TL;DR: AI is pushing companies to reshape their workforces quickly, and those moves get expensive to walk back when leaders act without knowing what their people can actually do, how roles are shifting, and who can adapt. The 55% regret figure points straight at that blind spot. Many organizations are committing to AI decisions before they have real evidence about who to redeploy, develop, keep, or hire.¹
What holds up is a measurable read on capability across new hires and current employees alike, rather than strategy built on titles, org charts, or skills taken at face value. That visibility lets leaders spot the potential already inside the company and see where outside hiring is genuinely needed. So for CHROs, CEOs, and the wider leadership team, the work starts before the next hard-to-reverse AI decision: map which capabilities to protect, build, and redeploy. Cost savings matter, but they make a thin strategy on their own.
The 55% regret figure is not a cautionary tale about AI. It is a visibility failure. Organizations are making workforce decisions based on what AI might eventually replace, without enough evidence about what employees currently do, which capabilities drive performance, and who can adapt as the work changes. Without that visibility, even well-intentioned decisions become expensive ones.
In Predictions 2026: The Future of Work, Forrester identified a costly reversal point in the AI labor story: 55% of surveyed employers regretted laying off workers because of the promise of AI, according to Computerworld’s reporting on the Forrester analysis.¹ Forrester’s own 2026 future-of-work forecast adds that half of AI-attributed layoffs are expected to be quietly reversed as companies confront the operational reality of replacing human talent prematurely.²
Every CEO, CFO, CHRO, COO, and talent leader under pressure to turn AI investment into productivity, cost control, and growth capacity should pay attention. An AI-driven workforce decision made without a clear view of workforce readiness risks knowledge loss, backfill expense, execution risk, and slower time-to-capacity when the business discovers the work still needs people. Return on Talent Investment (ROTI) reframes the question leaders should be asking before the next decision reaches the boardroom: what is this workforce actually capable of becoming, and what will it cost if we misread that?
Replacement thinking and augmentation thinking require different evidence
Forrester’s 2026 AI job forecast says AI will augment 20% of jobs over the next five years. Its future-of-work predictions also warn that over half of layoffs attributed to AI will be quietly reversed as companies realize the operational difficulty of replacing human talent too early.² ³
Replacement decisions and augmentation decisions require different evidence. Replacement thinking asks which roles can be removed. Augmentation thinking asks which work can be redesigned, which employees can perform at a higher level with AI, and which capabilities need to be measured before leaders approve redeployment, reskilling, or external hiring.
The organizations now regretting AI-driven decisions likely had no regrets about wanting to improve efficiency. Where things went wrong was acting before the organization fully understood the work. AI changes tasks before it cleanly eliminates roles. Many roles contain judgment, customer context, exception handling, institutional knowledge, risk awareness, and collaboration patterns that are difficult to see in a cost-reduction model and even harder to rebuild once lost.
AI workforce decisions need a performance framework
McKinsey’s 2025 global AI survey found that 88% of organizations now report regular AI use in at least one business function, while only about one-third have begun scaling AI programs and 39% report any enterprise-level EBIT impact from AI.⁴ BCG found that only about 5% of organizations have achieved substantial financial gains from AI, and approximately 70% of AI value comes from rethinking the people component rather than the algorithm or technology alone.⁵
Both findings point to the same conclusion: AI value depends on the operating model around the technology. Leaders need to understand how work changes, which people can adapt, which teams need development, and where external hiring is truly required.
Five diagnostic questions for any AI workforce decision
- Which tasks are actually changing, and which role outcomes remain business-critical?
- Which capabilities predict success in the redesigned work?
- Which existing employees already show the aptitude, judgment, and learning agility to work effectively with AI?
- Which employees can be reskilled, redeployed, or moved into adjacent roles faster than the business can hire externally?
- Which workforce decisions carry execution risk, customer risk, compliance risk, or productivity drag that could offset expected gains?
Working through these questions moves the conversation from AI enthusiasm to workforce discipline. It also makes the talent function more relevant to the CEO and CFO because the discussion becomes measurable: capability, time-to-capacity, early attrition risk, redeployment potential, and business continuity.
Measuring AI readiness across new hires and existing employees
Most organizations think about AI readiness in two separate tracks. When hiring, they look for candidates who list AI tools on their resume or have credentials from programs with AI in the title. For existing employees, they rely on manager perception, self-reported familiarity, or participation in a training module. Neither approach actually measures readiness. Both create blind spots that become expensive.
The more useful frame is to treat AI readiness as a single workforce question that applies equally to candidates coming in and to the people already in the building. A high-performing employee in a pre-AI workflow may need more support to adapt than their track record suggests. A newer hire with a less traditional background may have exactly the learning agility and practical judgment the role now requires. A strong internal candidate may be invisible to workforce planning because no one has measured what they can actually do with AI-augmented work.
Without consistent, objective measurement across both populations, leaders end up making redeployment and hiring decisions based on assumptions rather than evidence. They underestimate internal capability and overpay to source it externally. They retain people in roles they will struggle in and miss the ones who could step into something harder.
This is what AI PREVAIL was built to address.⁶ It measures AI skills, aptitude, and adaptability using the same science-backed assessment methodology across both hiring and existing workforce contexts, so leaders are working from a consistent view of capability rather than two disconnected data sets. Redeployment decisions become defensible. Hiring decisions become more precise. And the organization stops treating its internal talent pool as an afterthought to external recruiting.
The better path: measure, redeploy, then hire with precision
The World Economic Forum’s Future of Jobs Report 2025 found that employers expect 39% of key skills required in the labor market to change by 2030. Among the skills rising in importance: AI and big data, technological literacy, creative thinking, resilience, flexibility, agility, curiosity, and lifelong learning.⁷ Navigating that scale of disruption requires a workforce system that can continuously measure what people can do and where they can go next.
