And just 17 percent said they enjoy using AI and seek new ways to apply it, down from 21 percent in 2025. Incidentally, Accenture also noted signs of good old-fashioned FUD (fear, uncertainty and doubt). The IT professional services firm found that about half of workers feel secure in their jobs – down 11 percentage points from the 59 percent who said the same as recently as Summer 2025. About six in 10 workers also believe that young professionals are having a harder time finding jobs due to automation and AI. The percentage of workforces that said they “enjoy using AI and seek new ways to apply it” fell from 21 percent last summer to 17 percent this January. Overall, studies suggest AI implementations have focused primarily on lower-level, lower-risk activities such as answering questions, summarizing information and automating repeated tasks. AI agents currently operate at the edge of business processes rather than within core, mission-critical process flows and are heavily dependent on human oversight and approval for important decisions. There’s a growing understanding, however, that in order to justify deeper investment in AI, and achieve its massive potential, it must be better applied to strategic decision making and high value tasks and integrated deeper into business processes. And that will not be friction free. About a third of IT decision makers surveyed for process optimization company ABBYY reported struggles integrating GenAI into business practices, while training GenAI models was harder than expected. One root cause of such struggles is staff not having the skills necessary to train models and integrate data within business-critical business processes, Camunda executives argued. And just because business leaders say their staff is being properly trained, don’t automatically believe it. “While two-thirds of leaders (66 percent) cite skills training as a top investment priority, that investment is not consistently reaching the employees most exposed to rework,” reported the Workday study. “Among employees who use AI the most, only 37 percent report increased access to training – a nearly 30-point gap between stated intent and lived experience.” The issue is compounded by lagging role redesign. “(N)early nine in 10 organizations report that fewer than half of their roles have been updated to include AI-related skills,” Workday researchers continued. “AI has been layered onto roles that were never updated to accommodate it – forcing employees to use 2025 tools within 2015 job structures. Rather than reducing effort, this mismatch often increases it, as employees are left to reconcile faster production with unchanged expectations around accuracy, judgment and accountability.” For employees already doing a large share of rework, “outdated role definitions make it harder to capture AI’s benefits,” the study warned. “Without clear expectations for how AI should be used – and where human judgment must apply – employees default to verification and correction, absorbing the cost of low-quality output themselves.” At the same time, technology stacks are becoming more distributed and the number of endpoints involved in each process is growing, as a full 76 percent of organizations surveyed by Camunda said the volume and diversity of endpoints is increasing exponentially. “As a result, 85 percent say they Which, if any, of the following challenges have you encountered when implementing generative AI tools within your company? Training our Generative AI models was harder than we expected 31% Our staff did not have the necessary skills to deploy it 29% It was difficult to integrate within our business processes 28% We did not have proper governance or AI policies in place 26% Our staff misuse the tools 21% Paying for access was too expensive 19% It was not able to accurately incorporate our company data 19% Our staff ended up spending more time on tasks 18% It used too much energy 16% It experienced hallucinations 16% N/A. We have not encountered any challenges in particular 19% Source: Opinium Research; ABBYY The AI Adoption Gap Source: Gartner (October 2025) If an AI bubble were to bust, how would it impact your organization’s Investment strategy? Among C-suite repondents Source: Accenture Which, if any, of the following have you done to improve the outputs of Generative AI tools in your company? Source: ABBYY 9-Step PCI Compliance Checklist 7% 31% 16% 35% 10% Significantly decrease investments (20% or more) Significantly decrease investments (up to 20%) No changes Somewhat increase investments (up to 20%) Significantly increase investment (20% or more) Training staff to better use AI Encouraged teams to share tips and examples for effective use of AI Used AI agents to improve outputs Use process intelligence to improve analysis Used Document AI/intelligent document processing to improve the outputs Set up regular check-ins to review AI use Considering alternative AI tools Used Retrieval augmented generation (RAG) to improve the outputs Asked staff to manually check and correct the outputs Scaling back or removing AI tools that weren’t working We got rid of all AI tools and stopped using AI completely N/A. We have not done anything in particular to improve the outputs of AI tools 50% 43% 36% 35% 35% 33% 28% 25% 25% 20% 18% 2% AGI Neurosymbolic AI Ambient AI Agentic AI AI Adoption Gap Time AI Innovation Race (Providers) AI Outcome Race (Customers) AI Agents GenAI Traditional AI (ML, NLP) 18 CHANNELVISION | WINTER 2026
RkJQdWJsaXNoZXIy NTg4Njc=