Skip to main content Skip to footer
  • "com.cts.aem.core.models.NavigationItem@6b20d332" Careers
  • "com.cts.aem.core.models.NavigationItem@2a48a446" News
  • "com.cts.aem.core.models.NavigationItem@570f19ab" Events
  • "com.cts.aem.core.models.NavigationItem@65200914" Investors

New frontiers,

new frictions

Mapping the DNA of AI adoption

Our global study of 10,000-plus workers reveals the hidden segments in the workforce—and $6 trillion in trapped productivity—behind AI's adoption gap.

Download full report

New frontiers,

new frictions

Mapping the DNA of AI adoption

Our global study of 10,000-plus workers reveals the hidden segments in the workforce—and $6 trillion in trapped productivity—behind AI's adoption gap.

Download full report
Enterprises have invested heavily in AI. Yet at a macroeconomic level, productivity data remains flat. Our latest research, which draws on a survey of over 10,000 workers across more than 20 key geographies and every major industry, reveals why: The challenge is not with the technology; it’s with the people who need to use it. If a high level of friction stops people from using AI, its economic impact will never arrive.

That friction, it turns out, is both widespread and consistent. Our research reveals that most of the workforce wants to use AI far more than current operational realities allow. Additionally, most are unhappy with the AI tools their organization makes available.

Further, while enterprises tend to roll out AI as if they are working with a single, generic employee, AI adoption varies greatly among different segments of the workforce. A one-size-fits-all deployment can be a key cause of friction, itself. 

As part of our research series on closing the gap between AI's technical capabilities and achieving real-world results, we explored the role that workers themselves play in either speeding or slowing adoption. 

By asking respondents detailed questions about their overall disposition toward technology, work and AI, we identified over 60 personas in the workforce today. These coalesced into seven archetypes, each with a distinctively different level of AI engagement and experience—and each with its own need for a specific intervention to boost AI adoption. 

Across them all, several systemic AI adoption challenges emerged, each pointing to a specific, solvable problem that is currently suppressing trillions of dollars of potential value.

Employees want to adopt AI but aren’t equipped to use it effectively

80%

of employees record a positive Activation Gap, a measure of the gap between a worker’s AI aspirations and their actual use of it.

This represents a huge amount of untapped potential from employees who want to adopt AI but are unable to do so because of a wide range of factors, such as skills shortages, low access or concerns about outputs.

Up to 82%

of employees are unsatisfied with the AI tools they’re given to use.

Across all archetypes, the vast majority of employees believe their current AI tools do not meet their needs.

72%

of the workforce falls into just two archetypes: Stalwarts and Navigators.

Because both of these groups also have the largest Activation Gaps, they are the most important for businesses to focus on. Even a uniform 5% rise in global labor productivity could contribute roughly $6 trillion to the world’s GDP.

Understanding the archetypes

Seven workforces, not one

From the outset of our study, we wanted to break away from the limitations of viewing the workforce as a single homogenous entity. Each of us approaches work, risk and technology in unique ways. We knew it would be key to understand how these perceptions and approaches change employee willingness to use AI. 

To develop the archetypes, we first asked a series of detailed questions about respondents’ overall attitude toward technology (its risks and rewards) and work itself. We then modeled these insights against their experience with using AI.

To understand the dynamics within the archetypes, we created the Activation Quotient (AQ), which scores how engaged employees are with AI, and the Activation Gap (AG), which measures the distance between aspiration and action.

The figure below shows the seven archetypes that make up the modern workforce, ordered on the basis of their Activation Quotient.

Understanding the Activation Quotient

The Activation Quotient assesses the degree to which employees are engaged with AI, based on their current use, future expectations, skill levels and response to their organization's operating conditions.

The Activation Quotient’s four building blocks are Action (what they are doing today), Aspiration (what they wish they could do), Ability (what they have access to) and Conditions (how aligned they are with their organization’s view of AI adoption).

An archetype that scores above the workforce average of 61 is more likely to adopt AI with the right supports in place. Archetypes with below-average scores will require more effort to be engaged.

Understanding the Activation Gap

Employee engagement is only half the story. To understand the ease of enabling the workforce and the potential gains of doing so, we created the Activation Gap, a metric that measures the distance between how much a worker aspires to use AI and how much they actually do. In effect, this represents how much trapped potential still sits at the employee level.

A positive Activation Gap value indicates untapped potential. A negative value means the group is already pushing the boundaries of AI adoption.

The seven archetypes

Figure 1 

*Activation Quotient and Activation Gap

Note: For AQ, an archetype that scores above the workforce average of 61 is more likely to adopt AI. Archetypes with below-average scores will require more effort to be engaged.


Base: 10,000 workers in 10 geographies and multiple industries


Source: Cognizant Research

The seven archetypes

Figure 1 


Note: For Activation Quotient, an archetype that scores above the workforce average of 61 is more likely to adopt AI. Archetypes with below-average scores will require more effort to be engaged.


Base: 10,000 workers in 10 geographies and multiple industries


Source: Cognizant Research

Key findings

Unlocking trapped AI potential by focusing on four challenges in worker adoption

FINDING 01

Tool dissatisfaction is rampant

Across every archetype, employees express widespread dissatisfaction with how their organization is deploying AI

Between 72% and 82% of employees in any segment say their current AI tools do not meet their needs. Precisionists have the highest confidence in the tools they use, but they barely clear the bar at 28%, saying they are satisfied with current AI tools—the highest figure of any archetype but still resoundingly low. The failure to provide access to relevant tools is uniform across the workforce; Precisionists have just built more compensatory practices than any other group.

