The best tutor in the world is free — and you’ve probably already got it open in a tab

Right now, you have access to the best tutor or mentor in the entire world, and it’s free. If you want a slightly better version, it’s $20 a month. Push hard and you might spend $100–200 a month. That’s the whole price range.

Of course, we’re talking about AI. And the claim isn’t hype — it’s provable. AI has the entire web at its fingertips, does remarkable pattern recognition, summarizes beautifully, and explains things at whatever depth you want. It meets you exactly where you are: total beginner, advanced, expert, anywhere in between. You control how it responds and how long its answers are. You can feed it context — documents, YouTube transcripts, your own notes — and it will teach you from them. It can build quizzes, turn a topic into a full study curriculum, make flashcards, and go back and forth with you in voice mode.

And it never gets tired. It’s perpetually patient.

If you have high agency and curiosity, and you’re willing to spend real time in conversation with it, AI is the most powerful learning tool you can touch. Here’s how to set it up properly — as a dedicated AI mentor project — plus the automation tips, agents, tools, and resources that turn it from a chatbot into a genuine learning partner.

Watch the full walkthrough:

Start with deep research: your on-demand expert briefing

Before building anything, learn how to learn. In ChatGPT, open a new chat, hit the + button, and select Deep Research. Now prompt it like you mean it — long, detailed, thorough. Ask for best practices, recent studies, what the experts are actually saying (podcasts, write-ups, GitHub discussions, posts on X — everywhere).

If you spend the time to actually read the resulting report like an article, you’ll come out more knowledgeable about the topic than most people — grounded in recent data, with sources you can click through to fact-check against hallucination.

Tip: speak your prompts. Tools like Whisper Flow (or the built-in dictate button) let you ramble out a rich, detailed prompt in seconds. You’ll write fatter, better prompts than you’d ever type.

Build your AI mentor project (step by step)

This works in ChatGPT (Projects), Claude, Gemini (Gems), or Grok — the concept is the same everywhere. Here’s the setup:

  • Create a new project and name it something like “AI Mentor.”
  • Decide on memory scope — project-only keeps things tidy; default lets it draw on your wider history.
  • Run a deep research pass asking for best practices on tailoring an AI project to be the ultimate tutor and mentor about AI itself: how to learn it, use it better, understand large language models, troubleshoot problems, and become more valuable by leveraging AI.
  • Download the research as Markdown (AI models tend to work best with Markdown) and add it as a source in your project.
  • Have the AI write its own custom instructions. In the project chat, attach the research file and ask it to write custom instructions that make the project function as your AI tutor and mentor about AI. Copy the result into your project settings. (Watch the character limit — ChatGPT caps project instructions at 8,000 characters; if it errors, ask the AI to shorten it.)
  • Save, and start asking. “I don’t understand what a large language model is — explain it so I can grasp it and explain it to friends.” Because the project has your research as context plus tailored instructions, the answers teach better than a generic chat ever would.

You can do this for any topic — and even tailor the voice. Want it to respond like a favorite business thinker or podcaster? Do a voice-analysis deep research pass and plug that in too. The whole game is: curate the sources, let the AI write the instructions, paste, save, learn.

Two power features most people skip

Voice mode as a conversation partner. Go back and forth by voice — have it search things up mid-conversation. One practical setup: map your phone’s action button to your AI app’s voice mode (Perplexity’s is excellent for research and fact-checking) so your research assistant is one press away.

Temporary chat for sensitive material. If you’re feeding the AI private information, turn on temporary chat first. The conversation won’t stay in history, won’t train the models, and fades after 30 days.

What is an LLM, anyway?

Here’s the clean explanation worth stealing for your next dinner conversation:

A large language model is an AI system trained to predict the next piece of text. It has read a staggering amount of text — books, websites, articles, code, conversations — and learned the patterns of language: what words come next, how sentences are structured, how ideas relate. When you type a prompt, it predicts the most useful response one piece at a time.

Large means it learned from massive data with billions or trillions of internal settings. Language model means it models patterns in language. Think of it as extremely advanced autocomplete — but because language contains knowledge, reasoning patterns, code, and human behavior, predicting text well can look like thinking.

It doesn’t know things the way a person does. It generates answers from learned patterns — which is why it’s incredibly useful and sometimes confidently wrong. LLMs are powerful thinking aids, not truth machines.

AI automation tips: where the real time savings live

Learning with AI is step one. Automating with AI is where the hours come back. A few high-leverage moves:

  • Automate the boring middle. Summarizing meetings, drafting follow-ups, triaging email, turning rough notes into polished docs — these are the highest-ROI automations for most people.
  • Use the automation features already in your AI tools. ChatGPT, Claude, and Gemini all now ship built-in automation and scheduled-task features — set up recurring research briefings, weekly summaries, or reminder workflows once and let them run.
  • Chain tools, don’t just chat. Connect your AI to your calendar, inbox, and docs. An automation that watches for a trigger and acts (drafts, files, summarizes, reminds) beats a brilliant one-off answer.
  • Start with one workflow. Pick the task you repeat weekly that you hate most. Automate that one thing end to end before touching anything else.
  • Keep a human in the loop for anything public-facing or irreversible — review before it sends, posts, or spends.

AI agents: your mentor’s bigger sibling

An AI agent is AI that doesn’t just answer — it acts: planning multi-step work, using tools, browsing, and completing tasks with minimal supervision. Practical ways to use them:

  • Research agents (deep research modes in ChatGPT, Gemini, Grok) that spend minutes doing hundreds of searches and return sourced reports.
  • Coding and build agents that write, test, and iterate on software from a plain-English description.
  • Workflow agents that handle multi-step jobs — prospect research, lead enrichment, content drafting pipelines — while you do something else.
  • Voice agents you can talk to hands-free while walking, driving, or cooking.

The pattern from the mentor setup applies here too: give the agent great context, clear instructions, and good sources — then let it work. Garbage in, garbage out has never been more literal.

Tools and resources worth your time

  • ChatGPT (Projects + Deep Research) — the mentor-project setup described above.
  • Claude (Projects) — excellent for long, nuanced documents and careful reasoning.
  • Gemini (Gems) — Google’s take on customizable AI assistants, strong with Google Workspace context.
  • Grok — X-native, fast-moving, with real-time social context.
  • Perplexity (voice mode) — superb for research, fact-checking, and cited answers on the go.
  • Whisper Flow — speak-to-text dictation that makes long, rich prompts effortless.
  • NotebookLM — upload your sources and interrogate them; a brilliant study companion.

And the meta-resource: the deep-research habit itself. Any time you want to get smart on something fast, run the research, read it like an article, check the sources. Push the limits of your plan’s research credits — curiosity compounds.

Where a human mentor still wins

Here’s the honest coda: AI is the best tutor — infinitely patient, endlessly knowledgeable, always available. But a tutor gives you knowledge. A mentor gives you judgment: the scar tissue, the pattern-matching from lived experience, the “don’t do what I did” that no training corpus fully captures.

The winning combination isn’t AI or humans — it’s both. Use your AI mentor to learn faster than ever, and a human mentor to make sure you’re learning the right things. That’s exactly what MentorNet is being built for: pairing the speed of AI with the wisdom of people who’ve actually walked the path.

Not sure where to start? Take the 60-second Mentor Match Quiz and tell us what you’re working on — we’ll help you find the human guidance to go with your new AI superpower.

LET’S KEEP IN TOUCH!

We’d love to keep you updated with our latest news and offers 😎

We don’t spam! Read our privacy policy for more info.

MentorNet
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.