Full AI Toolkit for Work and Side Income

Instruct the model to interview you first, so it gathers the missing detail before committing to an answer.

Give the model a template with gaps to complete, which locks the structure while letting it supply the content.

Ask for the same question answered from several viewpoints to surface angles a single response would miss.

Turn ChatGPT into a critic of your own work, asking for specific weaknesses instead of encouragement.

Set two or more options side by side and have the model weigh them against criteria you define.

Hand the model an output and ask it to reconstruct the prompt behind it, a fast way to learn prompt patterns from work you admire.

The Role, Goal, Context pattern: a repeatable prompt skeleton that covers what most weak prompts leave out.

Assign the model a persona to shift its vocabulary, assumptions, and depth for specialist tasks.

Set custom instructions once so ChatGPT carries your context and preferred output style into every new chat.

Upload an image and have the model read it, describing, extracting, and reasoning about what it sees.

Work with images in both directions on 4o, feeding visuals in and generating them back inside a single conversation.

Hand 4o a chart or a dataset and get trends, outliers, and plain-language interpretation back.

What 4o's wider knowledge base and cross-modal handling change in practice for research and analysis work.

Use Canvas to edit a document or block of code side by side with the model, instead of regenerating the whole thing each time.

What the o1 reasoning models do differently from 4o, and the kinds of problems that justify the slower response.

o1 in practice: worked examples showing where the reasoning models earn their cost, and where 4o mini is already enough.

Publish a conversation as a shareable link to hand a colleague the full context rather than a pasted excerpt.

Use temporary chats for one-off or sensitive questions you do not want kept in history or reused as context.

Set up the desktop app and the shortcuts that make it quicker to reach than a browser tab.

Generate images inside ChatGPT with DALL-E 3, refining them conversationally instead of rewriting prompts from scratch.

Use generation IDs to hold a character or style steady across a series of images.

Select and regenerate part of an image, changing one element while leaving the rest of the composition intact.

Set up a shared team workspace with pooled billing, shared GPTs, and conversations kept out of model training.

Install the browser extension that shows ChatGPT's answer next to your normal search results.

Compress long or technical documents to the level of detail you actually need, from a single line to a full brief.

Take one piece of writing and retune it for a different reader, adjusting vocabulary, assumptions, and depth.

Use ChatGPT to organize health and fitness information into plans and comparisons, and see clearly where a qualified professional is still required.

Translate between languages while preserving tone and intent, and check the result rather than trusting it blindly.

Turn a video or its transcript into a summary, key points, and timestamps in a fraction of the watch time.

Catch grammar, consistency, and clarity problems, and get the reasoning behind each suggested change.

Reshape messy pasted data into clean, consistent structure you can actually work with.

Pull names, companies, dates, and places out of unstructured text into a labelled list.

Sort free-text records into categories you define, with rules the model applies consistently across the whole batch.

Run a draft through a structured review for argument, flow, and word choice, then apply the edits selectively.

Generate assessment items of several types from source material, complete with answer keys and controlled difficulty.

Improve content that already exists for clarity, structure, and search performance, rather than starting over from a blank page.

Feed the model your domain material so its output uses your terminology and reflects how your industry actually works.

A worked example turning existing video scripts into book chapters, including where the format has to change.

Capture your brand voice in a reusable prompt so generated copy sounds like you rather than like a default model.

Use your own transcripts as source material so the model writes from your ideas and phrasing instead of generic filler.

A practical editing pass for AI drafts: what to check first, and the mistakes these models make most often.

Consolidate the prompting techniques so far into a repeatable method you can apply to any new task.

A full worked example, from weak first prompt to strong final one, showing exactly what each revision fixed.

Improve content that already exists for clarity, structure, and search performance, rather than starting over from a blank page.

Feed the model your domain material so its output uses your terminology and reflects how your industry actually works.

A worked example turning existing video scripts into book chapters, including where the format has to change.

Capture your brand voice in a reusable prompt so generated copy sounds like you rather than like a default model.

Use your own transcripts as source material so the model writes from your ideas and phrasing instead of generic filler.

A practical editing pass for AI drafts: what to check first, and the mistakes these models make most often.

The shifts worth tracking in AI capability and policy, and how to keep your own skills current as the tools change.

