How can we help?
AI impact on jobs & workforce
AI Impact Modelling
Understand how AI changes your jobs, roles and staffing needs before deciding on tools, positions or budgets. Harbinger breaks work down into tasks, assesses conventional digitisation, automation and generative AI, and translates the impact into future job profiles, skills, capacity scenarios and a practical workforce roadmap.
Explore AIM Preview · launch 20 January 2027 ↗
For executive management, HR, Finance and transformation leaders who want to understand which work will remain, change or emerge over the coming years, and what that means for FTE, skills and value creation.
Experience & trust
We connect workforce strategy, HR Analytics and practical organisational work
AI Impact Modelling builds on Harbinger's experience at the intersection of executive management, HR, workforce metrics, workforce management, transformation and daily operations . This is where AI impact needs to become concrete decisions about work, roles and capacity.
Our expertise
From Strategic Workforce Management to AI Impact Modelling
Harbinger has spent years modelling staffing needs, roles, KPIs and transformation. AI Impact Modelling adds a critical variable: How does AI change the task mix within a job?
“AI delivers little if nobody pays attention or decides what to do with it. Without a clear target state, a model can be made to mean anything.”
Tim Grove · CEO Harbinger AG
When AI moves faster than your job architecture
Most businesses already invest in AI. Many still do not know what work will remain afterwards.
Copilot, Gemini, ChatGPT and specialist AI agents are entering organisations. Meanwhile, job descriptions, headcount plans and competency models often stay the same. The risk is clear: Technology changes tasks faster than businesses rethink their roles and workforce.
New tools arrive. Old ways of working remain.
Licences have been distributed, use cases collected and training delivered. Yet it remains unclear which tasks will disappear, where AI will support people and which new responsibilities will emerge.
Workforce plans rely on yesterday's roles
FTE and budgets are rolled forward from historical positions. When 20, 30 or 50 per cent of a task mix changes, an old job description is no longer a reliable planning basis.
Efficiency gains need a destination
Time saved becomes valuable when it improves decisions, quality, customer or employee contact, or changes staffing requirements. That translation is what makes the business case.
AI Impact Modelling starts with the work. We examine which tasks create value today, which technology can take them over or accelerate them, and how that can produce better jobs and a more effective workforce.
Why this belongs on the management agenda now
AI is changing the architecture of work.
Recent labour-market research points in a common direction: GenAI primarily affects bundles of tasks within existing jobs, skill requirements are changing faster, and employers are beginning to factor automation potential into workforce decisions. A roadmap focused only on tools is therefore insufficient.
1 in 4
workers worldwide
is in an occupation with at least some GenAI exposure, according to the ILO. Its emphasis is important: in most cases, jobs are likely to be transformed rather than replaced entirely.
Source: ILO, 2025 40 %
of employers
expect to reduce their workforce where AI can automate tasks, according to the World Economic Forum. Reliable scenarios matter before headcount decisions are made.
Source: WEF, Future of Jobs 2025 7 → 20 %
AI adoption in businesses
Across OECD countries, the share of businesses using AI rose from around 7 per cent in 2021 to around 20 per cent in 2025. Skills and organisational development need to keep pace with adoption.
Source: OECD, 2026 ~225
new skills per highly exposed occupation
PwC reports a substantial rise in skill requirements in highly AI-exposed occupations in Germany between 2019 and 2025. Role profiles need to become more adaptable.
Source: PwC AI Jobs Barometer 2026Turning AI impact into a future role
An HR generalist example: rebuilding the task mix within the job
The useful unit of analysis is the bundle of tasks and their share of working time. A typical office-based role combines administrative routines, rules-based processes, knowledge work, communication and complex decisions. AI Impact Modelling separates these components and recombines them into a future role.
Today · illustrative
HR Generalist
- Prepare documents & master dataTemplates, amendments, follow-up and data maintenance.
- Draft routine communicationsLetters, queries, information and recurring coordination.
- Prepare reports & informationCollect, structure, summarise and present data.
