AI Will Reshape More Jobs Than It Replaces, BCG Study Finds
Artificial Intelligence is at the center of one of the biggest labor market transformations in recent decades. And when the Boston Consulting Group, the well-known BCG, points out that between 50% and 55% of jobs in the United States will be reshaped by AI over the next two to three years, the natural first thought is: is my job going to disappear?
But hold on.
That eye-catching number is not a sign of professional apocalypse. It actually points to something far more complex and, depending on how you look at it, even encouraging: most jobs are not going to vanish — they are going to change shape. For many professionals, this means keeping the same role or a similar one, but facing radically new expectations about how they work and what they deliver.
The difference between reshaping and replacing is huge, and understanding that makes all the difference for anyone wondering what lies ahead.
Here is a summary of what you will find in this article:
- What the BCG study really says about the future of jobs
- The two factors that define how AI will impact each role
- The 6 labor market impact segments, with real-world examples
- Why task automation does not necessarily mean job loss
- The big unknowns that will influence the pace of this transformation
- What business leaders need to do now to avoid falling behind
If you work in tech, lead teams, or simply want to understand where the market is heading, this one is for you. 🚀
What the BCG Study Is Really Saying
First things first — it is worth understanding what is behind the number that generated so much buzz. The Boston Consulting Group is not saying that half of American workers will lose their jobs. What the study indicates is that the nature of these roles will change significantly, and that this shift will happen at a much faster pace than any previous technological transformation the labor market has ever faced.
The analysis, based on microeconomic modeling, identified a substantial portion of the workforce for which AI will augment the capabilities of professionals in their current roles, rather than simply eliminating them. Beyond that: when the productivity gains generated by AI use trigger an increase in demand for the final product or service and the potential for expansion is high, BCG believes there will be a need for more human roles and, in some cases, entirely new ones.
While reshaping and the creation of new jobs will happen relatively quickly, the outright replacement of jobs by AI will be slower. The study estimates that within five years or perhaps further out, between 10% and 15% of jobs in the U.S. could be eliminated. That number considers 165 million American jobs spread across roughly 1,500 different occupations. It is a considerable level of potential job loss, but very different from the 50% that many people interpreted too hastily.
An important point: BCG makes it clear that this analysis is not an unemployment forecast. It does not account for macroeconomic factors like geopolitics or inflation, nor does it factor in the impact of potential new AI breakthroughs beyond the capabilities of current frontier models. It also cannot resolve powerful unknowns, such as the future impact of AI on job accessibility or the speed at which the technology will be adopted at scale.
The Two Factors That Determine Everything
Understanding how AI will impact a specific profession inevitably comes down to two concepts that BCG places at the center of the analysis.
Replacement Versus Augmentation
The first factor is the relationship between replacement and augmentation of human capabilities. To illustrate this difference, the study uses two examples that are already experiencing the adoption of agentic AI at scale: call center representatives and software engineers.
A call center representative is typically responsible for resolving a defined set of customer queries within established workflows. A large portion of the work involves structured interactions such as account inquiries, policy explanations, and troubleshooting following scripts. When AI systems can reliably handle these repetitive queries end to end, fewer representatives are needed. The workflow can be cleanly divided: AI handles frontline interactions and humans take escalations and exceptions. Overall, employment in the call center representative role will decline as the more structured tasks are absorbed by the system.
A software engineer, on the other hand, produces a very different kind of output. Although programming includes routine elements, the core value of the role lies in systems design, architectural judgment, performance-versus-cost tradeoffs, and translating business needs into technical solutions. AI can dramatically accelerate code generation and testing, but with current capabilities, it cannot replace the system-level judgment needed to be accountable for the end-to-end outcome. The work cannot be cleanly divided between system and engineer. Instead, software development becomes a continuous interaction where engineers set objectives, refine outputs, validate results, and integrate components into broader systems.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
Limited Demand Versus Expanded Demand
The second factor is what the study calls demand expandability, and this is where many people are surprised. When AI reduces the cost of delivering a business outcome or end product, the central question is: does demand for that outcome expand or remain limited?