OpenAI's own prompting guidance, worth reading alongside this section to reinforce the techniques with the vendor's recommendations.

In this lab, you build a small reusable prompt library for one role you have or want, such as marketing manager, freelancer, or student. You identify the role's recurring tasks, write a prompt for each one with input variables and an expected output structure, and note when each prompt applies.

How generative AI changes the copywriting workflow, and which parts of content creation it genuinely speeds up.

Take an article or blog post from topic to finished draft, with the model handling structure, transitions, and first-pass wording.

Draft new copy and rephrase existing lines for length, register, or emphasis without losing the original meaning.

Build a product ad from a short brief, generating several angles so you can test rather than guess.

Write short-video scripts built for the first three seconds, where attention is won or lost.

Generate copy that reflects a specific product's features and buyer, rather than interchangeable filler.

Produce the title, description, and tags for a video in one pass, aligned to what people actually search for.

Use ChatGPT for keyword work and weave the results into copy that still reads naturally.

Turn an episode concept into a podcast script with segments, talking points, and natural spoken phrasing.

Write direct-response copy that leads with a clear offer and a reason to act now.

Generate original lyrics with a chord progression to match, and see how far the model can take a musical brief.

Build a resume from your raw history, targeted at a specific role rather than sent out identical to everyone.

Write a personal bio in several lengths and registers, from a one-line byline to a full speaker profile.

Draft a speech or presentation from a blank page, with an argument that builds and lands.

Produce a complete e-book, chapter by chapter, holding voice and structure consistent across the whole thing.

Generate social posts that fit each platform's length, tone, and conventions instead of one message copied everywhere.

Adapt content for a specialist niche, giving the model the vocabulary and context it needs to sound informed.

Find keywords and content ideas in one workflow, so research and planning happen together rather than separately.

Outline articles, videos, and podcast episodes before writing, which is where most of the quality is decided.

Write a complete article end to end, then edit for the things the model reliably gets wrong.

Generate titles and headlines in volume, then judge them against a clear standard rather than picking the first one.

Generate social post ideas in bulk so a content calendar starts from a long list rather than a blank page.

Write platform-specific post copy for Instagram, Facebook, and the rest, matching each one's length and tone.

Create polls and quizzes designed to pull replies, not just impressions.

Surface YouTube video ideas with real search demand behind them instead of guessing at topics.

Turn a video idea into a full script with a hook, body, and call to action.

Generate relevant YouTube tags that describe the video honestly and help it surface in related results.

Write YouTube descriptions that serve both the viewer and the search algorithm, with the key information up top.

Clean up auto-generated captions for accuracy and readability, which improves both accessibility and search.

What the SEO track covers, and how ChatGPT fits into a search workflow without replacing the fundamentals.

Where a language model genuinely helps in SEO, and where relying on it will actively cost you rankings.

Generate and refine keyword sets with ChatGPT, then sanity-check them against real search data.

Build long-tail keyword lists that target specific intent, where competition is lower and conversion is higher.

Turn keyword research into content ideas people will come back for, rather than one-off traffic grabs.

Write content that reads well for humans and still covers what search engines look for.

Audit and rework existing pages that underperform, which is usually faster than publishing something new.

Rewrite headlines and subheadings so they carry the keyword and still earn the click.

Generate meta titles and descriptions within the character limits, written to improve click-through rather than just fill the tag.

Use ChatGPT to structure backlink research and identify realistic link targets.

Analyze a competitor's site and content to find the gaps you can credibly fill.

What the social media marketing track covers, and how AI changes the production side of a content calendar.

A starter set of prompts for social content, ready to adapt to your own brand and platforms.

What ChatGPT is genuinely good at in social media work, and which tasks still need a human at the wheel.

Why prompt quality sets the ceiling on everything downstream, shown through side-by-side results.

How Gemini, Claude, and other models compare for social work, and when to reach for something other than ChatGPT.

Generate posts and captions tuned to each platform's format, length, and audience expectations.

Write longer blog content designed from the start to be cut down into social promotion.

Adapt one core idea across platforms and formats, including video scripts, without rewriting it from scratch each time.

Produce a month or a year of post ideas in one session, then refine the calendar rather than inventing it daily.

Specific prompt modifiers that improve readability: clarity scoring, active voice, conversational register, and contractions.

