These companies are actually upskilling their workers for AI - here's how they do it
Major companies are building structured AI training programs for their entire workforces, from entry-level hires to senior leadership, rather than replacing workers with automation. A new report reveals that only 1 percent of companies believe they have fully realized their AI in...
According to ZDNet AI's latest coverage, business leaders across industries are making a deliberate bet that upskilling existing employees is smarter than chasing a shrinking pool of experienced AI specialists. The story draws on research from Comparably, McKinsey, IBM, and the National Academies of Sciences, Engineering, and Medicine to map out what actual corporate AI training looks like right now, and more importantly, what is actually working.
Why This Matters
The gap between AI spending and AI results is not a technology problem, it is a people problem. McKinsey's January 2025 report "Superagency in the Workplace" found that virtually every large company is investing in AI, but only 1 percent believe they have reached anything close to full potential. That number is damning. Companies that figure out workforce development before their competitors do are building a durable advantage that cannot be copied overnight by buying more software licenses.
Daily briefing from 50+ sources. Free, 5-minute read.
The Full Story
The core finding here is deceptively simple: buying AI tools is easy, building a workforce that knows how to use them is hard. Companies like LexisNexis Legal and Professional are not just rolling out software and calling it a day. They are running enterprise-wide literacy programs designed to give every employee, regardless of job function or technical background, a working understanding of what AI can and cannot do. The philosophy driving this is less about turning accountants into prompt engineers and more about making sure every person in the organization can make informed decisions about when to trust an AI output and when to question . What makes the current wave of corporate AI training different from past technology rollouts is the explicit rejection of one-size-fits-all approaches. Sales teams are being trained on AI-assisted pipeline analysis and customer relationship tools. HR departments are learning how AI supports recruitment screening and engagement measurement. Finance teams are developing skills around forecasting models and anomaly detection. The training is built around the actual work each group does, which sounds obvious but represents a meaningful departure from the generic software training webinars most employees have sat through and immediately forgotten.
Governance frameworks are getting as much attention as the technical training itself. These are internal rulebooks that answer the questions employees inevitably have: What data can I feed into this tool? What do I do when the AI gives me a weird result? Who is accountable when an AI-assisted decision goes wrong? Companies building these frameworks are essentially pre-answering the questions that cause hesitation and misuse, which speeds up adoption without sacrificing responsibility. Sandboxed experimentation environments, where employees can test tools without any risk to live systems or customer data, are also proving effective because hands-on experience accelerates learning far faster than classroom instruction alone.
Perhaps the most counterintuitive finding in this reporting is that companies are not pulling back on entry-level hiring. The early narrative around AI automation predicted that junior roles would be the first to disappear. Instead, business leaders are describing a deliberate strategy of hiring early-career workers and pairing that investment with structured development programs that build AI fluency from day one. The logic is straightforward: it is cheaper and culturally healthier to develop talent internally than to compete for the small pool of workers who already have deep AI expertise.
The National Academies of Sciences, Engineering, and Medicine published research in December 2025 specifically addressing workforce retraining for the AI era, which signals that this has moved beyond a corporate HR conversation into national policy territory. When a body of that stature publishes on a workforce topic, it is because the underlying disruption is real and large enough to require systematic societal responses. IBM has similarly repositioned AI upskilling as a core business priority rather than a peripheral training function, a framing shift that carries real budget and leadership attention with .
Key Details
- McKinsey's January 2025 report found only 1 percent of companies believe they have fully realized their AI investment potential.
- The National Academies of Sciences, Engineering, and Medicine published workforce retraining research specifically addressing AI in December 2025.
- LexisNexis Legal and Professional is among the named companies running enterprise-wide AI literacy programs.
- IBM has publicly repositioned AI upskilling from an HR function to a core business priority.
- Comparably research documents successful programs operating across at least 4 distinct dimensions: enterprise literacy, role-specific training, governance frameworks, and sandboxed experimentation.
- Companies are continuing to hire for entry-level roles despite predictions that AI would eliminate early-career positions.
What's Next
Watch for consulting firms like McKinsey and IBM to expand their AI workforce enablement service lines significantly through 2025 and 2026, as corporate demand for external curriculum design and governance framework development is already strong enough to constitute its own market. Universities and community colleges are also modifying programs to embed AI literacy at the curriculum level, which means employers will face a lower baseline training burden within three to five years as graduates arrive with foundational AI knowledge already intact. The competitive gap between companies that build this infrastructure now and those that delay will be measurable in productivity metrics by the end of 2025.
How This Compares
Compare this movement to what happened with cloud computing adoption between 2012 and 2016. Companies that invested early in cloud training and internal certification programs came out of that period with structural speed advantages that competitors could not close quickly. The companies profiled here are making an analogous bet on AI, and the McKinsey data showing a 99 percent failure rate on realizing AI potential suggests the window for building that advantage is still wide open. Most organizations are still fumbling with tool deployment while the leaders are already on workforce capability.
The approach also contrasts sharply with what some earlier automation waves looked like. Robotic process automation adoption in the late 2010s frequently proceeded without meaningful employee development, which created resentment, low adoption rates, and a wave of expensive re-implementations. The explicit emphasis on responsible use, ethical guidelines, and employee agency in current programs reflects hard lessons from that era. Organizations are building AI programs that employees buy into rather than fear, which is a structural improvement over past practice.
IBM's own public positioning on this topic is worth noting separately from the companies it advises. A technology company of IBM's scale treating workforce development as a core business function rather than a support activity sends a signal to every firm watching. When the companies building AI tools say their own employees need structured upskilling, smaller organizations have no credible argument for skipping that investment.
FAQ
Q: Are companies actually still hiring entry-level workers despite AI? A: Yes, according to this reporting, business leaders are continuing to hire for entry-level positions as a deliberate strategy. Rather than replacing junior roles with automation, companies are pairing early-career hires with structured AI training programs to develop talent internally over time.
Q: What does an AI upskilling program actually look like day to day? A: Programs typically include enterprise-wide literacy training for all employees, role-specific instruction tailored to functions like sales or finance, internal governance guidelines on responsible AI use, and sandboxed environments where workers can experiment with AI tools without risk. Check out AI Agents Daily guides for practical walkthroughs of how these tools work.
Q: Why are companies bringing in outside consultants for AI training? A: Firms like McKinsey and IBM have expanded service offerings around AI workforce enablement because designing effective curricula, skills assessments, and governance frameworks is genuinely complex work that most HR departments are not equipped to build from scratch. The external market growing around this problem is evidence that the challenge is both real and widespread.
The companies investing in workforce capability right now are not doing it out of altruism. They are doing it because the data, including McKinsey's finding that 99 percent of organizations are leaving AI value on the table, makes the business case impossible to ignore. Follow related AI news as more companies publish results from these programs over the next 12 months. Subscribe to the AI Agents Daily weekly newsletter for daily updates on AI agents, tools, and automation.
Get stories like this daily
Free briefing. Curated from 50+ sources. 5-minute read every morning.