This dynamic is not new. Economists have long observed that efficiency improvements can increase total consumption rather than reduce it, a phenomenon often called the Jevons Paradox. When the cost of a resource falls, usage can rise. The same logic applies to labor: if productivity increases, the impact on employment depends on how demand for the output responds.
Software engineering illustrates expandable demand. Organizations across every sector continue to face persistent unmet needs for digital products, automation, and new features. As AI reduces the cost and time required to build software, organizations often build more. Total output volume expands and overall employment can remain stable or grow, even as individual engineer productivity rises. The continued growth in the number of software engineers in the years following the introduction of ChatGPT in 2022 illustrates this phenomenon.
Call center representatives illustrate limited demand. The volume of incoming interactions is largely determined by the size of the customer base and the frequency of service needs. When AI reduces the cost of handling routine queries, the number of interactions does not expand proportionally. In this context, productivity gains tend to reduce the number of representatives needed.
The 6 Labor Market Impact Segments
The map BCG proposes organizes occupations into six distinct groups, each telling a different story about how the relationship between humans and automation will play out over the coming years. These segments form what the BCG Henderson Institute calls AI Work Disruption Segments.
Amplified Roles (5% of current jobs)
When AI augments human capabilities and demand expands, employment can remain stable or grow. Humans stay central to value creation, and there may be upward wage pressure as higher productivity increases competition for skilled talent.
Software engineering fits into this category. So do many lawyers, particularly those in advisory and practice areas requiring a high degree of judgment. Investment in legal tech startups, such as Harvey AI, reached record levels in 2025, generating significant discussion about how AI will reshape the field. As AI accelerates research, document drafting, and case preparation, legal services can become more accessible. In domains where unmet demand exists — such as compliance, regulatory review, and contract management — lower costs can increase the volume of legal work.
Rebalanced Roles (14% of current jobs)
When AI augments work but demand is limited, headcount may remain stable while roles are redesigned. Routine tasks are automated and more complex responsibilities expand. Upskilling becomes essential as work shifts toward higher-value activities.
One example is content marketing, where demand is constrained by marketing budgets and strategic priorities. At the same time, with audience fragmentation and influencers and language models reshaping customer journeys, brands need more content, not less — delivered in a more targeted way to the right audience at the right time. Marketing professionals, rather than being siloed by channel as they often are today, will see their roles expand to become omnichannel specialists capable of thinking through campaigns end to end.
Divergent Roles (12% of current jobs)
Where AI replaces human tasks but demand remains expandable, the effect on employment becomes uneven. In these roles, entry-level and junior positions are more exposed to automation in the short term. A considerable share of the structured tasks traditionally performed at these levels are among the first to be automated, which means there will be some job losses in this category. At the same time, responsibilities at senior levels persist and may grow.
Insurance sales agents fit into this category. AI automates routine activities such as lead qualification, quote generation, and policy comparisons. At the same time, significant protection gaps remain, particularly in broader life insurance coverage and among small businesses that continue to be underinsured. By reducing distribution and service costs, AI enables insurers to reach previously underserved customers, expanding overall market participation.
IT support technicians offer a similar example. AI can resolve routine tickets and automate diagnostics. As digital infrastructure expands and systems become more complex, the need for advanced troubleshooting and systems oversight may grow.
Replaced Roles (12% of current jobs)
Only when demand is limited and AI directly replaces human workers in core tasks do roles fall into this category. Efficiency gains convert into net job losses and downward wage pressure develops for the positions that remain.
Certain financial analyst roles fit here, as do call center representatives. The volume of financial analysis is largely tied to existing reporting cycles, investment mandates, and internal decision-making processes. When AI automates routine modeling, data aggregation, and initial interpretation, output does not expand proportionally.
It is important to note that replacement does not imply permanent exclusion from the workforce. As augmented, less automatable, and emerging roles expand in other parts of the economy, workers in positions exposed to replacement can transition into those roles with targeted reskilling and mobility support.
Enabled Roles (23% of current jobs)
AI will become part of the daily routine for 23% of jobs, reshaping how tasks are performed but without fundamentally altering how the work is structured. Workers will be expected to use AI to improve efficiency, accuracy, and decision-making, while the core responsibilities — which often involve the physical presence of a human worker, interpersonal interactions, or domain-specific expertise — remain human-led.
