Workers using AI tools are saving an hour a day on average, according to Adecco's 2025 Global Workforce of the Future survey. One in five reports saving two hours daily. Across a workforce, that's five to ten hours a week of productive capacity per person, unlocked.
Three out of four of those workers were never trained to do it.
This isn't a story about AI capability. The tools work. The savings are real. This is a story about organizational intent — about what it means that the single most significant productivity gain in a generation is happening by accident inside most companies.
The Accidental Productivity Boom
The number comes from Adecco's recent workforce research: employees across industries are independently discovering, experimenting with, and deploying AI tools — and generating measurable productivity gains. The one-hour average isn't a projection; the top quintile saves twice that. Either way, it's what workers are actually reporting.
But 75% of those workers didn't get a training program, a policy, a use-case guide, or even a single meeting where their employer acknowledged that AI tools exist. They figured it out themselves, on company time, with company goals, producing company results.
Which means most companies are benefiting from AI by accident. Their employees are getting smarter about it. Their workflows are shifting. And their leadership doesn't know where the gains are coming from or how to scale them.
What This Creates
The accidental approach has a ceiling. Informal AI use produces inconsistent outcomes — the person in accounting who discovered the right prompts produces twice the output of the person in the next office who's still working the old way. Same job, same tools, different results, no system for the gap.
It also creates a talent sorting mechanism nobody asked for. The employees who are most aggressively self-educating on AI tools are the ones getting calls from companies that have figured out how to systematize this. They have skills that are worth more than their current employer knows. And the moment a company offers them a role where those skills are recognized, valued, and built upon, the calculus changes.
This is how you lose your best people to a competitor with a better AI training program. Not to a competitor with better pay or better culture — to a competitor that simply knows what their employees are capable of.
The Employer Half of the Equation
Here's what a systematic approach looks like, and it's not complicated:
- Define which AI tools are approved for which use cases
- Build a shared prompt library for your most common workflows
- Give people structured time to experiment and document what works
- Create a simple feedback loop — what's working, what failed, what to standardize
That's it. That's the difference between accidental AI adoption and intentional AI adoption. It doesn't require an AI strategy team. It requires someone to own the question and a few hours to build the initial structure.
The companies doing this are documenting the gains, scaling them across teams, and building a compounding productivity advantage. The ones not doing it are getting the gains unevenly and burning through the people responsible for them.
The Hiring Implication
The talent market is developing a new category: the AI-fluent operator. Not an AI engineer. Not a data scientist. Someone who runs a normal business function — finance, operations, HR, marketing, supply chain — but has built genuine fluency in applying AI to that work.
These people are harder to find than AI specialists because nobody knows what the job posting looks like. "Proficient in Excel and AI tools" doesn't capture it. And the interview processes that identify true AI fluency don't exist yet at most companies.
The ones who can find them first — and create the environment where they stay — are building a talent moat that compounds every quarter as AI capabilities expand.
The 75% who were never trained? Some percentage of them have become deeply fluent anyway. Those are the ones worth finding.
VC5 Consulting works with companies building AI-augmented teams — sourcing the operators, engineers, and specialists who are actually fluent in this transition. If you're trying to find people who know the difference between using AI and using it well, let's start there.