State of enterprise AI adoption in 2026: current trends, barriers & best practices

November 19, 2026
AI consulting services & expertise

Current state of enterprise AI adoption

As AI technologies continue to evolve, their adoption is becoming increasingly widespread across organizations. Enterprises are gradually moving from experimenting with AI to putting AI solutions into actual use across the organization. However, many still struggle to scale AI and drive enterprise-wide transformation.

Although 88% of organizations use AI regularly in at least some business functions or departments, AI adoption has not yet reached an enterprise-wide scale. In fact, 86% of leaders express doubts about their organizations’ full readiness to integrate AI into their daily operations.

McKinsey

In 2025, 25% of organizations deployed over 40% of their AI experiment projects into production, with 54% of companies planning to achieve this level within 3-6 months.

Deloitte

40% of large organizations are scaling AI agents, up from 27% in 2025, while smaller organizations report the same level of adoption (22%).

McKinsey

Employee access to sanctioned AI tools has risen by 50% in organizations over the last year.

Deloitte

In the US, overall AI usage averaged from 17% to 20% across firms and reached 37% in companies with over 250 employees.

US Census Bureau

37% of organizations use AI primarily within existing workflows without substantially changing them, 30% redesign their key business processes around AI, while 34% fundamentally change their business model through AI.

Deloitte

AI adoption by technology type

Some of the AI technologies that have gained significant traction in recent years include generative AI (often referred to as GenAI), agentic AI, and physical AI. Since its introduction and rapid rise in 2022, generative AI has evolved into a strategic investment for enterprises. Generative AI chatbots and copilots, such as ChatGPT, Gemini, and Claude, help employees generate content based on user prompts and perform tasks like creating marketing materials, resolving customer issues, or automatically generating reports.

The growing generative AI adoption is also laying the groundwork for implementing agentic AI in organizations. AI agents can independently plan and execute multi-step tasks and workflows in situations that previously called for human judgment and coordination. However, adoption of agentic AI is still at an early stage as companies are exploring its capabilities.

Beyond AI operating primarily in digital environments, physical AI is bringing AI capabilities into the real world by combining intelligent software with sensors, robotics, autonomous vehicles, and other machines that can perceive their surroundings and take physical action.

In 2025, around 62% of organizations invested more in generative AI than in the previous year. Even among organizations where top managers are not particularly supportive of generative AI, 60% still increased their spending on it.

Capgemini

The share of organizations with partial or full-scale generative AI deployments in 2025 reached 36%, showing a 20% year-over-year growth. Among companies with relatively mature generative AI adoption, 30% utilize AI agents in their operations.

Capgemini

79% of senior executives say their companies have already adopted AI agents, while 88% plan to increase their spending on agentic AI over the next 12 months.

PwC

61% of executives say their organizations are actively adopting AI agents and preparing to move them past the experimentation phase to implementation at scale.

IBM

58% of organizations worldwide use physical AI in at least some limited use cases, and 80% plan to adopt the technology within the next two years.

Deloitte

Scheme title: Phase of organization’s use of AI tools, by organizations’ revenue size, % of respondents
Data source: McKinsey, The state of AI in 2026: On the road to ROI

Adoption across industries & business functions

Industries

AI has moved from a competitive differentiator to an operational standard in many industries. Despite strict compliance requirements, highly regulated sectors are among the most active AI adopters as the technology helps manage risks and make decisions more efficiently.

For example, financial services organizations utilize AI most often for fraud detection and wealth management, while healthcare and life sciences companies apply it for faster drug discovery and clinical research analysis. Manufacturing companies, in turn, use AI solutions to optimize supply chain management and enable predictive maintenance, and in education, teachers leverage AI to create learning materials and personalize assignments.

As of May 2026, the information (39.7%) and finance and insurance (33.9%) sectors show the highest AI use rates in the US, significantly exceeding the national average of 19.8%. In contrast, around 14% of retail businesses are currently using AI, and about 17% plan to adopt it in the next six months.