Clinical assistants and lab technicians illustrate this dynamic. Their work remains manual or patient-facing, but AI increasingly supports documentation, workflow coordination, and aspects of diagnostic analysis. For instance, clinical assistants may use AI for real-time note-taking and patient triage, and lab technicians may leverage AI to interpret test results or flag anomalies.
Limited Exposure Roles (34% of current jobs)
In the remaining share of jobs with lower automation potential, both the technical feasibility of automation and the scope for AI-driven productivity gains remain limited. The work is often highly contextual, relationship-driven, or dependent on human physical presence in ways that are difficult to codify or scale through AI.
Physicians and teachers are illustrative. Both require the ability to form complex judgments, engage in interpersonal interactions, and adapt in real time to individual needs. AI may assist in limited, task-specific ways, but it will not significantly reshape either of these roles in the short term. The core of the work — patient care and instruction — remains fundamentally human.
The Scale of Impact: The Numbers in Perspective
Looking across all six segments together, between 10% and 15% of jobs are vulnerable to elimination. These are the jobs in the replaced and divergent categories, weighted by their automation potential and adjusted for demand expansion. At the same time, 50% to 55% will be reshaped, including jobs in the amplified, rebalanced, and enabled categories, plus the remaining portions of jobs in the divergent and replaced categories where the work is not vulnerable to elimination. The remaining roles lack significant short-term automation potential.
Meanwhile, an unspecified but significant number of jobs will likely be created as new demands emerge and new capabilities are enabled by AI. This is a part of the equation that often gets left out of alarming headlines.
The Big Unknowns That Will Influence the Pace of Change
The BCG analysis makes it possible to categorize roles, but that is not the whole story. Each segment, and potentially individual roles within each segment, will evolve and have different cascading effects that need to be managed across different time horizons.
The Side Effects of Agentic AI Transformation
First, a central challenge in the AI era will not simply be the number of jobs affected, but the speed at which workers can be reskilled and repositioned into redesigned roles. Absorbing a workforce shift of this magnitude will require deliberate investment in reskilling, mobility, and capability building.
Second, as AI absorbs much of the routine work that historically justified large entry-level hiring cohorts, fewer execution-focused positions will be needed. In the short term, the volume of entry-level jobs may decline. Over time, however, these jobs will be redefined and filled by candidates prepared to take on higher-order tasks — such as overseeing AI outputs, managing exceptions, and contributing to more complex problem-solving from the very start of their careers.
In this environment, AI fluency may become an increasingly important complement to experience when assessing readiness for expanded responsibilities. In some cases, this will create opportunities for junior candidates, including recent graduates, who may be more familiar with AI than workers with more years under their belt. 💡
Third, skill thresholds will rise. Redesigned roles will require employees to demonstrate greater expertise, oversight, and accountability, raising the premium on domain knowledge and sound judgment. The most durable roles will tend to require higher credentials and greater seniority.
Finally, cognitive load will intensify. Many roles currently balance structured execution with higher-level thinking. As repetitive tasks are automated, the remaining work will concentrate on problem-solving, decision-making, and integrating complex inputs. While some workers will thrive in more judgment-driven roles, others may struggle with the shift to sustained high-level cognitive engagement.
The Lag Between Potential and Actual Adoption
Another fundamental unknown is timing. Economic impact often lags behind model capability, because it also depends on application maturity, workflow redesign, integration with legacy systems, and the availability of human capital capable of deploying and managing these AI systems.
Contact center tools are among the most mature applications, yet overall market penetration remains limited relative to total industry size. In BCG’s experience, full worker replacement tends to advance more slowly than augmentation. Replacement often means fewer humans are kept in the loop, which requires extensive process redesign and formalization of tacit knowledge.
Scaling agentic systems requires specialized integration talent, including field engineers, systems integrators, and project managers who adapt systems to company-specific contexts. The supply of these professionals remains limited relative to demand, making implementation capacity a central bottleneck. These roles are also an example of new jobs emerging from AI adoption.