- Coordinate onboarding & processesManage status, handovers, deadlines, checklists and interfaces.
- Advise managers & employeesHandle exceptions, context, conflicts, decisions and personal communication.
Illustrative range of change 45–60 %
of today's task mix could change substantially through standardisation, automation and GenAI. That does not necessarily mean it disappears.
Digitise / standardise 10–15 percentage points Automate 10–15 percentage points Augment with GenAI 25–30 percentage points
Tomorrow · possible future role
People Operations & Workforce Partner
- Advise on complex cases & leadershipMore time for situations where context and human judgement matter.
- Validate AI & workflow outputsOwn quality, plausibility, compliance and escalations.
- Improve processes & AI use casesIdentify automation opportunities continuously and redesign workflows.
- Develop workforce insightsTranslate data, patterns and capacity into practical decisions for managers and HR.
- Shape the employee experienceUse released capacity for better communication and more effective services.
People Operations PartnerHR Operations & AI PartnerWorkforce Services Partner
Please note: This example is illustrative and is not an industry benchmark. In a project, we establish your actual activities, time allocations, process maturity and technology conditions. Those findings determine the credible range of change.
Several technologies can change work
Three technology levers, with different effects on jobs
AI projects often group every efficiency opportunity under the same label. That can overstate potential, lead to poor technology choices and blur accountability. Our models distinguish conventional digitisation, automation and generative AI.
1 · Digitisation
Standardise work & make it digital
What does this mean?
Digital representation, standardisation and visibility of work steps, without automated decisions.
- digital forms and structured inputs
- consistent templates and standard text
- central records and tracking
- clear handovers between roles and systems
Boundary: No automated process control or independent decision-making.
2 · Automation
Automate rules-based, high-volume work
What does this mean?
Recurring steps run automatically according to defined rules, for example through workflows or robotic process automation.
- status changes, handovers and reminders
- deadline and process monitoring
- transfer of standardised datasets
- triggers, escalations and rules-based routines
Boundary: No substantive judgement on complex exceptions or ownership of accountability.
3 · Generative AI
Augment thinking, structuring, writing & analysis
What does this mean?
GenAI supports knowledge work by condensing information, generating alternatives, identifying patterns and preparing decisions.
- summaries and structuring
- drafts and writing support
- pattern, variance and information analysis
- preparation of decision materials
Boundary: No accountability, no guaranteed truth and no substitute for professional judgement in critical decisions.
What matters: These three lenses together reveal what may actually happen to a job. Sometimes sound standardisation offers the greatest benefit. In other cases, GenAI profoundly changes a role even though no process is fully automated.
Explained in four minutes · German-language video
How AI changes jobs, roles and staffing needs
In this German-language video, Tim Grove and Jeanette Weissgerber explain why analysis needs to focus on the tasks changing within each job and how those changes lead to future roles, skill profiles and capacity decisions.
Harbinger perspective: AI rarely replaces a whole job at once. It shifts tasks. We make that shift possible to model before it creates unplanned consequences for headcount, skills or transformation.
Our decision framework
What should work look like tomorrow?
AI Impact Modelling is a task-based assessment of AI impact. We combine job and task analysis, automation assessment, workforce scenarios and job redesign to provide a practical basis for decisions by management, HR and Finance.
1
Make work visible
We capture real tasks, responsibilities, time allocations, inputs, outputs and interactions, then compare them with existing job descriptions.
2
Assess the impact
We assess each task for value contribution, standardisation, automation, GenAI augmentation and the human judgement it requires.
3
Redesign jobs
We build future roles from tasks that remain, change or emerge, including their skills, interfaces and accountability.
4
Decide on workforce implications
Changes to roles inform scenarios for FTE, capacity, reskilling, redeployment, organisation and implementation priorities.
What AI Impact Modelling provides
- a structured assessment of AI impact at task and role level.
- a job redesign approach for more rewarding roles that create greater value.
- a scenario framework for skills, capacity and staffing needs.
- a management and workforce project that supports concrete decisions.
Where the scope ends
- selection or technical implementation of AI software.
