The conversation around AI in HR has become dominated by a very specific image: a knowledge worker at a laptop, using an intelligent assistant to draft job descriptions, summarise CVs, or pull analytics from an HRIS dashboard.
It is a seductive image. It is also profoundly incomplete.
Roughly 2.7 billion people go to work every day without ever opening a laptop. They operate manufacturing lines, stock supermarket shelves, deliver parcels, care for elderly patients, build infrastructure, and keep supply chains moving. They are the deskless workforce — and they are, by a considerable margin, the largest segment of the global labour market.
For this population, the standard pitch for “AI for HR” is not just irrelevant. It is invisibilising. It assumes a model of work, and a model of HR, that simply does not apply.
This piece is about why that matters, what a genuinely AI-first approach to HR looks like when designed around frontline and deskless workers, and why getting this right is arguably the most important challenge in the future of work. It is also about what it means — concretely — to help companies work smarter, starting with the people everyone else forgot to design for.
The gap nobody talks about
Here is a striking data point: despite representing nearly 80% of the global workforce, deskless workers receive a disproportionately small share of enterprise technology investment. Most HR platforms were built for, and continue to prioritise, the 20% who sit at desks.
The practical consequences are significant. A logistics worker who needs to report a near-miss incident, understand a change to their shift pattern, or access their payslip faces a different UX challenge than a desk-based employee. They may be working in loud environments, under time pressure, with limited connectivity, on shared devices, or in their second or third language. The friction that a knowledgeable worker barely notices becomes, for a frontline worker, a genuine barrier to engagement.
This is not a niche problem. It is a structural one — and it has meaningful downstream effects on retention, compliance, safety, and operational performance. Solving it is exactly what working smarter means for the majority of the world’s workforce.
What “AI-First” actually means for this cohort
The phrase “AI-first” is used so broadly that it risks losing all meaning. In the context of HR technology for frontline workers, we would define it precisely: an AI-first approach means that intelligence is embedded into the interaction layer itself, not bolted on as an add-on to an existing workflow.
This distinction is critical.
Most HR platforms add AI as a feature — a chatbot alongside a portal, an analytics module layered over a database. An AI-first platform, by contrast, is one where the natural language interface is the primary interaction model. Where asking a question and getting an accurate, personalised answer in under three seconds is the default, not the premium add-on.
For the deskless workforce, this shift is transformative for three reasons:
- It eliminates navigational complexity. Traditional HR portals require workers to know where to look. AI-first interfaces require only that workers know what they need. A warehouse operative who wants to know whether they are eligible for overtime this weekend does not need to navigate to a policies tab, find the correct document, and parse dense legal language. They ask. They get an answer. The system handles the rest.
- It removes language as a barrier. Multilingual workforces are the norm, not the exception, in frontline industries. An AI-first HR interface that operates naturally across languages — not through clunky drop-down locale selectors, but through conversational intelligence — fundamentally changes access. A worker asking in Spanish, receiving a response in Spanish, without having to toggle anything, is a different experience entirely from one that requires English proficiency to navigate.
- It scales personalisation without scaling cost. Every frontline worker has a different employment contract, schedule, location, tenure, and benefit entitlement. Traditionally, getting personalised answers to individual questions required either a knowledgeable HR business partner or a very well-configured self-service portal. Neither scales to tens of thousands of workers spread across dozens of sites. AI does. That is what smarter work looks like at scale — not harder, not louder, just more intelligent.
The nuances that most vendors miss
There is a meaningful difference between building AI for HR and building AI into HR. The former treats HR as the customer; the latter treats HR as the service channel through which workers are actually served.
When we speak to HR leaders in retail, logistics, healthcare, and manufacturing, a consistent pattern emerges. They are not, primarily, looking for tools that save their own teams time (though that is welcome). They are looking for tools that improve the experience of their workforce at scale — because they understand, correctly, that frontline worker experience is directly correlated with business performance.
This changes the design brief considerably.
An AI for HR that improves recruiter productivity is useful. An AI for HR that enables a night-shift warehouse worker to resolve a payslip query at 2 AM, in their own language, without calling anyone — that is genuinely transformative. The former optimises an existing process. The latter changes the nature of access.
