How to Use AI to Generate Public Speaking Ideas
Liam Sandford
Liam Sandford is a Head of Marketing, public speaking expert, and 2x Best Selling Author including the book Effortless Public Speaking. He helps ambitious professionals and entrepreneurs communicate with impact to get noticed, grow their career, and build their business.
Coming up with something worth speaking about is often harder than delivering it. Most people are not short of knowledge, experience, or opinions. They get stuck deciding what to speak about, how to frame it, and which angle will land. That uncertainty is what produces forgettable, overstuffed presentations and the last minute rewrites that quietly wreck your confidence.
This is where AI earns its place, not as a replacement for your thinking but as a partner that helps you surface, organise, and challenge ideas when you are stuck, overwhelmed, or too close to a topic to see it clearly. Used with intention, it gets you off the blank page and into structured exploration fast, so you can compare directions before committing to one. The aim is not to outsource creativity, it is to widen it.
One principle sits underneath all of it: it is not about you, it is about your audience. Every idea is only as good as its usefulness to the people in the room. Before AI ever existed, the way I found ideas worth speaking about was by paying attention to that audience, and the method I describe below still does most of the heavy lifting. AI just makes it faster. Finding ideas is one stage of the wider AI workflow for public speaking.
How to Use AI to Generate Public Speaking Ideas
AI produces better ideas when you give it direction. With no boundaries it returns broad, generic content disconnected from your context, which is why people conclude it "is not very good" when really it was never told enough to be useful. A small amount of thinking up front sets the constraints AI then works within. Skip it and AI gets noisy. Do it well and AI gets focused.
Clarify the Audience and Context First
An idea that kills at a conference can die in a team meeting, and one that lands with beginners can insult experts. Before you prompt, be clear on who the audience is, why you are speaking, and the situation you are walking into. Context is more than demographics: it is their emotional state, their expectations, and their constraints. Are they here by choice or by obligation? Is this a high stakes decision or a place to learn?
There is a shortcut I lean on for this: the 2 Year Test. Think back to where you were two years ago and the problem you were wrestling with. More often than not, your past self is sitting in your audience, facing the thing you have since worked out. Ask AI to help you reconstruct what you struggled with back then, and you often land on the most useful angle in minutes. You do not need perfect answers, you need constraints, and even rough ones, the experience level, the industry, the time limit, the goal, turn abstract suggestions into usable ones.
Define the Outcome You Want From the Speech
An idea is only useful when it serves an outcome. Before you ask AI for anything, decide what you want the audience to think, feel, or do differently by the end. That single outcome becomes the filter every idea has to pass through. Without it, AI hands you a pile of interesting concepts that pull the speech in five directions at once. With it, you can see at a glance which ideas support the outcome and which are just distractions wearing a clever hat. A quick example. Say you are speaking to new managers. "Help them feel more confident" is too vague to filter anything. "Get every person to run their next one to one using the three questions I give them" is specific enough that you can hold each idea against it and see at once whether it earns a place. The sharper the outcome, the faster the editing, because most ideas disqualify themselves the moment you ask whether they move the audience towards it.
Using AI to Generate Core Speech Ideas
With the foundations set, this stage is about breadth before depth. You are not hunting for the one idea yet, you are mapping what is possible so you can step outside your usual thinking. AI is a strong accelerator here, as long as you keep feeding it the right raw material.
Mine Your Audience's Real Problems
The best ideas are not invented, they are noticed. Here is the method I used long before AI, and still do: the rule of three. When a question or frustration comes up in your audience's world once, note it. When it comes up a second time, it is worth a short social post. When it comes up a third time, it has earned a place in a presentation or a proper resource, because three sightings means it is a real, recurring problem rather than a one off.
AI turns this from a slow habit into a fast one. Collect the questions, comments, and objections you keep hearing, the replies under your posts, the things people ask after a meeting, the threads in your industry, and paste the batch into AI. Ask it to group them into themes and show you which themes recur most. The recurring ones are your strongest topics, because they are the problems your audience keeps running into. You are not asking AI to invent ideas, you are asking it to organise the evidence you already gathered. A prompt that works well here: "Here are 40 questions and comments my audience has raised over the past month. Group them into themes, rank the themes by how often they appear, and for the top three suggest a presentation angle that would genuinely help." What comes back is not a list of clever topics, it is a map of what your specific audience needs, ranked by demand, which is a stronger starting point than anything invented from a blank page.
