AI literacy for HR teams and managers

Build practical AI understanding through source checking, safe information handling and scenario-based learning. This practical guide to AI literacy brings together a step-by-step approach, illustrative examples, a reusable worksheet and answers to common questions. Start with the section closest to your current challenge, then use the working session to turn the guidance into a clear next action.

Teach the decisions people need to make

AI literacy is the ability to use and question AI appropriately in a specific working context. Employees do not need to become model researchers to recognise an invented policy statement or an unsafe data-sharing request. Start with the tasks they actually perform.

Teach four practical questions: Is this use approved? Is this information appropriate to enter? Can I verify the result? What should I do if something is wrong? Connect each answer to the organisation’s real tools and processes.

Explain limitations through examples

Generated text can be plausible and still contain errors. Fluency is not evidence of accuracy. An output may omit context, repeat a bias in supplied material or invent a source. Demonstrate these issues using fictional records that do not expose employee information.

Show the difference between summarising a supplied observation and making an unsupported inference. “The deadline changed twice” is a possible source fact. “The employee lacks commitment” is an interpretation that may not be supported at all.

Practise meaningful review

Give learners an approved source and a draft containing deliberate mistakes. Ask them to identify unsupported claims, missing qualifications and an incorrect date. Require them to point to the evidence for each correction.

Include a case where the correct action is not to use AI, such as an unapproved tool receiving sensitive information. Include a case where the source is insufficient and the user should ask a person or locate the authoritative policy. Do not reward confidence when the evidence is missing.

Worked example: a policy answer

Illustrative exercise: an assistant answers a leave question using a policy for the wrong location. The answer is clear and cites a real document, but that document does not apply to the employee.

The learner checks the source’s audience, effective date and scope, then routes the question to the appropriate policy owner. The lesson is that a citation is a starting point for verification, not a guarantee that the answer is relevant.

Learning-session worksheet

• Role and task: Which decision will the learner make at work?
• Approved route: Which tool and information rules apply?
• Failure example: What realistic mistake should they be able to recognise?
• Verification: Which source or calculation resolves the uncertainty?
• Escalation: Who can help when the answer cannot be verified?
• Practice evidence: Can the learner explain the correction in their own words?
• Follow-up: What reminder or job aid supports the next real task?

Use short sessions with hands-on practice. Attendance records alone do not show that someone can apply the rules.

Reinforce learning as tools change

Collect common questions and anonymised examples of errors. Update job aids when approved tools or workflows change. Give managers a way to ask for help without encouraging improvised exceptions.

Review whether people can detect mistakes under ordinary time pressure. If checking is too difficult, redesign the use case rather than telling people to be more careful. The NIST AI Risk Management Framework provides a broader reference for risk awareness; the acceptable-use framework connects learning to workplace rules.

Put the guide into practice

Set aside a working session with the people who own this process and one or two people who experience it. Use a fictional or appropriately authorised case, so the discussion can be specific without sharing unnecessary personal information. The purpose is to leave with a usable decision or document, not just agreement that the topic matters.

Prepare the case

Choose a role and one approved task rather than designing a generic technology lecture. Identify the mistakes the person must recognise and the decisions they must make. Gather fictional source material and a deliberately flawed output. Make sure the exercise can be completed without entering real employee information into a training tool.

Write the starting assumptions down before discussing solutions. If the group disagrees on what happened, identify the information needed to resolve that difference rather than building a plan on an untested story.

Work through the decision

Ask learners to compare the output with the source and explain each correction. Include a missing qualification, an invented detail and a statement that is true but irrelevant to the audience. Discuss why fluent wording can make these errors easy to miss when someone is working quickly.

Ask each participant to explain the proposed decision in their own words. Differences in interpretation often reveal an unclear criterion, a missing responsibility or an instruction that will be difficult to follow.

Test an exception

Introduce a task outside the approved scope or a question that the source cannot answer. The learner should identify the limitation and use the correct help route. Do not reward an answer simply because it is confident or polished. Knowing when to stop and ask is part of practical competence.

Record what changes in this situation and what remains the same. An exception should lead to a clear next step, with an owner, rather than an informal workaround that nobody can explain later.

Agree the handoff

Give learners a short job aid with the approved tool, information boundary, source checks and escalation contact. Ask them to explain how they will use it in their next task. Managers should provide time for review and reinforcement rather than treating training completion as the end of responsibility.

Finish by confirming the owner, the next action and the date when the result will be reviewed. Give the person receiving the work enough context to continue without repeating the whole discussion.

Frequently asked questions

Does everyone need the same AI training?

A shared foundation can explain common limitations and workplace rules, but practical training should reflect the role and task. A manager reviewing performance text needs different examples from an analyst checking a query. Train people on the decisions they actually make, and provide deeper support for those designing, approving or operating AI-enabled processes.

How can we assess AI literacy?

Use realistic exercises where learners must identify an error, verify a claim, handle information appropriately and choose an escalation route. Ask them to explain their reasoning in plain language. A quiz can check basic recall, but attendance or a pass mark alone does not demonstrate safe use under normal workload. Observe whether the learning transfers to the approved task.

How often should training be refreshed?

Review learning when tools, policies, tasks or important failure patterns change. A short targeted update can be more useful than repeating an entire course. Use questions from employees and findings from pilots to choose examples. Keep job aids current and make it easy for people to ask for help when an unfamiliar situation arises.

Review the first cycle

Sample how learners apply the checks after training and ask which steps are difficult in real work. If source verification is too slow or confusing, improve the workflow as well as the instruction. Track recurring misunderstandings and update examples without exposing the personal information involved in actual incidents.

Keep a brief record of what was tried, what participants found useful and what needs to change. Compare the result with the original problem rather than judging success only by completion. If the process created extra work without improving clarity, quality or support, simplify it and test again. Share the agreed change with the people who will use it, and name the person responsible for keeping the guidance current.

About New Dynamics

New Dynamics connects goals, feedback, recognition and reviews around the way organisations work. This guide is published by the New Dynamics Editorial Team as part of our practical library for HR leaders, managers and People teams.

Use the examples and worksheets to structure your own discussions and adapt them to your organisation. Illustrative scenarios are not customer case studies. Policy and employment guidance needs appropriate local review before adoption.

For questions about this guide, corrections or a conversation about your performance management process, email contact@new-dynamics.com. Explore the complete guide library for related resources.