- prompt training or general AI awareness courses.
- a blanket assessment claiming a fixed percentage of jobs can be removed.
- forecasts without validation by business functions and against company context.
The connection to Strategic Workforce Management
Understanding how technology changes future staffing needs
Strategic Workforce Management addresses the broader planning question: What workforce will we need? AI Impact Modelling examines how AI changes the task mix inside individual roles. Together, the two approaches show whether the business will need more, less or simply different capacity in the future.
Strategic Workforce Management
How many people and which skills will we need?
- Business strategy and growth or contraction scenarios
- Demographics, attrition and the labour market
- FTE and skill requirements
- Make, Buy, Build and location decisions
+
AI Impact Modelling
How does AI change work within these roles?
- Break down tasks and understand time allocations
- Assess digitisation, automation and GenAI
- Redesign job architecture and future roles
- Derive capacity effects, skills and redeployment options
A shared objective
A credible future workforce model connecting the business, work, technology and workforce planning.
Concrete project outputs
What you receive from an AI Impact Modelling project
The project should leave you able to make better decisions. Its outputs connect work, roles and workforce implications to budget planning, organisational design, skills strategy and transformation.
Task Map & Job Decomposition
A consistent picture of the work actually performed in the roles under review, the time it takes and where it creates value.
- Task, responsibility, frequency and time allocation
- Input, output and degree of interaction
- Actual work compared with existing job descriptions
AI, Automation & Digitisation Portfolio
Each task is assessed against distinct technology options, showing which intervention makes sense alongside its AI exposure.
- conventional digitisation
- workflow / RPA / automation
- GenAI augmentation
Role Redesign & Future Job Descriptions
The analysis produces future roles with an appropriate task mix, new responsibilities and possible role titles.
- tasks that remain or disappear
- new AI and governance responsibilities
- greater scope for valuable work
Capacity & Headcount Scenarios
We translate changes in task allocation into several realistic workforce scenarios, making the assumptions and uncertainty clear.
- Base / Advanced / Transformational Case
- FTE and capacity ranges
- assumptions, risks and dependencies
Skills, Reskilling & Redeployment
New tasks create new skill requirements. We show which capabilities need to be developed, redeployed or sourced externally.
- skill gaps by future role
- reskilling and learning priorities
- redeployment options before considering workforce reductions
Executive Roadmap & Decision Framework
Management receives clear priorities: what to decide now, what to pilot, what to prepare technically and what to scale later.
- quick wins and structural opportunities
- budget and organisational decisions
- governance, change and implementation
The scope can scale. Start with a clearly defined area such as HR Operations, Finance or Procurement, then extend the method to further functions or the wider corporate workforce.
Assessing potential with appropriate precision
The ranges an AIM project can examine
The following figures come from earlier AIM communications and are presented as potential ranges to investigate . In a project, they must be validated or rejected against your own roles, processes and technology conditions.
60–80 %
tasks with optimisation potential
In highly administrative or knowledge-intensive areas, a substantial share of tasks may be suitable for standardisation, digitisation, automation or augmentation.
20–35 %
time savings in administrative task groups
Research, documentation, information preparation, routine communication and recurring coordination often offer useful opportunities.
10–15 %
capacity for higher-value work
Some released time can be redirected into advice, quality, customer or employee contact, management and new responsibilities.
30–50 %
better use of existing AI tools
When job design, use cases and responsibilities fit together, existing AI tools can be used more effectively.
These ranges are neither guarantees nor general benchmarks. Their feasibility depends on task mix, process maturity, systems, data access, regulation, adoption and implementation quality.
Job quality
Less routine, greater contribution
Job reinvention rebuilds roles with more room for judgement, customer and employee contact, responsibility and development.
Capacity
More output with less friction
Technology can release capacity. Management decides whether to use it for growth, service, quality improvements or a change in staffing requirements.
Planning confidence
Headcount decisions grounded in the work
FTE decisions are connected to task analysis and scenarios, providing a stronger basis than general cost targets or tool promises alone.