US Census Bureau

Consumer products (73%), insurance (70%), banking (67%), aerospace and defense (65%), and telecom (64%) are the sectors that increased their investment in generative AI the most over the previous year.

Capgemini

High tech (45%) leads among industries with the highest AI agent adoption, reflecting the need of high-tech companies for innovation and data-driven decision-making. Other leading sectors include industrial manufacturing (28%), consumer products (25%), energy and utilities (21%), and pharma and healthcare (19%).

Capgemini

65% of financial services institutions actively deployed or used AI tools in 2025, with capital markets firms (68%) being top adopters among the industry segments. The financial sector continues to invest in AI, with the spending expected to surpass $123 billion by 2028 at a CAGR of 29%.

Statista

Companies in the healthcare and life sciences industry show strong adoption of AI as evidenced by the highest usage of Python libraries devoted to natural language processing (69%). Given that healthcare companies generate 30% of the global data volume, AI tools are becoming essential for analyzing vast datasets and extracting insights.

Databricks

Manufacturing and automotive companies lead in using pre-trained AI foundation models and publicly available large language models (LLMs). In 2025, natural language processing (NLP) adoption in the industry grew by 148% compared to the previous year, the highest increase among all sectors analyzed.

Databricks

In response to customers’ transformed shopping habits, retailers and consumer packaged goods (CPG) companies apply AI across multiple business functions, including marketing, merchandising, replenishment, and innovation. 45% of retailers and 18% of CPGs are scaling their AI initiatives, with 75% of CPGs and 40% of retailers still in the phase of AI exploration.

BCG

AI usage in education spiked in 2025, with 86% of educational organizations reporting the use of generative AI across an array of use cases, from personalized K-12 learning to AI-assisted administrative operations. Among those who use AI daily for school-related purposes, 53% are education leaders, 36% are educators, and 30% are students.

Microsoft

Business functions

Organizations apply AI across a wide range of business functions to address specific operational tasks. For example, predictive models help optimize inventory management and enhance resilience in supply chain management. In customer service, AI tools route customer queries to the most suitable rep and assist with handling routine inquiries.

Within finance and accounting operations, AI agents can process invoices, assess regulatory compliance risks, and send payment reminders to customers. Similarly, AI-powered tools help HR departments source and evaluate candidates and provide insights into employee performance while IT departments use AI to detect technical issues and automate responses to security threats.

53% of enterprise leaders say that IT departments accelerate AI adoption the most at their companies.

Zapier

Among companies that are actively using or planning to use AI agents within six months, 57% will use them in customer service departments, 54% in sales and marketing, and 53% in IT and cybersecurity. Other areas of application include HR, finance and accounting, and product and service development.

PwC

Key focus areas for implementing AI agents include demand forecasting and planning (94%), customer support (91%), financial planning and analysis (89%), recruitment and hiring (84%), and employee performance management (74%).

Capgemini

Scheme title: AI agent use that has reached the scaling phase, by industry and business function, % of respondents
Data source: McKinsey, The state of AI in 2025: Agents, innovation, and transformation

Streamline your enterprise operations with AI
Contact us

AI business value & ROI

To measure the value AI delivers to business operations, organizations track various KPIs and metrics, including productivity, process efficiency, and cost savings. However, as the majority of enterprises still integrate AI across fragmented use cases, its impact often remains limited to improving the performance of individual employees and teams. As a result, companies can’t fully realize substantial business benefits and returns on AI investments at scale.

At the same time, there’s a correlation between the level of investment in AI and positive returns. According to EY, those organizations that allocate 5% or more of their total budget to AI see better outcomes than those who spend less than 5%.

The ROI on AI in business operations averages 1.7x in organizations at different stages of AI adoption, from pilots to scaled deployments across various business functions.

Capgemini

39% of organizations admit that AI use impacts their EBIT level, while less than 5% of their EBIT comes from AI use. At the same time, the share of high-performing companies that attribute at least 5% of EBIT to their use of AI stands at 6%.

McKinsey

48% of companies say AI helped them significantly improve productivity, while only 11% agree that they’ve achieved substantial enterprise-wide impact.