Step 1: Define how work is actually changing
Before deciding whether a role should be reduced, redesigned, redeployed, or rehired, the organization needs to understand the actual tasks, outcomes, risk points, and capabilities required for success. Job analysis is the starting point, not headcount spreadsheets.
Step 2: Measure AI readiness across the existing workforce
The goal is to see who can move, who needs development, and where the organization has capability it has not yet fully deployed. Readiness cannot be inferred from tenure or title. It requires objective, science-backed assessment.
Step 3: Let workforce data drive the decision
The most useful data is the data that changes a decision. Readiness insight should shape reskilling investment, internal mobility, role redesign, hiring profiles, and succession planning. If it only sits in an HR dashboard, it is not doing its job.
Step 4: Hire externally with precision, not volume
External hiring should focus on the capabilities the internal market cannot supply quickly enough. Skills-based hiring becomes more important as AI changes work because static credentials lose relevance faster. Harver’s skills-based hiring approach centers on job analysis, predictive assessments, evidence-based validation, human oversight, and the connection between assessment data and post-hire outcomes such as retention and performance.⁸
What better talent visibility looks like in practice
ADT used Harver to modernize its contact center hiring model after a shift from four regional call centers to nationwide hiring created pressure on recruiter capacity, hiring timelines, and early attrition. The results: a 13% reduction in attrition, 13,000 recruitment hours saved, a threefold increase in recruiter capacity, and a reduction in time-to-hire of more than 50%.⁹ Beyond the numbers, the change elevated the talent function itself. As the Senior Director of Talent Acquisition at ADT put it: “Harver transformed the recruiter role from reactive recruiters to true talent advisors.”
A second example shows the same principle through a skills-based hiring lens. Paramount Advertising used Harver to bring more objective, standardized data into entry-level sales hiring, moving away from outdated criteria that limited candidate pass-through and made it harder to identify high-potential talent. The company reported a 254% increase in ethnic diversity, a 56% reduction in attrition, a 12% increase in employee performance, and a 30% reduction in recruiting time. As Tanmay Manohar, VP of People Analytics & Workforce Planning, put it: “The objective data we got from Harver was the validating catalyst for us to add in even more rigor and standardization.”¹¹
When organizations replace subjective evaluation with job-relevant, predictive insight, they make better workforce decisions at scale and make them with confidence. It applies to new hires and equally to decisions about how to get the most from the talent already inside the business.
Organizations with a clear, data-driven view of workforce capability are better positioned to move quickly, invest wisely, and build a workforce that grows in value over time. The ones without that visibility are the ones making decisions they will later need to reverse.
The executive test before the next AI decision
AI will continue to reshape work, and many organizations will need to redesign teams, roles, and operating models. Bain argues that companies need to modernize workflow and workforce in parallel, because workforce modernization that trails workflow redesign leads to lower adoption, weaker scaling, disappointing return on investment, workforce disengagement, and technology skepticism.¹⁰
The organizations that get AI workforce decisions right are not the ones moving fastest. They are the ones moving with the best evidence. That is the real lesson behind the 55% regret figure.
Four sign-off questions before approving any AI workforce decision
- What work will AI actually change, and over what timeframe?
- Which human capabilities will matter more as a result?
- Who in the current workforce has the readiness to grow into that future?
- What is the full opportunity cost if the organization acts without that knowledge?
Organizations that treat AI readiness as a measurable business discipline will know where capability already exists, where it needs to be developed, and where external hiring creates the most strategic value. They will make AI decisions with workforce evidence and build the kind of workforce that makes every AI investment worth making.
See what your workforce is capable of before the next AI decision is on your desk.
Sources
¹ Computerworld. “Forrester: Companies regret many AI-related layoffs.” 2025. https://www.computerworld.com/article/4084372/analysts-companies-will-face-setbacks-after-ai-layoffs.html
² Forrester. “Predictions 2026: The Workforce Muddles Through Ambient Disruption.” 2025. https://www.forrester.com/blogs/future-of-work-predictions-2026-whats-coming-for-work-and-the-workforce/
³ Forrester. “Forrester: AI-Led Job Disruption Will Escalate, While Fears Of A Job Apocalypse Are Overstated.” 2026. https://www.forrester.com/press-newsroom/forrester-impact-ai-jobs-forecast/
⁴ McKinsey & Company. “The State of AI: Global Survey 2025.” 2025. https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai
⁵ Boston Consulting Group. “AI Transformation Is a Workforce Transformation.” 2026. https://www.bcg.com/publications/2026/ai-transformation-is-a-workforce-transformation
⁶ Harver. “AI PREVAIL.” 2026. https://harver.com/ai-prevail/
⁷ World Economic Forum. “Future of Jobs Report 2025.” 2025. https://www.weforum.org/stories/2025/01/future-of-jobs-report-2025-jobs-of-the-future-and-the-skills-you-need-to-get-them/
⁸ Harver. “Skills-Based Hiring.” 2026. https://harver.com/solutions/use-case/skills-based-hiring/
⁹ Harver. “How ADT Used Predictive Hiring to Reduce Attrition, Save Thousands of Hours, and Triple Recruiter Output.” 2026. https://harver.com/clients/how-adt-used-predictive-hiring-to-reduce-attrition-save-thousands-of-hours-and-triple-recruiter-output/
¹⁰ Bain & Company. “Want More Out of Your AI Investments? Think People First.” 2026. https://www.bain.com/insights/want-more-out-of-your-ai-investments-think-people-first/
11 Harver. “How Paramount Advertising increased workforce ethnic diversity by 254% and reduced attrition by 56%.” 2026. https://harver.com/clients/tech-media-telecommunications/paramount/