Ask ChatGPT to critique your content against explicit criteria, and get usable feedback instead of praise.

Build content for different platforms, goals, and outcomes, so each post has a job rather than just filling a slot.

Break an effective prompt into its parts and see what each one contributes to the result.

Coach the model toward a specific tone through examples and correction, rather than one-word style instructions.

Generate ad and sales copy with a clear offer, audience, and angle defined before you prompt.

The scheduling and automation tools worth adding once AI is handling the writing.

Mine customer reviews for patterns, then act on what they say about your product and your copy.

Where copywriting demand actually sits on Fiverr, and how AI changes what you can deliver and charge.

A worked freelance example: writing a full sales page for a client with ChatGPT, start to finish.

The social content services freelancers can realistically sell, and how AI changes the economics of each.

A worked example building a client content calendar and the posts to fill it.

The video scripting work available to freelancers, and what clients expect from a finished script.

A complete worked example: one YouTube video script from concept through hook, body, and outro.

In this lab, you pick a keyword relevant to a niche you know something about, run ChatGPT keyword research, draft a 1,200 word blog post built from that research, optimize it for the keyword, and document every prompt you used along the way.

Google's own quality rater guidelines, the closest thing to an official statement of what the search engine rewards.

How to feed large volumes of source material to ChatGPT and get back a synthesis you can actually act on.

The questioning technique matters more than the tool: how to frame a question so the answer is worth having.

A set of diagnostic questions to point at your own business, using the model as a structured thinking partner.

Run a sharper job search: target roles, decode listings, and identify what a posting is really asking for.

Prepare for an interview with role-specific practice questions and rehearsed answers to the hard ones.

Write outreach to a potential employer that reads as genuine and specific rather than mass-sent.

Break down how other people achieved a result, and extract the parts that transfer to your own situation.

Use historical events as case material, asking the model to explain causes, consequences, and competing interpretations.

Think through social and workplace dynamics with the model, rehearsing difficult conversations before having them.

Practical personal uses: gift ideas, planning around other people's interests, and thinking through how to say something well.

Learn unfamiliar software quickly by having the model explain it at your level and answer follow-ups as you go.

Build a self-guided tutorial on almost any topic, paced and pitched to what you already know.

Analyze an article or document for argument, evidence, bias, and what it leaves out.

Work through multi-step math problems, and understand why you must check the working rather than trust the answer.

Compare options against criteria you set, and get a structured answer instead of a wall of prose.

Take a comparison and rewrite it into finished content, moving from analysis to something publishable.

Use ChatGPT as a language partner for vocabulary, grammar explanation, and practice conversation.

Plan a DIY or home improvement job step by step, including materials, sequence, and the points where a professional is the right call.

Build a travel itinerary shaped around your dates, budget, and interests rather than a generic top-ten list.

Generate fun facts, quotes, and jokes on any topic, useful filler for newsletters, posts, and presentations.

Turn a list of commitments into a realistic schedule, with the model handling sequencing and time blocking.

What a Custom GPT is, how it differs from a well-written prompt, and when building one is worth the effort.

Build your first Custom GPT end to end: instructions, knowledge files, and the settings that shape its behaviour.

Navigate the GPT Store to find what already exists before building your own, and judge which entries are worth using.

Build a business assistant GPT loaded with your own documents, so it answers from your material rather than the open web.

How retrieval augmented generation lets a chatbot answer from your documents, and why it is the backbone of most business AI assistants.

Call your Custom GPTs from inside a normal ChatGPT conversation and switch between them mid-task.

A Custom GPT that recommends the right AI tool for a described task, useful in a landscape that changes monthly.

A Custom GPT that writes prompts for you, turning a rough intention into a properly structured request.

Use VoxScript to search and transcribe YouTube videos, pulling their content into a conversation.

Order groceries through Instacart from inside ChatGPT, an example of an action-taking integration rather than a text one.

Use the Show Me plugin to generate diagrams and visual explanations from a written description.

Plan and price a trip with the Expedia integration without leaving the chat.

Pull structured courses and learning paths through the edX integration.

Generate a rough video from a text brief with Visla, straight out of a ChatGPT conversation.

Use Kraftful for product research and planning, aimed at teams making build decisions.