KPMG

44% of executives name improved employee experience and 42% name talent retention as major benefits from using agentic AI. In addition, 67% of leaders say agentic AI helps reduce operational costs through automation.

IBM

66% of companies that adopt agentic AI report increased productivity, 57% report cost savings, 55% report faster decision-making, and 54% report improved customer experience. About a third of companies also mention improved profitability (39%), improved cyber posture (37%), innovation in products and services (35%), and faster time-to-market (32%) as the value achieved.

PwC

AI is already delivering measurable cost reductions across individual business functions. In supply chain management, AI-driven route optimization and warehouse layout design can generate an average cost reduction of 23% during the fulfillment process.

In finance, audit compliance automation powered by generative or agentic AI reduced the cost of the record-to-analyze process by up to 25%.

The integration of AI into HR workflows, such as talent screening and résumé analysis, helps cut personnel and compensation costs by over 15%.

In customer service, AI capabilities like call summarization, chatbots, automated responses, early problem detection, and post-contact analytics help lower operational costs by up to 22% and labor costs by 20%.

Capgemini

Scheme title: Revenue increases within business units from AI use, past 12 months, by function, % of respondents
Data source: The state of AI in 2026: On the road to ROI

AI adoption barriers & risks

According to recent studies, enterprises struggle most often with high costs of AI solutions, security risks associated with deploying AI across the entire organization, shortage of AI-skilled workforce, absence of solid AI governance practices, and the complexity of integrating AI tools with existing IT infrastructures.

Vendor-related risks

The high cost of vendor AI solutions is the top constraint to using AI in organizations. 45% of leaders cite this as a barrier because they need to spend more on an AI tool to achieve more value and prove its ROI.

Zapier

20% of companies limit their AI usage due to AI-related operating costs, including token costs.

McKinsey

On top of that, 38% of enterprise leaders don't fully trust AI vendors to maintain security, with another 33% fearing AI solution vendor lock-in.

Zapier

Security

62% of companies report security as the key obstacle to scaling agentic AI, as they don't feel confident they'll be able to roll out autonomous systems safely at scale.

McKinsey

72% of respondents still see cybersecurity as a highly relevant risk, which increased by 6 percentage points since 2025. This risk remains foundational, since agentic AI and the autonomy it introduces can expand the potential threat surface.

McKinsey

37% of organizations indicate that data privacy and security are the top concerns when considering LLM providers.

Google

AI skill gaps

Enterprises find it challenging to redesign the operating models of teams and build AI capabilities quickly and at scale, partly due to a global shortage of talent with advanced AI skills, including in edge computing, big data, and machine learning. Currently, there's only half as many AI specialists with the expertise needed to meet the demand.

McKinsey

While many organizations face difficulties with integrating AI into the business due to insufficient worker skills, fewer than half of them adjust their talent strategies, focusing mostly (53%) on educating employees to build AI fluency rather than redesigning roles, the way work is organized, and the career paths around AI.

Deloitte

AI governance

Since agentic AI systems can now perform tasks across multiple applications, make decisions, and handle sensitive data with minimal human involvement, their growing autonomy raises new concerns around governance and accountability.

In 2025, organizations paid more attention to responsible use of AI, with AI-specific governance roles growing to 17% from 14% and the share of businesses without responsible AI policies dropping to 11% from 24% the previous year. Still, gaps in knowledge (59%), budget constraints (48%), and regulatory uncertainty (41%) constrain the implementation of AI governance across organizations.

Stanford University

Only 21% of companies have a mature model for the governance of autonomous AI agents, while 74% plan to adopt agentic AI in the next two years.

Deloitte

60% of organizations are predicted to fail to achieve the anticipated value from AI by 2027 due to incohesive ethical governance frameworks.

Gartner

Legacy infrastructures

78% of enterprise leaders say they're struggling to integrate AI tools with their existing systems, while AI integration with current infrastructures is one of the top barriers for 23% of companies.

