AI’s impact on work: hard truths & fresh lessons for leaders.

One of the most alarming developments is evidence that AI may already be eroding entry-level opportunities for young professionals. A comprehensive study of payroll data finds that since late 2022 – when tools like ChatGPT first went mainstream – early-career employees in highly AI-exposed jobs have seen significant employment declines. In particular, entry-level positions in fields such as software development, marketing, and customer support have fallen by roughly 16–20% for workers aged 22 to 25 ([1]).

Compounding this challenge, some employers are responding to the shrinking junior talent pipeline by narrowing their recruitment efforts. A 2025 survey found that 26% of large companies were recruiting primarily from a short list of elite universities (up from 17% in 2022) ([2]) – a retreat from the “talent is everywhere” hiring approach of recent years. One practical reason, recruiters say, is that AI-written résumés have made many candidates look indistinguishable from each other, pushing companies to fall back on familiar signals like degree prestige to identify top talent ([3]).

Why are junior staff bearing the brunt? Simply put, generative AI is now adept at automating the routine, “textbook” tasks that entry-level workers typically do – from writing basic code to drafting marketing plans. This is knowledge that college graduates have just learned, and AI can replicate much of it almost instantly. As study co-author Erik Brynjolfsson explained, “These large language models are trained on books, articles and written material found on the internet and elsewhere… That’s the kind of book learning that a lot of people get at universities before they enter the job market, so there is a lot of overlap between these LLMs and the knowledge young people have” ([4]). In contrast, experienced employees bring deeper expertise, tacit knowledge, and organizational insight that AI cannot easily mimic – which helps explain why their employment prospects have so far remained more resilient.

Some experts warn that aggressively automating entry-level roles could be a costly long-term mistake. MIT’s Andrew McAfee cautions that by removing too many junior roles, companies risk undercutting their future leadership pipeline ([5]). After all, today’s new hires and trainees are tomorrow’s managers and innovators – but only if they have opportunities to develop skills on the job. Recognizing this, forward-thinking organizations are taking action to bridge the AI skills gap instead of simply shrinking their intake. For example, design platform Canva recently gave its 5,000 employees a full week off from regular work to focus exclusively on learning AI skills – a “bold experiment in team development,” according to co-founder Cameron Adams ([6]). Similarly, companies like IBM, Microsoft, and Amazon have launched massive internal AI upskilling programs, and new roles such as AI trainers and prompt engineers are being created to ensure human talent remains at the core of the AI revolution. By investing in employee growth and adjusting career pathways, leaders can help mitigate the early-career crunch and cultivate the next generation of AI-ready talent.

As AI tools weave into everyday workflows, they are surfacing new cultural challenges in the workplace. One striking example is a recent experiment that revealed a gender-based double standard when employees use AI. Researcher Zehra Chatoo had two job seekers – one woman and one man – submit identical résumés generated with AI (differing only in the names “Emily” and “James”). The result was eye-opening: reviewers were 22% more likely to question the woman’s credibility and were twice as likely to doubt her competence, despite her qualifications being identical ([1]). Gen Z male reviewers were the most skeptical of the female candidate’s AI-assisted résumé, describing it as 'weak' 3.5 times more often than the identical male version. In fact, James’s résumé received a 97% approval rating, compared to just 76% for Emily’s ([2]). This kind of bias could discourage women from using AI tools – or from crediting those tools – which in turn might hamper their productivity or visibility. It’s a reminder that technology doesn’t automatically eliminate bias; sometimes it uncovers biases we didn’t know we had.

At the same time, many employees remain uneasy about the growing role of AI in their working lives. Nearly two-thirds of Americans (64%) believe AI will lead to fewer jobs in the long run ([3]), and even among those whose jobs are not immediately threatened, there’s a “trust gap” emerging. One recent global survey found that AI adoption in the workplace is now outpacing employees’ confidence in the technology’s benefits ([4]). In practice, this anxiety can translate into resistance. For instance, faced with an automated video interview or chatbot screener, a substantial share of job seekers are simply opting out: roughly 38% have already withdrawn from a hiring process because it required an AI-driven interview, and another 12% say they would likely do the same if asked ([5]). Of those who went through AI-based interviews, about half say the experience did not result in a job offer ([6]) – a discouraging outcome that can fuel further distrust. Clearly, if organizations push AI tools on employees or candidates without addressing their concerns, they risk attrition and hiring challenges.

These sentiments are not going unheard. Employee unions and regulators are increasingly vocal about the need to ensure a "just" AI-driven transition in the workplace. Labor groups are pressing employers to be transparent about how algorithms are used in decisions affecting jobs, pay, and working conditions. In the US, policymakers are taking note: the White House recently called on Congress to establish a legal framework to protect individuals from the unauthorized use of AI-generated replicas of their voice or likeness ([7]) – a response to union outcry in industries like media and entertainment over deepfakes and digital impersonation. The overall message is that managing the human side of AI is now a leadership issue. Companies that treat AI adoption purely as a technical upgrade, without regard for fairness, transparency, and employee involvement, may face growing backlash. By contrast, those who proactively engage their workforce, address bias, and build trust stand a far better chance of achieving sustainable, people-positive AI transformation.

With AI now rewriting business processes, leadership itself must evolve. Many organizations are appointing new roles and rethinking team structures to steer this transformation. A new global survey of 2,000 CEOs by IBM found that 76% of companies have created a Chief AI Officer (CAIO) position in 2026, up from just 26% in 2025 ([1]). Nearly two-thirds of these CEOs also say they are comfortable using AI-generated insights to inform major decisions at the leadership level ([2]). Boards are asking tougher questions about AI strategy, and companies increasingly realize that guiding AI deployment can’t be left to IT departments alone – it requires engagement from the C-suite and directors.

However, pouring money into intelligent software alone is not a silver bullet. A sobering analysis by Gartner recently found that only one in fifty AI projects truly succeeds in delivering transformative value ([3]). Often, the difference between organizations that thrive with AI and those that struggle lies in the human element of integration. Businesses that invest in reimagining workflows, upskilling employees, and aligning job roles to complement AI are seeing better returns. Indeed, IBM’s study notes that companies embracing an "AI-first" operating model – even restructuring leadership teams around AI – have scaled 10% more AI initiatives across their enterprises than their peers ([4]).

Forward-thinking leaders are also adapting how they measure performance in an AI-enabled workplace. Some firms have begun rewarding employees for integrating AI into their work. For example, Google has reportedly started to include AI tool usage as a factor in performance evaluations for its engineers ([5]). On the other hand, not everyone believes in tracking individual AI metrics. Goldman Sachs’s tech chief Marco Argenti argued that monitoring each employee’s AI usage is misguided – "like looking at only one player on the field" – and instead measures how fast teams move from idea to product as the key indicator of success ([6]) ([7]). These contrasting approaches underscore that there’s no one-size-fits-all playbook. What’s clear is that successful leadership in the AI era means fostering a culture of innovation and learning, where employees are empowered (and incentivized) to use AI ethically and effectively.

As IBM’s Vice Chairman Gary Cohn put it, the companies that win will be those that treat AI "not as a layer of technology, but as a new operating model" that accelerates decision-making and dissolves silos ([8]). Leading an organization through AI-driven change is a test of adaptability and vision. By pairing technological investment with purposeful role redesign, continuous learning opportunities, and an empathetic change management strategy, executives can unlock AI’s benefits while bringing their people along on the journey.