Find and book restaurant reservations through the OpenTable integration.

Apply Coupert to find discount codes automatically while shopping online.

Search flights and hotels through the Kayak integration inside ChatGPT.

Hand calculations, unit conversions, and hard maths to Wolfram, which computes rather than predicts the answer.

Use Scholar AI to find and summarize academic sources with real citations behind them.

Use WebPilot to read live web pages inside a chat, working around the training cutoff.

What the marketing track covers, and which parts of a marketing workflow AI meaningfully compresses.

Why a tightly defined audience sells more than a broad one, and what a usable audience definition contains.

Use ChatGPT for audience research, turning assumptions about your buyer into something structured and testable.

Build a full content calendar with themes, formats, and cadence, rather than deciding what to post each morning.

Generate keyword-led blog titles that have a reason to rank, not just a keyword dropped into a sentence.

Draft an SEO-optimized blog post quickly, then apply the editing pass that makes it publishable.

The mechanics of email marketing that AI does not change: list, offer, timing, and deliverability.

Build a full email sequence where each message has a job and the whole series builds toward one action.

Write Facebook ad copy with the platform's format, audience targeting, and creative conventions in mind.

Write Google Ads copy inside the character limits, matched to search intent rather than brand messaging.

Draft sales page copy with a structure that carries a reader from problem through proof to offer.

Map a complete marketing funnel stage by stage, then use ChatGPT to fill in each piece.

Build a buyer persona detailed enough to change how you write, not just a demographic sketch.

Create a lead magnet worth an email address, and the copy that gets people to claim it.

Write the landing page and social copy that feed the funnel, kept consistent in voice and offer.

Write the full email sequence behind the funnel, from first delivery through follow-up and close.

How AI fits into everyday office work, and which routine tasks it removes rather than merely speeds up.

Use ChatGPT to research and work through problems at work, with a method for checking what it tells you.

Turn raw meeting notes into structured minutes: key points, decisions, and owned action items.

Draft email replies that match the sender's register and actually answer what was asked.

Rework your resume against a specific job description, so the relevant experience is the visible experience.

Generate product and business name candidates, then filter them against domain, trademark, and pronunciation checks.

Write taglines and slogans in volume, then judge them on clarity and memorability rather than cleverness.

A workshop building a real slogan and tagline for a business, from brief through shortlist to final choice.

Write product descriptions for an online store that sell the benefit and still carry the search terms.

Generate an FAQ from real customer questions, which cuts support load and helps the page rank.

Analyze customer reviews to find the recurring complaints and improve the product, not just the copy.

Research podcast concepts and generate names that are memorable and still available.

Outline and script podcast episodes with a structure that holds attention across the runtime.

Brainstorm interview questions and build a realistic list of guests worth approaching.

Build a professional-looking website using AI site builders, from layout through to copy.

Customize an AI-generated site so it stops looking like a template and starts looking like a brand.

Where ChatGPT fits into a website project, from planning the page structure to writing every block of copy.

Write the core pages of a site: homepage, About, and Contact, each doing a distinct job.

Write landing page headlines and body copy focused on one action rather than several.

Generate meta descriptions that fit the character limit and earn the click from a results page.

Find a course idea with real demand behind it, and validate it before you build anything.

Build the course itself, using ChatGPT for structure, lesson content, and worked examples.

Write the course landing page and the promotional emails that drive enrolment.

Produce supporting educational content and the email sequence that keeps students moving through it.

Draft answers to student questions quickly while keeping the personal tone that makes them land.

Generate course quizzes with answer keys, distractors that are actually plausible, and controlled difficulty.

Use ChatGPT for research, structure, and feedback while keeping the work your own, and understand where academic integrity rules draw the line.

The responsibilities that come with AI writing tools: disclosure, accuracy, attribution, and knowing what you are accountable for.

What humans still do better, what the model does better, and how to divide a writing task between them.

Slot AI into an existing writing workflow at the points where it helps, instead of rebuilding how you work around it.

The developments worth watching in AI capability and regulation, and how to keep your skills from going stale.

Three prompts that pay for themselves immediately, plus what the rest of this section will cover.

What ChatGPT is under the hood, and a walkthrough of creating an account and getting to your first prompt.

The single habit that separates people who get value from ChatGPT from people who try it twice and give up.

