Workforce software giant Workday has switched on its overhauled learning platform for customers around the world, and the headline feature is an artificial intelligence tutor built by Sana, the company it snapped up earlier this year. The move, reported by Australian Mining, lands the American company squarely in one of the hottest and most contested corners of enterprise technology: how staff actually learn on the job.
For readers who do not track the human resources software beat, a quick primer helps. Workday sits behind the scenes at a large slice of the world’s big employers, running the systems that handle payroll, hiring, performance reviews and, increasingly, training. Learning has long been the least glamorous piece of that stack, typically a library of compliance videos and slide decks that employees click through and forget. Workday’s pitch is that an AI tutor changes the shape of that experience, turning a static catalogue into something closer to a personal coach that answers questions, drafts practice scenarios and adapts to the person in front of it.
What has actually launched
The new product folds in technology from Sana, a Swedish AI company Workday moved to acquire during 2025 in a deal widely reported to be worth well over a billion US dollars. Sana made its name building AI-native tools for knowledge work and training, and its tutor is now positioned as the centrepiece of Workday Learning rather than a bolted-on extra. The idea is that the tutor draws on a company’s own material, its policies, procedures and internal knowledge, so that answers reflect how a particular business actually operates rather than generic advice scraped from the open internet.
Going global is the meaningful part of the announcement. Enterprise software of this kind is usually rolled out in stages, with early access confined to a handful of markets before a wider release. A worldwide availability date signals that Workday believes the tool is ready for regulated, multinational customers, the sort of organisations that cannot afford a training system that hallucinates a safety procedure or misquotes an award.
Two ways to read it
Supporters of the approach argue that this is exactly where AI earns its keep. Training is expensive, uneven and hard to scale, particularly for large workforces spread across sites and shifts. An AI tutor that can answer a question at two in the morning on a remote site, in the worker’s own language, is a genuinely useful thing rather than a novelty. Backers also point out that Workday already holds the employment data needed to make learning relevant, so weaving a tutor into that context is a smaller leap than starting from scratch.
Sceptics are less convinced, and their concerns are not trivial. Corporate learning teams have watched a parade of technology fads promise personalisation and deliver dashboards. There is also the harder question of accuracy. A tutor that confidently gives a wrong answer about a hazardous task, a financial control or a legal obligation is worse than no tutor at all, because it carries the authority of an official system. Critics want to see how Workday handles the boundary between what the AI knows and what it merely guesses, and how easily a business can audit the answers its staff are being fed. There is a labour dimension too, with unions and learning professionals wary that AI tutors become a quiet excuse to thin out human trainers and mentors.
Why this matters in Australia
The choice of an Australian mining publication to carry the news is a telling one, because resources is precisely the kind of sector where the promise and the peril collide. Australian miners run large, dispersed and safety-critical workforces, often on fly-in fly-out rosters across the Pilbara, the Bowen Basin and beyond. Skills shortages have dogged the industry for years, and the pressure to bring new workers up to speed quickly, without cutting corners on safety, is relentless. An AI tutor that can shorten the path from induction to competence is an obvious drawcard for operators wrestling with those constraints.
Workday has a substantial local footprint, counting Australian employers across resources, banking, retail and government among its customers, so a global switch-on reaches this market directly rather than in some distant future release. That immediacy raises the stakes for local decision-makers. Australian workplace law is dense, with modern awards, enterprise agreements and state-based safety regimes that differ from the American context in which much of this technology is designed. A tutor trained on a company’s own documents still needs those documents to be current and correct, and it needs guardrails that reflect Australian obligations rather than assuming a United States default.
Data is the other Australian sticking point. Employee training records, performance information and internal know-how are sensitive, and the country’s privacy reforms have sharpened scrutiny of how such information is stored, processed and fed into AI systems. Boards and chief people officers considering the tool will want clarity on where the data sits, whether it is used to train broader models, and how it squares with the Privacy Act and sector-specific rules. Those questions are not unique to Workday, but they are the ones that will decide how quickly local adoption moves from pilot to production.
What happens next
The near-term test is straightforward: does the tutor survive contact with real workforces? Enterprise buyers will run limited trials, measure whether staff actually use the tool and whether its answers hold up, and compare it against rival offerings from the likes of SAP SuccessFactors, Cornerstone and a crowd of specialist AI-learning startups. Workday will need to show hard results on completion rates, time to competence and error rates, not just polished demonstrations, before finance chiefs sign off on wider rollouts.
For Australian organisations, the sensible path is neither breathless adoption nor reflexive caution. The upside for training-heavy, safety-conscious industries is real, but so is the downside of an AI system that gets a critical answer wrong. The employers who benefit will be the ones that treat the tutor as a tool to be tested and governed rather than a magic fix, insisting on human oversight, clear accountability and Australian-appropriate content from the outset. On that score, the global launch is less an ending than the start of a long assessment.
Sources: Australian Mining.
