AI for L&D: A Guide for Learning Teams
AI for L&D means using AI across the whole learning cycle: designing content, personalising paths, mapping skills, answering questions at the moment of work and showing what changed. Amit Kumar Soni led learning at Salesforce, Expedia Group and PepsiCo before founding Synottic. This page is what he tells an L&D team that is starting out.
Where AI helps learning teams
Six places where the time saved is real, and where an L&D team can see the difference within a quarter.
- 01
Instructional design
Turn a subject-matter interview into a first draft of the course, the facilitator guide and the assessment. Your time goes to instructional quality instead of the blank page. - 02
Personalised learning
Adaptive paths and AI coaching tied to a person's role and skills, so one programme does not have to serve everyone the same way. - 03
Skills intelligence
Map competencies, spot gaps and see which skills the business will need next, so capability spend follows evidence. - 04
Support in the flow of work
Knowledge assistants that answer a question when someone needs it, instead of sending them back to a course they finished months ago. - 05
Learning analytics
Turn learner data into a view that links training to a business outcome. Most L&D reporting stops one step short of that. - 06
L&D operations
Learner communications, scheduling, documentation and reporting: the admin load that eats a small team's week.
What stays with people
AI drafts fast and sounds sure of itself, including when it is wrong. Someone with instructional judgment has to review accuracy, tone and fit before anything ships.
IP and copyright, bias in learning content and learner privacy each need a named owner, not an assumption. And the decision about what a person should learn, and why, stays human.
Why L&D is the hardest seat in AI adoption
L&D is usually asked to teach AI to the whole organisation while also using it for its own work. That is two jobs, and the second one often gets skipped.
Teams that teach well have used the tools on their own deliverables first. Teaching a tool you have not used shows. And a course on prompts is awareness, not capability: people can finish it and change nothing about how they work.
AI capability building, from awareness to skill Why AI adoption stalls and what fixes itBuilt on running learning at scale
Amit led learning teams at Salesforce, Expedia Group and PepsiCo before he founded Synottic in January 2026.
How Synottic teaches it
Synottic, the Human-Centred AI company Amit founded, runs an AI for Learning and Development programme for L&D professionals. It has ten modules and five practical labs, and it ends with a capstone: each participant designs an AI-enabled learning and development transformation strategy.
The programme covers instructional design, personalised learning, skills intelligence, knowledge management and learning analytics. It is built on AI literacy, prompt and context engineering and responsible AI.
AI for Learning and Development at SynotticCommon questions
- 01
What is AI for L&D?
AI for L&D is the use of AI across the learning cycle, from designing content and personalising paths to mapping skills, supporting people at work and measuring impact. It speeds up the work around learning. It does not decide what people need to learn. - 02
Where should an L&D team start with AI?
Start with one deliverable the team already produces every month, such as a facilitator guide or a quiz bank. Use AI for the first draft and keep a person reviewing the result. A narrow start builds the habits and the trust a wider rollout needs. - 03
Will AI replace instructional designers?
It changes the job more than it removes it. First drafts get cheap, so judgment about what to teach, how to sequence it and whether it worked becomes the valuable part. Teams that treat AI as a drafting partner ship more and review more. - 04
How do you show the ROI of AI in learning?
Tie the learning to a business measure before you build it, then track that measure afterwards. AI makes it easier to pull learner data into one view, but the link between training and a business outcome still has to be designed on purpose.

Written by
AI educator, author, researcher and speaker. Founder and CEO of Synottic, and author of the book The Intelligence Mirror.
- PhD in AI, in progress
- P.G.P. in AI & ML, UT Austin
- ICF Professional Certified Coach
- 60K+ professionals trained
Last updated
Where this work lives
- SynotticThe Human-Centred AI company Amit founded: training, advisory and AI readiness for organisations. synottic.com
- The Intelligence MirrorAmit's book on how to think with AI without losing what makes you human. Read Chapter 1 free at theintelligencemirror.com
- Human-Centred AI newsletterEssays on AI adoption and the human side of work, on Substack.
Bring AI capability to your learning team.
Workshops, advisory and keynotes for L&D leaders who are building AI skill across the organisation.