027, 40% of AI projects will be canceled due to escalating costs. The deeper issue? AI hallucinations. When AI gives you different answers each time or can't explain where it got its information from, that's unacceptable in production environments. Leaders can't defend decisions they can't trust or verify. Today, data teams spend more time cleaning and managing permissions than delivering AI applications that drive business value. In 2026, the AI winners won't be the companies with the biggest models. Success will be defined by who cracked the code on giving AI systems governed, contextual access to enterprise data without hallucinations. That's how data becomes a competitive advantage instead of digital dust. 
Matt Belkin
President and GM, Data + Analytics, insightsoftware

DATA ARCHITECTURE

Over the next year, companies will discover that their biggest AI barrier is data architecture. Business data lives across countless disconnected systems, and this distributed information represents the critical context that AI needs to be truly effective, yet often remains largely inaccessible. AI is only as intelligent as the context it can access, and today's fragmented data landscape is the primary barrier preventing companies from unlocking AI's full potential. Organizations that prioritize building unified data and context frameworks will gain faster AI deployments, reduced security risk, and the ability to leverage organizational knowledge across their entire technology stack. Companies that solve for context architecture first will see their AI investments compound in value. Meanwhile, those taking ad hoc approaches will find diminishing returns, as context fragmentation slows AI adoption and innovation.
Michelle Gill
Sr. Director of Engineering, GitLab

THE ECHO CHAMBER

A return to revisit the basics to solving the Echo Chamber Problem in AI: The "AI echo chamber" problem happens when humans start to experiment on hypotheses to prove (or disprove) them, and write up these expectations in detail. The final results of these experiments may not always be cataloged in as much detail, or may be hard to correlate with the experiments. When LLMs start picking up this intermediate data, they will get more influenced by the hypothesis (as opposed to the actual results), which lead to poor or erroneous recommendations — defeating the core strength of LLMs. In 2026, I expect the tech industry to revisit these basics as it prioritizes solving the problem of "AI echo chambers." We will see more investment in, and a deeper examination of, the inner workings of a variety of topics, including input data quality, embellishing inputs with relevant data, and how vector store architectures can support GenAI applications that need rapid iterations and scaling
Karthik Ranganathan
CEO and Co-Founder, Yugabyte

CHANGE MANAGEMENT

Change management will become the primary barrier to AI adoption, not technology — organizations claiming "tech-first" strategies will continue operating "people-first."
Ritu Dubey
Market Head, Digitate

Go to: 2026 AI Predictions - Part 4, covering negative impacts of AI

Related Links

Hot Topics