IBM’s deployment of artificial intelligence to automate hundreds of human resources roles has produced an unexpected outcome: the technology giant’s total workforce has grown, not shrunk, as efficiency gains funded strategic hiring in software engineering, sales and marketing.
The development challenges simplistic narratives about AI’s impact on employment, demonstrating that automation’s effect on organisations extends far beyond mere headcount reduction. Instead, IBM’s experience suggests AI acts as a catalyst for workforce transformation rather than wholesale replacement.
In May 2023, IBM chief executive Arvind Krishna told Bloomberg the company expected to pause hiring for roles that could potentially be automated over a five-year period. The statement, focused primarily on back-office functions in human resources, sparked widespread attention as one of the first major acknowledgements by a technology leader that AI would directly affect corporate employment decisions.
Two years later, Krishna confirmed to The Wall Street Journal that IBM had used AI to replace the work of several hundred HR employees. The company’s proprietary system, AskHR, now handles approximately 94% of routine HR enquiries, from payroll questions to holiday requests and benefits documentation. In 2024 alone, the platform processed more than 11.5 million internal interactions.
The automation delivered measurable efficiency gains. IBM reported £3.5 billion in productivity improvements across more than 70 job roles. The company’s internal satisfaction metrics told an equally compelling story: AskHR’s Net Promoter Score climbed from –35 to +74, suggesting employees found the AI system more responsive than previous arrangements.
Yet IBM’s overall employment grew following the automation initiative. Krishna explained that efficiencies gained through AI freed resources for investment in areas demanding human expertise. “Our total employment has actually gone up, because what it does is it gives you more investment to put into other areas,” Krishna told The Wall Street Journal.
Those areas include software engineering, marketing and sales – roles Krishna characterised as requiring “critical thinking” and activities that “face up or against other humans, as opposed to just doing rote process work”. The company redirected savings from automated functions toward hiring specialists in technical, commercial and client-facing positions.
The outcome reflects broader patterns emerging across organisations implementing AI. Whilst automation eliminates certain tasks, it simultaneously creates demand for different capabilities. McKinsey research suggests that by 2030, up to 30% of current work hours in the United States could be automated, yet this technological shift will reshape rather than simply reduce workforce requirements.
IBM’s experience also highlighted automation’s limitations. Despite AskHR’s success rate, approximately 6% of HR enquiries still require human intervention. These cases typically involve sensitive workplace issues, ethical considerations or emotionally complex situations where empathy and discretion prove essential.
The distinction matters for workforce strategy. Organisations cannot simply automate their way to reduced headcount without consequences. IBM discovered that whilst AI excels at handling predictable, repeatable tasks, it struggles with ambiguity, context-dependent decision-making and situations requiring genuine human judgement.
Other technology companies have encountered similar realities. Language-learning platform Duolingo experimented with replacing tutors with chatbots, only to find user satisfaction declined. Several customer service organisations made comparable moves before quietly rehiring staff when performance metrics faltered.
Research from the World Economic Forum projects that 92 million jobs globally could be displaced by automation by 2030. However, the same analysis forecasts creation of 170 million new roles, yielding a net increase of 78 million positions. Many of these emerging roles – AI trainers, prompt engineers, ethics officers – did not exist five years ago.
For HR professionals and senior leaders, IBM’s experience offers several strategic insights. First, automation proves most effective when organisations view it as enabling workforce evolution rather than simple cost reduction. IBM’s approach – using efficiency gains to fund hiring in growth areas – demonstrates how automation can expand rather than contract overall employment.
Second, successful AI implementation requires careful workforce planning. IBM maintained its focus on roles requiring creativity, strategic thinking and interpersonal skills – capabilities that current AI systems cannot replicate. This meant identifying which functions genuinely benefited from automation versus those requiring sustained human involvement.
Third, the “human touch” remains crucial for complex scenarios. IBM’s 6% of queries requiring human assistance may seem small, yet those cases often represent the most sensitive and consequential interactions. Organisations that eliminate human capacity for handling such situations risk service quality deterioration and employee satisfaction decline.
The transformation also underscores the importance of transparent communication. When Krishna initially discussed potential automation in 2023, the statement generated significant concern. Two years later, the reality – several hundred roles automated, overall workforce growth, strategic reinvestment in high-value positions – presents a more nuanced picture than early projections suggested.
IBM’s workforce now exceeds 270,000 employees globally. The company’s experience demonstrates that even substantial automation need not result in net employment reduction when organisations approach AI strategically. The technology’s promise lies not in replacing human workers wholesale, but in enabling organisations to deploy human capital more effectively.
As Krishna noted, automation “allowed us to invest more in areas that need human creativity and interaction”. This balance between technological efficiency and human capability increasingly defines competitive advantage in AI-augmented organisations.
For business leaders navigating similar transformations, IBM’s journey provides a framework: identify genuinely automatable functions, measure outcomes rigorously, reinvest savings strategically, and maintain human capacity for complex decision-making. The alternative – viewing AI purely through a cost-reduction lens – risks missing automation’s fuller potential whilst creating workforce instability.
The broader implication extends beyond individual organisations. As AI capabilities expand, economies face the challenge of managing workforce transitions at scale. IBM’s experience suggests that whilst specific roles disappear, strategic organisations can create net employment growth by channelling automation gains toward emerging needs.
This pattern – technology eliminating certain jobs whilst creating others requiring different skills – has characterised every major technological transition. The telephone operators of 1920 became the computer programmers of 1970. Today’s automated HR tasks may fund tomorrow’s AI implementation specialists, customer experience designers and human-machine collaboration experts.
The critical variable remains whether organisations, workers and policymakers respond strategically to these shifts. IBM’s approach demonstrates one viable path: embrace automation for suitable functions, invest in human capabilities for complex work, and maintain focus on strategic outcomes rather than simple headcount reduction.
