What companies can really expect from artificial intelligence in 2025 and beyond
Reading time: 5 minutes | Keywords: artificial intelligence, AI adoption, generative AI, agentic AI, ROI, productivity, digital transformation
Artificial intelligence has gone from science fiction to business infrastructure in record time. In 2025, 88% of organizations report using AI regularly, and enterprise spending on generative AI reached $37 billion a 3.2x increase over the previous year.
But behind these impressive numbers lies a more complex reality: between 70% and 85% of AI initiatives fail to meet expectations. 42% of companies abandoned the majority of their AI initiatives in 2025, up from 17% in 2024.
How do we separate real opportunities from inflated hype? That’s the question every business leader must ask before investing.
AI adoption has accelerated dramatically. According to the latest McKinsey Global Survey, 78% of organizations use AI in at least one business function, compared to 55% just one year earlier.
But there are important nuances:
Wide but shallow usage: Most organizations have yet to integrate AI deeply enough into their workflows to achieve material enterprise-level benefits.
Maturity gap: Only 6% of organizations qualify as "AI high performers" with a 5%+ impact on EBIT. Just 26% have the capabilities to move from proof of concept to production.
Functional concentration: IT, marketing, and sales lead adoption, followed by service operations. The sharpest growth was in IT, where usage jumped from 27% to 36% in just six months.
Among employees, 56% of American workers now use generative AI tools for work tasks. Leaders use AI at 33% double that of individual contributors (16%).
The productivity promises of AI are enormous. But what do the actual data show?
Proven ROI: Companies that adopted generative AI early report $3.70 in value for every dollar invested, with top performers reaching returns of $10.30 per dollar.
Sector productivity: Industries that have adopted AI see their labor productivity grow 4.8 times faster than the global average. Sectors with high AI exposure show 3x greater revenue-per-worker growth.
Impact on code: Code has become the first real "killer use case" for AI. 50% of developers now use AI tools for coding daily (65% in top organizations). Teams report speed improvements of 15% or more.
However, most organizations achieve satisfactory ROI in 2–4 years much longer than the typical 7–12 month technology payback period.
AI creates material business value when it is embedded into core workflows, not merely adopted across functions.
The generative AI market is projected to reach $59 billion in 2025 and grow to $400 billion by 2031, with a compound annual growth rate of 37.57%.
McKinsey estimates that generative AI could unlock between $2.6 and $4.4 trillion in additional value, beyond what traditional analytical AI already contributes.
Key use cases:
AI agents—systems based on foundation models capable of acting in the real world, planning, and executing multiple steps—represent the next frontier.
23% of organizations report scaling agentic AI systems in some part of their business, with an additional 39% experimenting with agents. However, usage isn’t yet widespread: in no function does more than 10% report scaling agents.
Job postings mentioning agentic AI grew 985% between 2023 and 2024, signaling explosive corporate interest.
Agent usage is most common in IT and knowledge management, where cases like service desk management and deep research have developed rapidly.
“2025 will mark a significant milestone in AI agent adoption across industries like finance, supply chain, sales, services, marketing, and tax.”
— Igor Epshteyn, CEO of Coherent Solutions
Despite the enthusiasm, the risks are substantial:
Lack of trust: Trust in AI companies fell from 61% to 53% globally in 2024. In the U.S. specifically, trust dropped 15 points from 50% to 35%. 77% of Americans don’t trust companies to use AI responsibly.
Job concerns: 75% of Americans believe AI will reduce the total number of jobs in the U.S. over the next 10 years. However, the World Economic Forum projects a net gain of 12 million jobs.
ROI challenges: 66% of companies struggle to establish ROI metrics for AI initiatives. 74% of companies have difficulty achieving and scaling value, according to BCG.
Infrastructure costs: Data center electricity consumption in the U.S. reached 183 terawatt-hours in 2024 and is projected to reach 426 TWh by 2030. Training Google’s Gemini Ultra cost $191 million.
AI-related incidents are increasing sharply, but standardized responsible AI assessments remain rare among major model developers.
Based on the success patterns identified by McKinsey, high-performing organizations share distinctive characteristics:
Committed leadership: They are 3 times more likely to have senior leaders who demonstrate active ownership and commitment, including modeling AI usage themselves.
Defined processes: They have established processes for determining how and when model outputs need human validation to ensure accuracy.
Buy vs. build: 76% of AI use cases are now purchased rather than built internally, compared to 53% in 2024. Ready-made solutions reach production faster.
Our recommendation at Strolling Digital: start with proven use cases (customer service, data analysis, content generation), establish clear success metrics before implementing, invest in user training, and scale gradually based on measurable results.
AI is not a strategy; it’s a strategy accelerator. Organizations that treat it as an isolated technology project fail. Those that integrate it into their broader transformation vision thrive.
92% of firms plan to increase their AI budgets over the next three years, according to McKinsey. The question is no longer whether to adopt AI, but how to do so in a way that generates real, sustainable value.
“69% of leaders believe AI demands a complete rethinking of how their systems and processes are built and managed.”
It’s not about automating what exists; it’s about reimagining what’s possible.
Ready to turn AI ambition into measurable business impact?
If you want to define pragmatic AI use cases, build adoption-ready foundations, and scale value responsibly, get in touch with Strolling Digital. Let’s talk.