Our project approach
Four steps from today's work to a future workforce model
The method is modular and scalable. A focused pilot begins in one clearly defined function and quickly produces initial, testable hypotheses. You then decide how extensively to redesign roles and apply the scenarios to workforce, budget and organisation.
1Task Analysis & Mapping
Make work visible
We capture the main activities and structure them by responsibility, time allocation, frequency, inputs, outputs and interaction.
- Review documents and role lists
- Conduct expert interviews
- Validate current tasks
- Assess each task's value contribution
Outcome
A complete task map as the basis for assessment and scenarios.
A complete task map as the basis for assessment and scenarios.
2Automation & Prioritisation
Assess technology potential
We examine every task through digitisation, automation and GenAI, alongside risk and the human judgement required.
- Assess suitability for standardisation
- Examine automation and workflows
- Model GenAI augmentation
- Prioritise value contribution and feasibility of change
Outcome
An impact matrix or heatmap with clear priorities and boundaries.
An impact matrix or heatmap with clear priorities and boundaries.
3Future Scenarios & Impact
Rebuild the roles
We develop two or three future scenarios: what remains, what is automated or augmented, and which new tasks and skills emerge.
- Develop future job profiles
- Define new task bundles
- Describe skill shifts and responsibilities
- Model capacity and FTE scenarios
Outcome
Comparable scenarios for roles, skills, capacity, opportunities and risks.
Comparable scenarios for roles, skills, capacity, opportunities and risks.
4Roadmap & Recommendations
Implement workforce decisions
The target state becomes a prioritised roadmap for job redesign, technology, skills, reskilling, organisation, governance and change.
- Quick wins and pilots
- Reskilling / Redeployment
- Organisational and budget implications
- Change, governance and measurement
Outcome
A practical decision basis and an achievable transformation path.
A practical decision basis and an achievable transformation path.
Speed with a clear rationale: We use AI where it accelerates analysis and comparison. Assessments remain explainable and are validated with your business functions, keeping company context central.
Accountability remains human
Build AI governance, employee representation and change into job design from the start
As AI moves closer to workforce decisions, task allocation or performance assessment, governance and transparency become more important. We treat technology potential as a decision space with clear boundaries and accountability.
Human Accountability
We define which tasks AI may prepare or execute, and where human judgement, approval, responsibility or escalation must remain.
Data protection & employee representation
Data access, transparency, role changes and possible effects on employees are addressed early. Employee representatives can be involved according to the project context.
EU AI Act & application context
AI systems used in employment can raise particular regulatory questions. AIM distinguishes job and opportunity analysis from the later legal approval of a specific system. Specialist legal advice can be added where required.
The EU AI Act follows a risk-based approach; certain AI uses in employment may fall into a high-risk category. Harbinger advises on organisational and workforce implications and does not act as a law firm. Information from the European Commission →
When to start
Where AI Impact Modelling adds the greatest value
The strongest starting point is a real management decision. The approach is particularly relevant in complex organisations with many office-based roles, substantial knowledge work and active AI or automation initiatives.
A strong fit if you …
- are rolling out Copilot, Gemini, ChatGPT Enterprise or AI agents and want to understand the organisational implications first.
- need to replan headcount or functional budgets while the AI effect remains uncertain.
- have job descriptions and role architecture that are visibly outdated.
- have productivity objectives and want an evidence-based approach to workforce implications.
- need reliable plans for reskilling, redeployment or new job families.
- need a shared decision basis for the CEO, CHRO, CFO and business functions.
Outside the scope if you …
- primarily want to select or technically implement AI software.
- need prompt training or a general AI awareness session.
- have already fixed a workforce reduction target and seek retrospective justification.
- are unwilling to involve actual tasks, role holders and business functions in the analysis.
- expect an exact five-year forecast without scenarios, assumptions or uncertainty.
Questions from executives, HR & Finance
Frequently asked questions about AI Impact Modelling, job redesign and AI in the workforce
What is AI Impact Modelling?
