Click here for the original LinkedIn article
This article is based on and integrates seamlessly with Alexis Fink and my September 3 webinar on Driving Organizational Change with Data in the Age of AI, and the virtual workshop Maura Stevenson, PhD, Alexis Fink and I will be leading starting October 8 by the same name (Driving Organizational Change with Data in the Age of AI)
AI is going – and has already started – to transform how work is done within organizations. Anywhere people are asked to weigh in on decisions has the potential to be disrupted by AI. And in these early days, it’s really hard to say for certain what the limitations of AI will be in most cases.
The HR function is no exception. HR historically is organized around key people management processes such as recruiting, compensation, talent management, and learning and development. Each of these has tremendous potential to be transformed by AI, creating greater efficiencies and hopefully more effective outcomes. But if you take a step back from an HR process lens, and consider the disciplines that underpin strategic HR, the impacts of AI are less certain.
AI will quickly transform HR tasks, but not HR expertise
All the discussion about whether AI will augment or replace humans focuses on task-level views of what work is. And at that level, the potential and rate of change can seem both breathtaking and incredibly broad. Yet, zoom out to the level of HR expertise more generally and the situation doesn’t seem as dire.
What makes someone expert is a combination of tactical knowledge about work tasks, and broader conceptual knowledge about an entire body of work. For example, a solitary recruiter may be expert at certain recruitment tasks which can be easily automated, such as finding and engaging with candidates from a particular source like online or university campuses. AI could easily disrupt that work by providing an always-on, limitless resource to match potential candidates with job opportunities.
The work of an organization’s recruitment center of expertise (COE), in contrast, will always require humans responsible for the end-to-end processes that cover the full range of tasks that start with business partnership on priorities, nurturing recruiting channels and honing selection processes that are relevant to the organization and the roles before jobs are even posted, and then running the entire recruitment process through to a successfully onboarded candidate. AI can and will transform many of those tasks within the end-to-end processes, so the people will need to grow their skill sets to take on more strategic aspects of the work; but the most successful companies will keep humans at the center of and leading the integrated work of recruitment through onboarding.
The same applies to each core HR function: performance management, compensation, learning and development, etc. In each case, the work itself may be greatly transformed by AI. Yet human expertise in the functional discipline will always be at the center of the work of HR, no matter how advanced AI becomes. While AI can drive enormous efficiency and precision in task execution, human judgment and expertise are essential to making effective strategy and prioritization decisions.
AI will struggle to transform how work is integrated across HR expertise domains
Zoom out now to the HR strategy and operating model. Two expertise domains are central to integrating work across HR’s different processes and functional expertise: organizational development (OD) and people analytics.
The work of OD is the responsibility of the HR leadership team, and HR business partners (HRBPs). Integration across all HR and talent processes is the foundation of what they do, collaborating across the function and across the business to address critical talent issues. That integration is always bespoke, and requires direct interaction with humans in the system to gauge their perspectives and intentions, concoct workable solutions, and implement them – all tasks that AI cannot disrupt a human from doing, though it could augment at the edges of the work.
At first glance, people analytics may not appear to be as cross-functional as OD, but, done right, it has to address the full range of issues that impact the business (see Strategic Analytics for details), using perspectives and tools from a wide range of disciplines (see Workforce Analytics for details). Similar to OD, this work is bespoke, and requires direct interaction with humans in the system to understand the issues being addressed, conduct analyses that are both quantitative and qualitative, and influence how the results drive organizational change.
The reporting and analysis tasks at the heart of people analytics are ripe for disrupting with AI, but those are the only ones at risk for major transformation. AI can build dashboards to support queries about data, a task which currently is done much more manually by analytics professionals. Yet good people analytics is way more than dashboards, and includes addressing the different mental models stakeholders bring to the table – a task that AI cannot automate. For the rest of the job, just like OD, AI should help augment at the edges of the work, not the core.
As Maura Stevenson, Alexis Fink and I have argued previously (Are OD and Analytics Twins Separated at Birth?), the disciplines of OD and analytics share a lot in common … when done right. Yet most HR functions struggle to effectively integrate the work of OD and analytics because of deep-seated perceptions that OD is more of a “soft,” right brain discipline, while analytics is more of a “hard” left brain discipline.
AI has tremendous potential to transform a lot of the work of these two essential groups; but it will never fully disrupt them. And when it comes to integrating across OD and analytics, AI will struggle to make even a tiny impact.
AI won’t solve those foundational challenges … though it will accelerate the production of analyses that can be used to inform strategic HR decision making. Which means the risk will shift more into the “garbage in, garbage out” challenge: when the barriers to conducting sophisticated analysis are greatly removed by AI advances, the ability to interpret and act on them appropriately becomes even more essential. Which highlights the challenge of integrating analysis insights into organizational design and change, an area where AI cannot replace humans.
Resources cited here:
Alexis Fink and my webinar on September 3: Driving Organizational Change with Data in the Age of AI
Maura Stevenson, PhD, Alexis Fink and my virtual workshop, which starts October 8: Driving Organizational Change with Data in the Age of AI
Workforce Analytics: A Global Perspective, co-authored with Martin Edwards, Dana Minbaeva and Mark Huselid
Strategic Analytics: Advancing Strategy Execution and Organizational Effectiveness