Of most interest are the Architects. Despite trusting their organization’s governance framework and agreeing with their employer’s approach to AI risk, 87% are also reaching for non-sanctioned tools when approved options fall short. Either the governance frameworks or the offered tools need to change.

What this means

Tool dissatisfaction is a baseline condition across the entire workforce, so any adoption metric built on activity alone will overstate progress. Investment in the tool, workflow and context layer is the precondition for successful intervention.

FINDING 02

Productivity gains are waiting to be tapped

Across the workforce, 80% of employees record a positive Activation Gap

This means businesses stand to gain a vast amount of AI value by tapping this hidden potential. To ascertain where those pockets of potential were greatest, we asked respondents to compare their current productivity gains with what that boost would be if provided with optimal tools and resources.

The difference was dramatic. When employees were asked about their current productivity, many reported gains at the lower end of the spectrum, 0% to 10%. Imagining a future with ideal support, far more expected significant improvements of 10% to 20% gains or more. To illustrate, even a uniform 5% rise in global labor productivity could contribute roughly $6 trillion to the world's GDP.

What this means

Equipping workers with the right tools, resources, skills and motivation to embrace AI can unlock substantial economic gains. The Activation Gap shows the potential is already present in the workforce and waiting to be released.

The table below shows where the largest pools of latent potential sit:

Table showing potential productivity improvements by archetype

Figure 2 


Note: A productivity tier is one band of self-reported gain. The bands rise from 0%–10%, to 10%-20%, to more than 20%.
Because Accelerators and Precisionists have already adopted AI to their greatest extent, there are no measurable productivity gains from closing the gap between what they’re doing, and what they want to do.

Base: 10,000 workers in 10 geographies and multiple industries


Source: Cognizant Research

The table below shows where the largest pools of latent potential sit:

Table showing potential productivity improvements by archetype

Figure 2 


Note: A productivity tier is one band of self-reported gain. The bands rise from 0-10%, to 10%-20%, to more than 20%. 

Because Accelerators and Precisionists have already adopted AI to their greatest extent, there are no measurable productivity gains from closing the gap between what they’re doing, and what they want to do.



Base: 10,000 workers in 10 geographies and multiple industries


Source: Cognizant Research

FINDING 03

AI is doing the easy work

Across archetypes, about one-third of employees are using AI for low-complexity work

This includes data analysis and reporting, summarizing and presentations. Meanwhile, more complex, strategic work sees far less use of AI: strategic planning, budgeting, directing autonomous agents.

However, this changes significantly when comparing the most and least AI-engaged workers. On average, 28% of Architects and Accelerators are already doing more complex work with AI, compared with just 15% of Holdouts.

This indicates a rising divide between activated and non-activated workers. While Architects and Accelerators are redefining their roles by blending their human skills and machine capabilities, the slower-adopting groups often experience AI as merely an efficiency tool. This growing asymmetry will eventually be reflected in performance data, leading to workforce decisions that are harder to reverse than the targeted activation interventions that could have closed the gap.

What this means

Leaders should identify the Architects and Accelerators and look to them for use cases for doing more complex work with AI, while also showing slower-moving peers how to uplevel their application of AI.

FINDING 04

A diverse workforce requires diverse approaches

Until now, workforce AI adoption has been a uniform story. Our research changes that narrative: No single AI strategy works for every kind of worker.

Each of the seven archetypes demands a tailored approach to AI adoption. Accelerators and Architects benefit from governance and knowledge capture aligned with their current practices. Precisionists require organizations to first understand their self-built solutions before scaling them. Navigators need access and permission, while Pragmatists are motivated by peer evidence in trusted contexts. Stalwarts require assurance and career relevance, while Holdouts need foundational skills that are consistently reinforced. 

What this means

Organizations need to treat worker groups differently, according to a clear understanding of their needs. This requires behavioral segmentation and interventions that match the actual barriers each group faces.

81% vs. 56%


The training inversion

Even something as basic as training shows significant variance across the archetypes. Consider that among employees who have completed AI training, 81% of the fast-adopting Architects rate it as effective, compared with 56% of Holdouts, the slowest-adopting group. The workers who could most benefit from training are least well-served by it.

Further, only 53% of Holdouts complete mandatory AI training when offered versus 83% of Architects. Mandatory training, the primary skilling instrument for most organizations, is reaching the workers who need it least, while leaving the hardest-to-activate groups further behind.

$6 trillion rests on meeting the workforce where it is

Our research found that, despite their differences, most people in the global workforce want to use AI—and they want to do more with AI than they currently can. That should be a wakeup call to organizations everywhere.

AI value is not in the technology itself but in the adoption of it. When there’s too much friction in adoption, there is less AI impact. What remains, then, is the strategy: a strategy that strives to resonate with the real workforce, which is a diverse one.

The archetypes in this report are the map, and the organizations that read it will be the ones that convert AI’s theoretical capability into realized productivity, one archetype at a time. Because a workforce that wants AI and an organization that can deliver it is the fastest way to realize AI value.

Please download the full report, which contains detailed archetype profiles and intervention frameworks.

Research

Ollie O’Donoghue

Head of Cognizant Research

Duncan Roberts

Associate Director, Cognizant Research

Editorial

Mary Brandel

Editor