AI Impact Modelling is Harbinger's task-based consulting approach to assessing how AI affects work. We break roles into tasks, distinguish conventional digitisation, automation and generative AI, assess value and change potential, and develop future role, skill and capacity scenarios.
Why analyse tasks rather than job titles alone?
A title reveals little about where AI will have an effect. Some tasks follow rules and can be automated; others can be accelerated by generative AI or require human judgement. Task-level analysis shows what may disappear, change, become more valuable or emerge within a role.
Which jobs are most relevant for AI Impact Modelling?
Knowledge-intensive, office-based roles involving information, documentation, analysis or coordination are particularly relevant. Examples include HR, Finance, management accounting, Procurement, Legal, Sales, project management and corporate administration. The approach can extend across functions.
Is AI Impact Modelling about cutting jobs?
The analysis starts with future work. It may show reduced capacity requirements, a need to reshape roles or develop skills, or opportunities to redirect released time into higher-value activities. Headcount implications are an analytical outcome rather than a predetermined target.
How do automation and generative AI differ?
Automation follows defined rules and suits recurring workflows, handovers, status changes and triggers. Generative AI primarily supports structuring, writing, analysis and decision preparation. Conventional digitisation standardises work and makes it digital before either is applied.
How does AI Impact Modelling differ from Strategic Workforce Management?
Strategic Workforce Management considers future staffing more broadly, including volumes, skills, demographics, locations and business scenarios. AI Impact Modelling examines how AI changes tasks within roles and the resulting job design, skills and capacity needs. The services can be used independently or together.
What data and documents are needed to start?
Existing job descriptions, organisation charts, role lists and process information usually support a focused start. Actual tasks and time allocations are then validated with business functions and role holders. Workforce, FTE, skills or process data are added according to scope.
Do our job descriptions need to be up to date?
No. Outdated or generic descriptions are a common starting point. We treat them as hypotheses and compare them with actual work, producing a task map grounded in current activity.
How soon can we see initial results?
A focused pilot can produce initial task maps, technology assessments and role hypotheses within a few weeks. Overall duration and depth depend on the number of roles, available data, functional scope and scenario detail. The modular approach can scale across functions.
What do we receive at the end?
Typical outputs include task maps; digitisation, automation and GenAI assessment matrices; redesigned roles with future tasks and skill profiles; capacity and staffing scenarios; reskilling and redeployment options; and a prioritised implementation and governance roadmap.
How do you handle works councils, data protection and AI governance?
Employee representation, data protection, responsibilities and AI governance are considered early. We make the modelling transparent and distinguish opportunity analysis from later deployment decisions. Specific AI systems used in employment may require a separate legal or regulatory assessment. Harbinger does not replace legal advice.
Is AI Impact Modelling only relevant if we already use Copilot or Gemini?
No. It is often most valuable before or alongside deployment, helping clarify which work should change, which use cases have business value and what follows for roles and skills. This gives technology decisions a clear workforce purpose.
What is the best way to get started?
In an initial, no-obligation conversation, we clarify your decision, the functions and roles involved, and a proportionate scope. One clearly defined function often makes a useful pilot that can be extended later.
Further reading
Explore AI, roles and Strategic Workforce Planning
For decision-makers who want to understand how AI affects tasks, job profiles, workforce planning and leadership. Linked articles are in German.
Job Redesign & Role Profiles
How is AI changing role profiles? The quiet shift begins
Why the greatest changes happen within task profiles.
Workforce Planning & AI
Strategic Workforce Planning and AI: what changes?
How headcount planning becomes a dynamic model of tasks, skills and capacity.
Copilot, Gemini & Organisation
Copilot & Gemini in business: what AI tools really change
Why deployment needs to be connected to roles and the way work is organised.
Leadership Augmentation
AI for leaders: understand LLMs and strengthen leadership
Where human accountability remains and how leadership can become more effective with AI.
Next step
Before deciding on AI, headcount or future roles, get clear on the work
In an initial conversation, we discuss your decision, a suitable starting function, the data and roles required, and a pilot scope that produces useful evidence for management, HR and Finance.