Rethinking Pay in a World Where GenAI Redefines Value

Most compensation systems are still designed around jobs which are defined, evaluated, and priced based on role expectations. But Generative AI has the potential to change how work is done at a more granular level: the task level.

This shift creates a fundamental question:

If the job title stays the same, but the tasks and required skills change, how should pay adjust?

To answer that, we need to move beyond traditional job-based frameworks and look more closely at where value is being created, and by whom (or what).

The Four Task Types That Reshape Value

In a recent study on GenAI on the workplace, Deloitte categorizes tasks into four groups. Each one has different implications for compensation:

1. Automated Tasks: Machines do these best

   Example: Summarizing FAQs or generating standard reports.

2. Augmented Tasks: Humans + AI create better results together

   Example: Analysts using GenAI to extract insights from complex data sets.

3. Human-only Tasks: Best done by people using judgment, empathy, or contextual knowledge

   Example: Managing conflict or leading teams.

4. New AI-related Tasks: Emerging work, such as writing prompts, validating outputs, or managing AI risks

   Example: Prompt engineers or ethics reviewers.

   
   

Today’s compensation systems don’t reflect this variety. That gap is likely to grow as GenAI adoption increases.

From Jobs to Skills to Value Contribution

If work is increasingly made up of discrete tasks and, in turn, those tasks demand different combinations of human skill and machine input, then pay needs to be better aligned to:

• The type of tasks performed

• The level of skill required

• The impact of that contribution

This doesn’t mean abandoning job structures entirely. It means revisiting how roles are evaluated:

• Are they being rewarded for outcomes that now depend more on GenAI than on individual input?

• Are we recognizing new skill areas like GenAI fluency?

• Are we differentiating based on employees’ ability to apply AI in ways that improve performance?

The Role of HR Analytics: Measuring What Matters Now

HR and rewards leaders will need better visibility into:

• How tasks are shifting within roles

• Where GenAI is adding productivity

• Who is learning to use GenAI effectively

This is where HR analytics becomes essential. Task-level analysis, especially before-and-after GenAI integration, can reveal:

• Where the greatest value is being created

• Which skills are most predictive of performance

• Where to target pay differentiation or bonus eligibility

A recent MIT study showed a 35% productivity increase for less experienced customer service agents using GenAI. This kind of gain may warrant changes in performance incentives, career acceleration, or differentiated training investments.

A Case for Skill-Based Pay, With Practical Constraints

Skill-based pay is not new, but GenAI may give it new urgency. As skills, and not just jobs, become the main drivers of value, organizations may need to:

• Expand their use of broadbanding and skill premiums

• Create modular reward systems that reflect fluid task mixes

• Incorporate AI fluency and GenAI-specific capabilities into reward criteria

But there are limits. Most organizations are not ready for fully skill-based pay. What may be more realistic is a hybrid approach:

• Maintain job structures for governance and benchmarking

• Introduce task- or capability-linked incentives for certain roles or business units

• Use analytics to inform pay progression based on evolving skill sets

Ownership, IP, and Contribution: The Emerging Reward Frontier

New questions are also emerging around ownership of AI-enabled output. If an employee creates something or makes a decision with GenAI, who owns it? How is the value of that work recognized?

The Shutterstock model (compensating contributors whose content trains GenAI models) is one early example. While not widely adopted, it signals that rewarding digital contribution may evolve beyond traditional performance metrics. 

This is an area that could raise legal and ethical challenges, but these models are likely to become more common as content creation, product development, and even decision-making become partially AI-generated.

Avoiding Unintended Consequences

If reward systems don’t keep pace with how value is created, several risks arise:

• Employees may feel under-recognized for the work they now perform differently or more efficiently.

• Skill gaps may widen if there’s no incentive to develop GenAI fluency.

• Pay equity concerns may increase if new task configurations benefit only a subset of roles or workers.

Rewards leaders should also be aware that roles requiring AI fluency are beginning to command market premiums. This is especially true in hybrid roles (e.g., HR or marketing professionals who can effectively prompt and validate GenAI outputs).

Moving Forward: A Practical Agenda

To prepare compensation for the GenAI era, reward leaders can start by:

1. Auditing current roles for task shift. Where is GenAI changing what people actually do?

2. Linking pay and performance analytics to real outputs and impact, not just roles.

3. Piloting AI-relevant skill premiums in high-impact areas.

4. Updating incentive programs to reflect AI-enabled productivity.

5. Partnering with HR analytics to measure outcomes in more granular ways.

Conclusion

Pay systems are built to reflect how value is created. GenAI changes that equation. Some tasks are now faster, some require new skills, and others need to be re-evaluated entirely.

HR leaders, compensation professionals, and business executives should work together to ensure their reward models reflect this new reality. Getting ahead of this shift will improve fairness and engagement, and will help organizations retain the talent that learns how to work productively with GenAI.

The work is changing. Pay must change with it.