There is also a deeper nuance around trust. Frontline workers are often, historically, the segment of the workforce least likely to believe that HR systems are designed with their interests in mind. The perception — sometimes accurate — is that enterprise HR technology exists to manage them, not to help them. An AI-first approach that is genuinely worker-facing, that answers their questions rather than routing them to a document, that is available at the moment of need rather than during office hours — this has the potential to fundamentally shift that dynamic.
Trust is not built through features. It is built through consistent, accurate, fast, accessible help. AI, when properly deployed, is the only mechanism that makes that available at scale.
Practical examples: what this looks like on the ground
The following scenarios are illustrative of what a genuinely frontline-focused AI HR platform enables.
Retail chain, 12,000 employees across 200 locations. Workers use a mobile app to check shift schedules and request time off, and raise absence notifications. When a new absence policy is introduced, workers can ask questions about it in natural language and receive accurate, contextualised responses — rather than being directed to a 14-page PDF that nobody reads. HR business partners see a 40% reduction in inbound queries on policy questions within the first month.
Logistics operator, 8,000 deskless workers, 18 nationalities. Workers communicate in their preferred language via an internal social network that keeps the entire organisation connected — without requiring a company email or a desk. The AI layer handles translation and response generation, drawing from a single policy source but surfacing answers in the worker’s language of choice. Compliance with mandatory training completion rises from 67% to 91% because reminders and nudges are conversational, personalised, and in the right language.
Healthcare provider, high staff turnover environment. New starters receive onboarding support through an AI assistant that answers their questions in the first 90 days — about pay, time tracking, rota systems, uniform policy, and escalation procedures. Managers run performance reviews from day one, and recognition flows naturally through Kudos — small signals that tell a new worker they belong. Time-to-productivity improves. Early attrition falls.
These are not hypothetical outcomes. They reflect the pattern we see when AI is deployed at the interaction layer, with frontline workers as the primary beneficiary.
Humand’s approach: intelligence where the work actually happens
At Humand, we have built from a specific conviction: the most important place to put intelligence in HR is not the analytics dashboard or the ATS — it is the moment of need for the worker. The measure of success is not whether the platform has AI. It is whether every frontline worker’s day is meaningfully smarter because of it.
We call it the “3 AM test”: if a worker has a question at 3 AM on a Sunday, can they get an accurate, helpful response without waking anyone up or waiting until Monday? For shift workers, healthcare staff, and logistics operatives, that test is not hypothetical. The 9-to-5 availability model is a structural mismatch with how these workforces actually operate — whether the question is about a shift or absence request, a clocked hour that doesn’t look right, or simply needing to know where to turn.
Our AI-first approach means intelligence is in the conversation, not the configuration. We vouch for accuracy, tone, personalisation, and multilingual capability — because that is where workers experience the platform. We are also deliberate about what AI should not do: it should not replace the human relationships that matter. The line manager who recognises a job well done through a Kudos, the HR BP who runs a performance review with genuine care, the team that stays connected through an internal social network even across dozens of sites — these require human judgment, empathy, and presence. AI takes the transactional load so those people can focus on what only humans do well.
Why This Matters – and What We Are Building Towards
Frontline worker experience is a leading indicator of business performance. Turnover, absenteeism, safety incidents, and customer satisfaction all correlate with how supported and informed workers feel. HR technology that genuinely serves the frontline is not a people team investment — it is an operational one.
“AI-enabled” implies a product that has had AI added to it. “AI-first” means the architecture, interaction model, and quality investment were all designed around AI from the start. The difference is concrete: an AI-enabled platform surfaces an AI feature when you look for it. An AI-first platform means you never had to look.
The goal is not more technology. It is smarter work — for every team, in every shift, in every language. The gap between what knowledge workers experience in their HR platforms and what frontline workers experience is real, documented, and consequential. Closing it is not a nice-to-have. It is the operational imperative of the next decade.
That is what Humand is building towards. Helping companies work smarter — not just at the top of the org chart, but from the palm of every hand.
2.7 billion people go to work without a laptop. Most HR platforms weren’t built for them. Humand was. See it in action.