Write work emails that state the ask clearly, match the relationship, and get a reply.

Build a work presentation with AI, from structure and narrative through to the words on each slide.

Proofread and edit with a checklist the model applies consistently, rather than a vague request to improve the text.

Turn a messy brain dump into a to-do list with clear, actionable items instead of vague intentions.

Rank tasks by urgency and importance, and get a defensible reason for the ordering rather than a gut call.

Ask the model to review how you work and suggest where the time is actually going.

Draft reports and long documents with a structure that holds up, then edit for accuracy and voice.

Hand a dataset to ChatGPT and get patterns, comparisons, and plain-language insight you can put in front of a stakeholder.

Summarize emails, articles, and documents to the length and depth you need, whether that is a line or a page.

Translate work documents and emails while keeping professional register, with a check step for anything consequential.

Take a blog post from topic to draft in minutes, then spend your time on the edit rather than the blank page.

Get a fast, structured briefing on an unfamiliar topic mid-task, with sources to verify before you rely on it.

Build a personal learning plan with the model, pitched at your current level and the skill you actually need next.

Get a difficult concept explained at several levels of depth until one of them clicks.

Structure market research, from sizing a market to mapping competitors, with the model organizing what you find.

Prepare properly for a meeting and turn what happened into minutes people will actually read.

Convert raw notes into structured minutes: decisions made, points raised, and actions with owners attached.

Work through risks and forecasts systematically, using the model to surface scenarios you have not considered.

Troubleshoot technical problems step by step, with the model narrowing causes instead of guessing at fixes.

Tune your resume to a specific role so the relevant experience is the first thing a reader sees.

Write cover letters that say something specific about the company and the role rather than restating the resume.

Search harder: decode job descriptions, find the roles that fit, and target applications rather than mass-sending them.

Rehearse interview answers and salary negotiation with the model playing the other side of the table.

Produce user manuals, software documentation, and FAQs from source material, in a consistent structure.

Get the Excel formula you need by describing the problem in plain English, including nested and lookup formulas.

Write and understand basic code with ChatGPT's help, even without a programming background.

The responsibilities that come with AI writing tools: disclosure, accuracy, attribution, and knowing what you are accountable for.

What humans still do better, what the model does better, and how to divide a writing task between them.

Slot AI into an existing writing workflow at the points where it helps, instead of rebuilding how you work around it.

A worked case study on the front half of writing: brainstorming angles and building an outline worth drafting from.

A full worked example writing one article with ChatGPT, from outline through draft to final edit.

A case study on the highest-leverage lines you write: subject lines, titles, and headlines.

Write a video script two ways, from a blank page and from a reusable template, and compare the results.

A worked example producing social posts and promotional emails for one campaign, kept consistent across both.

In this lab, you pick one task you handle repeatedly at work or in your business, then build a Custom GPT for it with a system prompt, clear instructions, and a set of starter prompts, finishing with a short plan for refining it after real use.

OpenAI's own guide to building GPTs, useful reference material alongside the Custom GPT lessons in this section.

What the Gemini track covers and why a second frontier model is worth learning alongside ChatGPT.

The real differences between Gemini and ChatGPT: live search access, Google Workspace integration, and how each handles long context.

Recent Gemini additions worth knowing, including one-click export into Google Docs and interface options.

Research with Gemini's live access to Google search, which sidesteps the training-cutoff problem entirely.

Run digital marketing research in Gemini, pulling current market and competitor information rather than cached knowledge.

Do keyword research inside Gemini, with the advantage of Google's own search data behind the model.

Outline articles, videos, and podcasts in Gemini before committing to a draft.

Write a full article in Gemini, then edit for the patterns this model in particular tends to fall into.

Generate article titles that carry the keyword and still read like something a person would click.

Rework existing titles for search without gutting what made them readable.

Generate social post ideas in Gemini, informed by what is currently being searched for.

Write posts for Instagram, Facebook, and X in Gemini, matched to each platform's tone and length.

Create social polls in Gemini designed to prompt a response rather than a scroll.

Generate fun facts, quotes, and jokes in Gemini for newsletters, posts, and presentation openers.

Find video ideas in Gemini using live search interest rather than guesswork.

Write a full video script in Gemini, from hook through body to call to action.