Upskilling for AI: Paylocity's Workforce Training Strategy

Upskilling for AI works best when technical skills meet soft skills. Get Paylocity's three-step advice with plans for managers and workers.

By Cosby Khosa · Published 21 October 2026

Key takeaways

Pair technical skills, such as clear instructions for AI tools and AI-assisted analytics, with critical thinking, creativity, adaptability and ethical reasoning.
Start AI adoption with simple, repetitive tasks, and survey employees about their concerns.
Refresh training as tools change, since AI evolves fast.
Paylocity sells HR software, so compare its advice with other sources.

Upskilling for AI works best when technical skills are paired with human ones. Paylocity’s how-to guide, Upskilling for AI: Workforce Training Strategies, published around 15 September 2026, makes that case. It pairs technical skills, such as writing clear instructions for AI tools and AI-assisted analytics, with soft skills such as critical thinking, creativity, adaptability and ethical reasoning.

It also recommends three practical moves: start AI adoption with simple, repetitive tasks, survey employees about their concerns, and refresh training as tools change.

This article expands on those three pieces of advice with practical steps for employers and workers. Any examples are our own illustrations, and the advice is general, so every organisation will need to adapt it. Paylocity sells payroll and HR software, so its guidance also serves its business. We return to that point near the end.

What upskilling for AI means

Upskilling for AI means helping people build the skills they need to work well with AI tools. It does not mean turning everyone into an engineer. For most workers, it means learning to use AI in their own jobs, to judge its results and to work alongside it.

It is different from reskilling, which means preparing for a different role. Upskilling deepens the skills you already use in your current one. Our companion guide on AI reskilling and how to prepare your workforce covers the other side of the coin.

Why does it matter? AI tools are spreading across workplaces. Gallup finds that employees whose manager supports AI use are 7.4 times as likely to say AI helps them do what they do best, yet only 28% report that support. That gap suggests many organisations have tools but little training.

LinkedIn’s survey found that about 75% of workers felt unprepared for a job search in 2026, which hints at a wider confidence gap about skills.

Upskilling for AI can close the gap. It helps people feel capable, improves productivity and reduces anxiety. It can also help retain staff, who value employers that invest in them.

The rest of this article covers the three main ideas in Paylocity’s guide: pair technical and soft skills, start simple and listen to employees, and keep training current.

The technical skills: instructions for AI tools and AI-assisted analytics

Paylocity names two technical skills. The first is writing clear instructions for AI tools, which we call instruction writing here. The second is AI-assisted analytics, which means using AI to explore and interpret data.

Instruction writing matters because the quality of AI results depends heavily on the quality of the request. A vague request gives a vague answer. A clear request, with context, a goal, constraints and an example of the desired format, tends to give a better one.

For example, instead of asking an AI tool to “write a report”, a manager might say: Summarise these sales figures for a non-technical audience in 150 words, highlighting the two biggest changes since last quarter and one risk. This tells the tool the audience, length, focus and purpose.

AI-assisted analytics involves using AI to clean data, spot patterns, create charts and draft explanations. The user must still check the results. AI can make errors, including inventing facts or misreading data. Good analysts verify numbers and test conclusions.

Upskilling for AI in these skills should be hands-on. Provide real examples, let people practise, and discuss what worked. Share good instructions in a library for others to reuse.

Technical skills also change fast. What works today may be obsolete next year, which is why refreshing training is vital.

The soft skills: critical thinking, creativity, adaptability and ethical reasoning

Paylocity pairs the technical skills with four soft skills. This pairing is the heart of the guide.

Critical thinking is the ability to question, check and judge. With AI, it means not accepting results at face value. Is this accurate? What might be missing? Does it make sense?
Creativity means generating ideas and seeing new possibilities. AI can produce options, but humans decide what is worth pursuing and how to make it fit the situation.
Adaptability means adjusting to change. AI tools and tasks evolve, so people must keep learning and be comfortable with uncertainty.
Ethical reasoning means considering what is right and fair. It covers privacy, bias, honesty and effects on people. Laws on AI in hiring in several US states show how important this skill is.

These skills are hard to automate and they make people better users of AI. Someone with weak judgement may simply use AI to produce poor results faster.

How do you build them? Use case discussions, scenario exercises, reflection and feedback. For example, present a case in which an AI tool produced a biased result, and ask the team to identify the problem and propose a fix.