Exploring Fresh Angles on Familiar Topics
Plenty of speakers worry their topic has been done to death. The fix is rarely a brand new subject, it is a sharper angle on a familiar one. Ask AI how the topic plays out for different roles or experience levels, or where the common advice quietly oversimplifies reality. Ask it to surface the tensions, trade offs, and contradictions inside a topic, because those reflect how people really experience the problem rather than the polished theory. Fresh angles come from relevance and honesty, not novelty for its own sake.
Developing Stronger Ideas From AI Outputs
Raw AI ideas are rarely strong enough to stand on without your experience added. They tend to be vague, overgeneralised, and detached from a real audience. They need shaping, testing, and narrowing before they are ready to be spoken aloud. This is where speakers either rush forward with something half formed or polish an idea until it loses all clarity. Both wreck confidence and delivery.
Stress Testing Ideas for Relevance and Impact
Once you have a shortlist, pressure test it before you ever stand up. This is the prompt most people never use and the one I use most: paste your idea in and say, "act as a sceptical member of my audience, tell me where this loses you, where it sounds generic, and which claim you do not believe." It stings the first time, then it becomes the most useful pass you do. You can also ask it for the objections a specific audience would raise and the weak points in your reasoning.
Stress testing shows whether an idea has depth or collapses under scrutiny. Weak ones sound compelling at a glance and fall apart when examined, because they lean on assumptions or ignore the audience's context. Strong ones sharpen under pressure. The bonus is confidence: when you know where an idea might be questioned, you can address it head on rather than being caught out during the Q&A.
Narrowing Ideas Into a Clear Core Message
Most speeches fail not because the ideas are weak but because there are too many of them. Speakers try to say everything they know instead of deciding what matters most. Too much context is the killer of attention, and each time you switch between ideas you make the audience work harder to follow you. The discipline is to narrow to a single core message.
The test is simple and unforgiving: if you cannot say your main point in one sentence, you are not clear enough yet. Ask AI to express your idea in one line, to separate the essential supporting points from the optional ones, and to name the single thing the audience should still remember a week later. This forces prioritisation, and it exposes the points you are clinging to out of attachment rather than relevance. Once that core message is clear, everything else gets easier, and building the speech around it becomes simpler.
Using AI to Generate Supporting Material for Ideas
With a clear core idea, AI helps you build around it so the message is easier to understand, remember, and act on. Supporting material has one job: to serve the idea, never to compete with it. AI is useful for spotting which examples strengthen understanding and which just add noise.
Generating Examples and Analogies
A good example is what turns an abstract point into something the audience can picture, and the right one depends entirely on who is in front of you. Ask AI to brainstorm examples that fit your audience's world, and to offer two or three analogies for the same idea so you can feel which one clarifies and which one strains. This matters most when you are explaining a framework or a process that would otherwise stay abstract. These are the "eyes light up" moments, where the room suddenly gets it.
AI gives you options, not answers. Always rewrite an example in your own language and stand behind it, because authenticity comes from ownership, not originality. An example only becomes powerful when you can deliver it naturally and it is genuinely yours. Shaping an example into a story that lands is its own craft.
Identifying Common Questions and Misunderstandings
Strong ideas anticipate confusion before it happens. Ask AI where an audience might misread your message, which questions they are likely to ask, and which assumptions need clearing up. Address those points inside the speech and you guide the audience rather than lecture them, which lowers the pressure on you and makes the whole thing feel conversational.
One thing no prompt prepares you for is the question you genuinely cannot answer. Here, honesty wins. I used to try to bluff and feel terrible afterwards. Now the line is simply "I will take a look and get back to you", and then I follow through. Audiences respect that more than a waffled non answer, and it hands you a reason to reconnect later.