Zapier

Lack of AI trust

74% of companies say AI inaccuracy, including inconsistency and robustness failures, is their top-of-mind AI risk, up 14 percentage points from 2025.

McKinsey

28% of organizations rank lack of trust in AI agents as a top-three barrier. Leaders' trust is lower for higher-stakes activities, including financial transactions (20%) and autonomous actions in employee interactions (22%), while the highest levels of trust are reported in data analysis (38%) and performance improvement (35%).

PwC

Scheme title: Top barriers preventing organizations from adopting AI at scale, % of respondents
Data source: McKinsey, The State of Organizations 2026 report

Factors to consider before AI scaling

Successful AI scaling requires organizations to look beyond the technological capabilities of AI solutions and factor in the complexities of production-scale operations, shifts in workforce trends, and technological independence.

Pilot vs production gap

Pilot AI projects typically have much simpler requirements than production deployments and run on curated data for shorter periods of time and within isolated environments. In contrast, production-scale AI systems require infrastructure investment, integration with multiple systems, and ongoing monitoring and maintenance, so companies need to allocate much more resources and devote more effort to coordination across teams, systems, and processes. Beyond that, moving a pilot AI project to real-world production can reveal problems that aren't visible during small-scale deployments, turning potential failures into business risks.

Workforce shifts

Despite the widespread fear that AI will replace humans and cause job losses, organizations expect AI-driven automation to augment human capabilities and transform existing roles rather than eliminate jobs. In fact, companies should expect that productivity gains from AI use will boost business growth, prompting them to hire more employees. In addition, as AI adoption scales, companies will need new AI roles, further expanding their workforce.

Sovereign AI trend

Many companies express at least moderate concerns about reliance on foreign-owned AI technologies and infrastructure, often factoring in the country where an AI solution is designed and deployed when selecting vendors and technology stacks. By choosing local vendors or building proprietary AI solutions, enterprises can reduce the dependence on foreign vendors, retain control over their data and infrastructure, and strengthen strategic resilience.

Pillars of successful enterprise-wide AI adoption

To transition from AI experimentation to scaled deployments and ensure company-wide value, enterprises need not just to embed more AI tools across scattered processes but also shift to an AI-enabled operating model. The experience of AI high-performing companies shows that building a cohesive AI strategy, redesigning workflows, retraining employees, and establishing a solid tech and data foundation helps embed AI more fully into the enterprise and maximize its business value.

Build an enterprise AI strategy

A strategy that aligns AI initiatives with the organization’s business goals, strengths, industry context, regulatory requirements, and digital maturity and focuses on high-impact business domains helps set a direction for enterprise-wide AI scaling. In addition, leadership buy-in and C-level’s active support helps accelerate AI transformation through clear business priorities, informed funding decisions, and dedicated change management efforts.

Modernize your tech infrastructure

A flexible, secure, and scalable technology foundation is essential for deploying AI systems and integrating them into business workflows across the entire enterprise, making infrastructure a strategic business capability rather than just an IT initiative. Forward-looking organizations opt for composable, vendor-agnostic AI architectures that connect multiple AI agents into a single system and adapt dynamically to business needs and regulatory changes.

Redesign workflows

In the recent survey by McKinsey, around three-quarters of high performers reported they fundamentally redesigned their workflows because of AI use. To integrate AI into core business processes and create end-to-end agentic workflows, organizations need to move from traditional siloed business functions to new AI-enabled operating models based on business outcomes. The new workflows should be designed around the hybrid capabilities of humans and AI, balancing the need for human oversight and AI autonomy.

Retrain workforce

As AI profoundly transforms existing employee roles, enterprises should retrain and upskill employees to build the technical and soft skills needed to work effectively alongside AI. Workforce strategies should prioritize building AI skills in individuals while preserving critical human skills, such as emotional intelligence, to prepare employees for new AI-assisted roles and ways of working.

Establish an AI governance framework

As AI becomes more autonomous and deeply integrated into critical workflows, enterprises should establish a formal AI governance program early in the adoption process to ensure safe AI scaling and trusted use. An effective governance framework should set standards for AI model lifecycle management, assign clear accountability for decisions and outcomes, define when human oversight is needed, and establish systematic monitoring of both human and AI agent performance to ensure accuracy and decision quality.

Develop a data strategy

With a unified, trusted data strategy in place, enterprises can provide AI systems with accessible, high-quality data, reducing the risks of fragmented data like operational blind spots or flawed decisions. To strengthen data foundations, companies should integrate data from diverse sources to ensure data availability and accessibility, enforce standards for data quality, and embed security and privacy controls to support responsible data use and safe AI scaling.

Adopt AI with a trusted technology partner
Turn to Itransition

Real-life success stories of enterprise AI adoption

Take a look at the featured case studies from Itransition’s portfolio where we helped enterprises across industries adopt AI technologies to streamline operations and improve business performance.

8%

increase in visitor-to-buyer conversion rate

15%

more data collected

50%

lower monthly infrastructure costs
AI-powered BI platform for a fashion retailer

Itransition delivered a BI solution to help the ecommerce company gather and analyze data in near-real-time for further customer experience personalization. The solution features an AI-powered recommendation engine that suggests products based on users’ purchases, views, and clicks, as well as computer vision capabilities for automated product image recognition.

99.1%

defect detection accuracy

60%

reduction in equipment run-in time

Up to 85%

decrease in equipment configuration effort
AI-driven quality control enablement

We delivered a scalable, production-grade AI solution for a test and measurement equipment manufacturer serving the food and beverage industry, helping them enhance their test equipment offering. The new solution measures dimensions for standard and non-standard caffeine pouches, informs employees whether defects are present, and spotlights the exact defect location.

100%

of products validated automatically

2.6 p.p.

increase in defects detected before shipment

19%

annual revenue increase
AI solution for medical packaging defect detection

We implemented a computer vision solution for a medical storage container manufacturer, enabling them to achieve 100% product quality inspection without increasing the amount of manual labor. The solution analyzes photos of medical container surfaces, segments coating areas based on coating quality, decides whether the product is defective or not, and sends alerts to the conveyor system to remove defective containers.

75%

model accuracy up from 60%

Faster

decision-making

Lower

costs on computing power resources
Custom ML algorithms for an insurance platform

Our team helped Integrated Specialty Coverages revamp the machine learning model in their existing AI underwriting solution to help them make more accurate predictions of customer intentions. We developed a new ML model that makes decisions based on an optimal set of features, predicting the likelihood of users completing the application to buy an insurance policy.

50%

faster image processing

Streamlined

solution management
ML solution for brand analytics & reporting

Itransition delivered an ML-powered solution for a global event sponsorship valuation company to identify brands in sports-related images and help brands assess the effectiveness of their marketing campaigns. Powered by several ML and non-ML models, the solution recognizes brand logos in media, determines their exact location, and, based on the results, generates different types of bespoke reports with statistics and calculations.

Itransition helps enterprises smoothly incorporate AI solutions into their technical ecosystems and business workflows and get the most out of AI investments, providing both advisory support and end-to-end development services.

AI consulting
AI consulting

We provide expert advice to help you address possible challenges of AI adoption and ensure the AI solution is fully aligned with your business needs. Our consultants guide you through every step, from AI readiness assessment and business case definition to solution conceptualization, implementation, and ongoing optimization.

AI development

Our team delivers AI solutions tailored to your unique workflows, business specifics, and industry standards. We take care of data preprocessing, AI model training, backend and frontend development, solution integration with other systems, and post-launch support.

About Itransition

Providing IT consulting and software engineering services since 1998

5+ years in AI development and consulting

Internal AI/ML Center of Excellence

Holding a Microsoft Azure AI Platform specialization

Long-standing partnerships with Microsoft and AWS

Delivering AI solutions compliant with HIPAA, GDPR, and other relevant standards

ISO 9001 and ISO 27001-compliant quality and information security management practices

Awards and recognition from Forrester, Everest Group, Gartner, and Zinnov Zones