Communication, collaboration and ownership matter just as much when you are job hunting, as our guide on government job interview questions and answers shows.

💡 Did You Know? Manager Support Makes AI Training Work 7.4× Better

Gallup finds that employees whose manager supports AI use are 7.4 times as likely to say AI helps them do what they do best, yet only 28% report that support. The biggest upskilling lever may not be a new tool at all. It may be the person who manages your team.

Start AI adoption with simple, repetitive tasks

Paylocity recommends starting AI adoption with simple, repetitive tasks. This is sensible for several reasons.

They are low risk. If the tool makes a mistake, the damage is small and easy to fix.
They are easy to measure. You can compare time and quality before and after.
They are tedious for people. Automating them is welcome, which builds goodwill.
They build skills and confidence. Employees learn how the tool works on familiar ground, and early wins create momentum.

Good starting examples include summarising meeting notes, drafting routine emails, formatting documents, sorting data, answering standard queries and preparing first drafts of reports.

A simple way to begin:

Run a pilot. Choose one or two tasks and a small group. Set goals, track results such as time saved and error rates, and gather feedback.
Check quality. Keep a person in the loop to review outputs, especially at first.
Be careful with data. Do not put confidential or personal data into tools that are not approved. Set clear rules.
Expand gradually. As people gain skills and trust, move to more complex tasks, always keeping risks and the need for human judgement in mind.

This stepwise approach to upskilling for AI is safer and more effective than a sudden, large rollout.

Survey employees about their concerns

The guide recommends surveying employees about their concerns. This is a vital step that many organisations skip.

AI makes people anxious. Gallup finds that 18% of US employees think their job is likely to be eliminated within five years, rising to 23% where AI is in use. Even those who are excited may worry about privacy, fairness, workload or being monitored. Some do not understand the tools and fear looking foolish.

If these concerns are ignored, they do not go away. They can lead to resistance, rumours and poor adoption. A survey gives people a voice. It tells you what worries them, what they want to learn and what barriers they face, and it shows respect.

Design it well. Keep it short and anonymous if possible. Ask about confidence with AI tools, training needs, concerns about jobs and data, and ideas for use.
Act on the results. Share the findings and explain what you will do. If you cannot address a concern, say why.
Follow up. Repeat the survey after a few months to track change.
Add conversations. Managers can ask their teams in meetings. Small group discussions can reveal issues that numbers miss.

Being open about upskilling for AI plans builds trust, which is essential for successful change.

Refresh upskilling for AI as tools change

Paylocity advises refreshing training as tools change. AI is evolving fast. New tools and features appear constantly, and old methods may stop working. A one-off course will soon be out of date, so plan for ongoing learning instead.

Set a rhythm. Hold regular updates, such as quarterly sessions on new features and good practice.
Mix the formats. Short videos, written guides, live demos, peer sharing and hands-on workshops suit different needs. Keep materials short and practical.
Appoint champions. Enthusiastic early adopters can keep up with developments and share them. Give them time and recognition.
Gather feedback. Ask what is not working, what new questions arise and what people want to learn next.
Update policies. Rules on data, quality and use should be reviewed as tools and laws change. California, Colorado and Illinois, for example, have rules on AI in hiring that employers must follow.
Track tool use. See which tools are used and how. If a tool is not used, find out why.
Budget for it. Training is not a one-time cost. Plan for regular investment.

Remember that people learn at different speeds. Offer support for those who need more time, and stretch for those who want more.

A simple upskilling for AI plan for managers

Managers have a central role. Gallup’s research shows that manager support is strongly linked to people’s experience of AI. Here is a six-week plan, our own illustration, based on Paylocity’s ideas.

Week one: talk to your team. Ask about their experience with AI tools, their concerns and their goals.
Week two: choose one simple, repetitive task to try with AI. Set a goal, such as saving an hour a week.
Weeks three and four: run the trial. Let a few people use the tool, and have them record what happens.
Week five: review. What worked? What went wrong? What did people learn? Share the results.
Week six: add a soft-skills discussion on checking AI results, ethical issues and when to use human judgement.
Ongoing: hold a short monthly session to share tips and update practices.

Ask HR for support, such as training materials and policies, and if your organisation has no AI guidance, ask for it. Be a model too: use AI tools yourself, and share your own mistakes and lessons.

What workers can do to upskill for AI themselves

You do not have to wait for your employer. Here are steps for individuals.

Try tools safely. Use approved tools on non-sensitive tasks. Practise writing clear instructions, and compare results.
Learn the basics. Free courses and tutorials can explain how AI tools work and what their limits are.
Build judgement. Always check results. Look for errors, bias and missing context, and learn to ask follow-up questions.
Strengthen human skills. Practise critical thinking, creativity, communication and ethical reasoning. They matter as much as technical skills.
Share what you learn. Teach colleagues, and learn from them.
Keep records. Note the tasks you improved and the results you gained, such as time saved. Add them to your CV.
Follow developments. Read reliable sources and be cautious about hype.
Protect data. Know your employer’s rules, and do not share confidential information with unapproved tools.
Ask for support. Request training, time to learn and clear guidelines.

Upskilling for AI is an investment in your future. In a slow job market, being able to show AI skills can set you apart. If you are heading into a job search, our guide on how to apply for jobs and get your application noticed shows where to put those skills on your CV.

How to read Paylocity’s guide with care

Paylocity is a payroll and HR software company. Its guide is educational, and it also promotes the company as a trusted source. The advice is sensible and widely shared, but you should compare it with other sources.

It is a how-to piece, not research, so it does not provide evidence of results, such as how many organisations improved after following the advice.
The publication date, around 15 September 2026, is approximate.
The guide appears aimed at US employers. South African organisations should adapt it to local contexts, such as skills development rules, language diversity and varied access to technology.
AI systems can make errors and their behaviour may change, so training should include guidance on checking results and handling mistakes.

Finally, upskilling for AI is not just a technical project. It is also about culture, trust and respect for people. Organisations that listen to employees and invest in their growth are likely to do better.

Quick Revision: Upskilling for AI Flashcards

Which two kinds of skills does Paylocity pair for AI upskilling?

Technical skills and soft skills

Which soft skills does it name?

Critical thinking, creativity, adaptability and ethical reasoning

Where should AI adoption start?

With simple, repetitive tasks

What should you do as tools change?

Refresh training

Frequently Asked Questions About Upskilling for AI

What does upskilling for AI mean?

Upskilling for AI means helping people build the skills they need to work well with AI tools: using them in their own jobs, judging their results and working alongside them. It does not mean turning everyone into an engineer, and it differs from reskilling, which prepares someone for a different role.

Which skills should employees learn first?

Paylocity pairs two technical skills, writing clear instructions for AI tools and AI-assisted analytics, with four soft skills: critical thinking, creativity, adaptability and ethical reasoning. Technical skills help people use the tools, while soft skills help them judge the results.

Where should a company start with AI adoption?

Start with simple, repetitive tasks such as summarising meeting notes, drafting routine emails or formatting documents. These tasks are low risk, easy to measure and tedious for people, so early wins build confidence and goodwill.

How often should AI training be refreshed?

AI tools change quickly, so a one-off course soon goes out of date. A sensible rhythm is a short update session each quarter, supported by a monthly tip-sharing session and a review of data and quality policies as tools and laws change.

Why survey employees before rolling out AI?

Surveys surface worries about jobs, privacy, fairness and workload that would otherwise turn into resistance or rumours. They also show what people want to learn. Keep the survey short, share the results and act on them.

Can individual workers upskill for AI without employer support?

Yes. Practise with approved tools on non-sensitive tasks, learn the basics through free courses, always check AI results, and keep a record of the time you save. Never share confidential information with tools your employer has not approved.

Final thoughts

Upskilling for AI is most effective when technical skills, such as writing clear instructions for AI tools and AI-assisted analytics, are paired with soft skills such as critical thinking, creativity, adaptability and ethical reasoning. Paylocity’s guide advises starting adoption with simple, repetitive tasks, surveying employees about their concerns and refreshing training as tools change.

Managers can run small pilots and share lessons. Workers can practise safely, build judgement and document results. Read the full guide, adapt it to your setting and treat upskilling for AI as an ongoing habit.

Sources and further reading

Paylocity, Upskilling for AI: Workforce Training Strategies (published around 15 September 2026)

Summary

Upskilling for AI: Paylocity's Workforce Training Strategy