What to Avoid When Using AI for Public Speaking Ideas
AI is a powerful tool for generating ideas, but left unchallenged it tells you what you want to hear. It is very good at producing language that sounds confident even when the idea underneath is weak. Knowing what not to do matters as much as knowing the prompts.
Avoid Treating AI Ideas as Finished Content
AI generates starting points, not conclusions. Lift an idea straight from the screen without reshaping it and it arrives on stage with no depth, no conviction, and no personal relevance. Ideas become powerful through understanding, not generation: you need to know why one matters, how it applies to your audience, and what makes it worth sharing.
Here is a rule of thumb that keeps me honest: if you would struggle to explain the idea clearly without notes, it is not ready to be spoken about. A finished idea feels owned and flexible. You can adapt it in real time without losing your footing, and that only comes from your own thinking, never from copying output. Treating its output as finished is one of the common mistakes that flatten a presentation.
Avoid Chasing Novelty Over Relevance
AI is excellent at producing unusual ideas, which is tempting and occasionally useful, but novelty alone does not make an idea work. Chasing originality for its own sake produces content that feels disconnected from the audience's real challenges. Audiences care about relevance, not cleverness. They want something that helps them understand, decide, or see a problem clearly.
An idea that feels new but irrelevant loses trust faster than one that feels familiar but deeply useful. So put clarity and usefulness ahead of originality, and ask the only question that counts: does this genuinely help this audience, in this context? The strongest presentations are rarely built on shocking insights. They are simple, grounded ideas delivered with intention, because in the end it is not about you, it is about them.
Using AI to Support Creative Thinking, Not Replace It
AI works best as a creative support system, not the source of your ideas. Used well, it helps you explore faster, spot patterns you might miss, and move through the early, doubt heavy stage with less friction. The danger is letting it become the place your ideas come from rather than a tool for developing the ones you have.
It can assist with exploration and refinement, but the insight has to come from you. The outcome of using AI well is not an AI generated idea. It is a speaker who has thought hard about their message, knows exactly why it matters, and can feel how it will land. Stay curious, intentional, and in control, and AI becomes a genuine ally for finding ideas that are clear, relevant, and worth an audience's time.
FAQs on How to Use AI to Generate Public Speaking Ideas
How can AI help me come up with speech ideas?
It is best as a thinking partner, not an idea machine. Give it your audience, context, and outcome, then use it to widen options, group the real problems you have noticed into themes, and pressure test a shortlist. The strongest ideas still come from your own experience.
What should I speak about if my topic feels overdone?
Change the angle, not the subject. Ask AI how the topic plays out for a specific role or experience level, or where the usual advice oversimplifies things. The tensions and contradictions inside a familiar topic are usually where the fresh, useful ideas live.
How do I know if an idea is strong enough to build a speech on?
Test two things. Can you say its core in one sentence, and could you explain it clearly without notes? If yes, it is owned and ready. If not, it needs more of your thinking before AI helps you build around it.
Does using AI make my ideas less original or authentic?
Only if you stop at copying. Authenticity comes from ownership, not novelty, so reshape everything in your own language and stand behind it. Used to develop your ideas rather than supply them, AI leaves the speech unmistakably yours.
TL;DR: How to Use AI to Generate Public Speaking Ideas
AI helps you find stronger public speaking ideas by widening your thinking, testing relevance, and sharpening direction, without replacing your judgement or your creativity.
Set the constraints first: audience, context, and the one outcome you want, so AI's ideas are usable rather than abstract.
Mine real problems with the rule of three, then have AI group what you have collected into recurring themes.
Pressure test a shortlist with the sceptic's prompt, then narrow to a single core message you can say in one sentence.
Build support that serves the idea: audience-fit examples, sharp analogies, and answers to the questions people will ask.
Keep ownership. If you cannot explain it without notes, it is not ready, and relevance always beats novelty.
More From Liam Sandford
Read my book: Effortless Public Speaking. Learn how to speak confidently, reduce stress, and turn public speaking into your competitive advantage. These actionable public speaking tips will help you improve your presentation skills for any audience.
Join the free 5-day email course: Get daily lessons packed with practical strategies to deliver effective presentations and speak confidently. This course is designed to build your public speaking skills step by step. Sign up below: