As of 13 August 2026, AI can recommend training courses for employees.

This still needs a person who signs their name to it.

Can you do it?

5 minutesto a draft.

30 minutesto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ita colleague

What the alternative costsThe supplied tool data gives no comparable price for a human training adviser.

If this goes wrong: you spend time and training budget on a course that does not address the employee's actual need, while the development need remains unresolved.

What to actually do

Hand it to a person
The route this page recommends
A person who owns the outcome does this end to end, worth it when the failure is dear.
Use a tool built for this
Second choice

Do it yourself

The distant third

A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

How to actually do it

Open the organisation's approved course catalogue, provider pages and internal learning policy, then copy the relevant course details into one document.
Gather the employee's role, development need, current skills, completed training, available time, delivery preferences, location constraints, budget and accessibility requirements.
Paste the employee information and course catalogue into a chatbot using the supplied prompt, and ask it to rank the courses without filling gaps by guesswork.
Open the provider page for every shortlisted course and compare the name, content, format, duration, price, dates, eligibility and accessibility information with the draft.
Send the checked shortlist to the employee's manager or learning specialist, who should test the recommendation against the employee's actual work, equal-access requirements and the organisation's training policy before approving it.

Prompt

Recommend training courses for the employee described below. Use only the course catalogue and employee information I provide. Do not invent course details, accreditation, prices, availability, outcomes or eligibility requirements. If important information is missing, state what is missing instead of guessing.

Employee role: [ROLE]
Employee's current skills and experience: [SKILLS AND EXPERIENCE]
Development need or performance goal: [DEVELOPMENT NEED]
Required or useful completion timeframe: [TIMEFRAME]
Available working time for training: [TIME AVAILABLE]
Delivery preference: [ONLINE, IN PERSON OR EITHER]
Location constraints: [LOCATION]
Budget: [BUDGET]
Accessibility or support requirements: [REQUIREMENTS]
Existing training already completed: [EXISTING TRAINING]

Course catalogue:
[PASTE COURSE NAMES, PROVIDERS, DESCRIPTIONS, PRICES, DATES, DELIVERY FORMAT, LOCATION, DURATION, ELIGIBILITY AND LINKS]

Return:
1. A ranked shortlist of up to five suitable courses.
2. For each course, explain which stated need it addresses and list any known trade-offs.
3. Identify any course facts that must be checked with the provider.
4. Identify any assumptions you made.
5. Suggest the best option and one alternative, but state clearly that the manager must make the final decision.
6. Flag any recommendation that could create unfair differences in access between employees.

Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

AI cannot know the employee's unspoken motivation, confidence, workload or relationship with the manager unless someone supplies that context.
AI cannot reliably judge whether a course teaches well or whether its level matches the employee beyond the information in the catalogue.
AI cannot take responsibility for deciding who receives training or whether different recommendations are fair across employees.
AI cannot guarantee that provider details, dates, prices or course availability remain current after the information was supplied.

Even on a YES, the friction has a name: judgement under ambiguity, context depth and verification cost.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

Axis	Score (0–2)
Output	2
Inputs	2
Verification	1
Liability	1
Effort delta	2
Total	8 / 10

The methodology and its thresholds are published in full.

FAQ

Can ChatGPT recommend training courses for my employees?: Yes. Give it the employee's role, development need, constraints and a current course catalogue, then ask it to explain and rank the matches. A manager still needs to check the facts, fairness and practical fit before approving a course.
Is it safe to use AI to choose employee training?: It is suitable for producing a shortlist, not for handing over the decision. Check the provider's current information and make sure the recommendation does not restrict fair access or rely on sensitive assumptions about an employee.
What information does AI need to recommend a training course?: Provide the employee's role, current skills, development goal, previous training, available time, delivery and location preferences, budget and accessibility requirements. Include the course catalogue with provider details, format, duration, eligibility and current availability.
Can AI match an employee to the right training course?: It can match stated requirements to course information and show the reasoning behind a shortlist. It cannot independently judge teaching quality, hidden development needs or the employee's readiness, so the manager or learning specialist must make the final choice.

Nearby answers

All Performance & training rulings

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.