As a result, there will likely be a multi-year lag between automation potential and actual labor market impact. Industries with high automation potential do not always demonstrate high levels of scaled AI adoption. In others, like technology and software, adoption is already well above average. Still other industries, including financial services and legal, have substantial automation potential, but at-scale implementation has not yet kept pace.
What Leaders Need to Do Now
For those in leadership positions, the BCG study offers clear guidance organized around four fundamental starting points.
Embed workforce strategy into competitive strategy
AI will reshape competitive dynamics and enable new business models. Companies that respond by accelerating innovation, redesigning offerings, or reconfiguring value delivery at the enterprise, business line, product, and function levels will fundamentally change the nature of work within their organizations. Workforce strategy cannot sit downstream of automation. It needs to be integrated into strategic planning.
Leaders should avoid reactive cost actions driven by headlines or competitor behavior. Workforce decisions need to reflect the specific mix of automatable and augmentable work within that particular company. Copying another company’s reductions without understanding your own exposure undermines both productivity and long-term competitiveness.
Focus automation on redesign, not just cost reduction
Agentic AI is not a blunt instrument. Different roles require different approaches. Sometimes cost is the dominant factor. Other times, different priorities take precedence — such as speed of work and quality. Cost actions like hiring freezes or headcount reductions are visible and straightforward. But when AI generates productivity rather than cuts, ROI becomes harder to define and defend.
In these situations, leaders need to redesign workflows and rethink how performance is measured. Capturing value requires new domain-specific KPIs that connect productivity gains to tangible outcomes — such as higher revenue per employee, more product delivered, or stronger customer impact.
Put upskilling, reskilling, and repositioning at the center of strategy
Just because a job remains does not mean employees are ready for it. Upskilling, reskilling, and structured repositioning pathways need to become central to workforce strategy. Additionally, workers will likely need more frequent requalification — not just a one-time effort — as the technology evolves.
For each of the five impact segments where AI is expected to materially affect work, BCG offers specific guidance:
- Amplified Roles: focus on retaining and developing talent, reinventing team-based workflows to maximize human-AI collaboration, and redefining what excellence looks like
- Rebalanced Roles: the opportunity lies in role redesign — identifying repeatable components that can be automated and reinvesting the time savings into higher-value activities
- Divergent Roles: the challenge is structural, requiring deliberate redesign of career pathways with accelerated development tracks and clear routes to higher-skill responsibilities
- Replaced Roles: reimagine end-to-end processes around AI agents while developing workforce transition planning in parallel with AI deployment
- Enabled Roles: focus on embedding AI into daily workflows and building broad AI fluency across the workforce
Shape the AI narrative to unlock performance
Sequencing and signaling matter. Leading with highly replaceable roles may deliver short-term efficiency gains, but it can create a demoralized environment that undermines the broader transformation. When employees associate automation with layoffs, engagement drops and the motivation to reskill disappears. Leaders need to communicate clearly that, if workers invest in development, AI in most roles will be about value creation, not replacement. The narrative set by leadership will determine whether the workforce embraces the transformation or resists it.
A Landscape of Opportunity, But Also Uncertainty
AI creates a massive opportunity for business leaders, but it also brings significant uncertainty about how to seize it. All of this is happening in a charged environment. In some cases, restructurings that would have occurred anyway — as part of the normal business cycle — will likely be attributed to AI, generating fear at a societal level.
At the same time, the impact of AI will vary significantly across companies. Some will embrace AI to drive innovation and growth, while others will focus on efficiency and automation. This can lead to very different talent strategies: some companies reducing headcount while others hire aggressively.
The imperative for CEOs today is to focus on striking the right balance between automation, upskilling, and deliberate talent planning — delivering ROI at scale for the business while helping their employees develop the skills they need to thrive in the AI era. Those who can navigate this transition with strategic vision and genuine care for their people will build more resilient and competitive organizations. And those who ignore the complexity of this moment, treating everything as a simple cost-cutting equation, will find out very quickly that the best talent does not stick around in environments that fail to invest in their future. 🎯
