Guide: 005

First 5 Things to Try

Never used AI? Start here. Five things you can do right now, in under 5 minutes each.

01

Ask ChatGPT to Explain Something

Go to chat.openai.com (free). Type "Explain [topic] like I'm 12." Try it with blockchain, quantum computing, or anything you've wondered about. Clear, jargon-free answer in seconds.

02

Have Claude Summarize Something Long

Go to claude.ai (free). Paste a long article, report, or email thread. Ask: "Summarize in 5 bullet points." Claude is especially good at handling long documents and pulling out what matters.

03

Generate an Image from a Description

In ChatGPT, type "Create an image of [description]." Try: "a cat in a spacesuit on Mars, oil painting style." Generated in seconds. This is DALL-E 3 working behind the scenes.

04

Let AI Help You Write an Email

Tell ChatGPT or Claude: "Help me write a professional email to [person] about [topic]. Keep it friendly but direct." Edit the result to match your voice. Save 10 minutes per email.

05

Try Perplexity Instead of Google

Go to perplexity.ai and search something you'd normally Google. Instead of 10 blue links, you get a direct answer with sources cited. Like having a research assistant who reads everything.

By Role
AI for Your Job

Specific AI use cases based on what you actually do for work.

Manager / Executive

  • Summarize meeting transcripts into action items
  • Draft strategy memos and board presentations
  • Analyze team reports and spot trends

Marketer

  • Generate ad copy variations for A/B testing
  • Create social media content calendars
  • Analyze competitor messaging and positioning

Developer

  • Autocomplete code with Copilot or Cursor
  • Debug errors by pasting stack traces into Claude
  • Generate unit tests and documentation

Writer / Creator

  • Brainstorm ideas and overcome writer's block
  • Edit and improve drafts for tone and clarity
  • Research topics with AI-powered search
Free Courses
Learn from the Best

World-class AI courses from MIT, Stanford, Harvard, and top researchers. All completely free.

Neural Networks: Zero to Hero

ANDREJ KARPATHY

Build neural networks from scratch in code — from backpropagation to GPT. By the former Tesla AI Director and OpenAI founding member. The best hands-on AI course that exists.

Free

6.S191: Intro to Deep Learning

MIT

MIT's fast-paced deep learning course. Covers neural networks, CNNs, transformers, generative models, and reinforcement learning. Updated every year with new lectures and hands-on labs.

Free

CS229: Machine Learning

STANFORD

Andrew Ng's legendary ML course. Supervised learning, unsupervised learning, reinforcement learning, and statistical pattern recognition. The course that launched thousands of ML careers.

Free

CS224n: NLP with Deep Learning

STANFORD

The best free NLP course. Word vectors, transformers, pre-training, reasoning, and agents. Updated annually — the 2024 edition covers LLMs, RLHF, and AI agents.

Free

CS231n: Deep Learning for Vision

STANFORD

Originally created by Fei-Fei Li and Andrej Karpathy. Convolutional neural networks, image recognition, detection, segmentation, and generative models. The definitive computer vision course.

Free

6.034: Artificial Intelligence

MIT OPENCOURSEWARE

Classic AI foundations — search, constraint satisfaction, logic, planning, and learning. Taught by the legendary Patrick Henry Winston. Full video lectures, assignments, and exams.

Free

CS50 AI with Python

HARVARD

Search algorithms, knowledge representation, probability, optimization, ML, neural networks, and NLP. Hands-on Python projects including game dev and handwriting recognition.

Free

Practical Deep Learning for Coders

FAST.AI

Created by Jeremy Howard (former Kaggle #1). Top-down practical approach — you build real things from lesson 1. Computer vision, NLP, recommendation systems using PyTorch.

Free

ML Specialization

ANDREW NG / DEEPLEARNING.AI

3-course specialization — the updated version of Ng's original 2012 course. Free to audit on Coursera. Over 4.8 million learners. Rated 4.9/5.

Free to audit

ML Crash Course

GOOGLE

15-hour self-study course with animated videos, interactive visualizations, and 130+ exercises. Covers ML fundamentals, large language models, and responsible AI.

Free

Elements of AI

UNIVERSITY OF HELSINKI

Non-technical intro to AI for everyone. No programming or advanced math required. Over 1 million students. Perfect starting point if you're completely new to AI.

Free + Certificate

Neural Networks for ML

GEOFFREY HINTON / U OF TORONTO

Learn neural networks from the "Godfather of Deep Learning" himself. 2024 Nobel Prize winner. Covers architectures, backpropagation, Boltzmann machines, and learning theory.

Free
YouTube
AI YouTube Channels

Learn from the actual researchers building AI. Lectures, paper breakdowns, and visual explanations.

Andrej Karpathy

~990K SUBSCRIBERS

Former Tesla AI Director, OpenAI founding member. Deep-dive lectures on LLMs, neural networks, and transformers. His "Zero to Hero" series and 3.5-hour LLM deep dive are landmark content.

LECTURES

3Blue1Brown

~6.5M SUBSCRIBERS

Stunning visual math explanations. His neural network series makes backpropagation, gradient descent, and attention mechanisms intuitive. Watch these before any course.

VISUAL MATH

Yannic Kilcher

~271K SUBSCRIBERS

In-depth 30-minute breakdowns of AI research papers. Transformers, diffusion models, new architectures. The best way to stay current on AI research without reading every paper.

PAPER REVIEWS

Two Minute Papers

~1.7M SUBSCRIBERS

AI and computer science breakthroughs explained in short, visual videos. Covers generative AI, simulation, robotics, and more. Complex research made fun and accessible.

QUICK EXPLAINERS

Lex Fridman

~4.9M SUBSCRIBERS

Long-form interviews with the biggest names in AI — Karpathy, LeCun, Altman, Sutskever, Hassabis, Hinton. Direct access to the thinking of the people building AI.

INTERVIEWS

StatQuest with Josh Starmer

~1.3M SUBSCRIBERS

Breaks down statistics and ML into bite-sized visual pieces. If you struggle with the math behind ML, this channel will save you. Unique style that makes complex concepts click.

ML FOUNDATIONS

Sentdex

HANDS-ON PYTHON AI

Project-based Python programming for ML and AI. You code along and build real things — neural networks from scratch, TensorFlow, reinforcement learning. Learn by doing.

CODE-ALONG

AI Explained

AI NEWS & ANALYSIS

Makes complex AI developments accessible to non-technical audiences. Clear, measured analysis of new model releases, benchmarks, and capabilities. No hype, just signal.

NEWS ANALYSIS
X / Twitter
Follow the Researchers

The people actually building AI. Follow them for insights you won't find anywhere else.

Roadmap

30-Day AI Learning Plan

If you feel overwhelmed, do less. One month of steady use will teach you more than a giant binge of random tutorials.

Week 1: basic prompting

  • Use ChatGPT or Claude daily for one real task
  • Try summarizing, rewriting, and brainstorming
  • Learn what the models do well and where they fail

Week 2: work workflows

  • Apply AI to email, notes, research, or planning
  • Measure time saved on one repeated task
  • Keep prompts simple and concrete

Week 3: tool comparison

  • Compare assistants, search tools, and one specialist tool
  • Use our AI tools directory to narrow choices
  • Stop using anything that adds friction

Week 4: go deeper

  • Take one free course or follow one strong YouTube channel
  • Build a weekly habit instead of starting over every month
  • Use our newsletter to stay current
FAQ

Learn AI FAQ

What is the fastest way to start learning AI?

Use one assistant, one AI search tool, and one real task from your daily work. Practical repetition beats passive reading at the beginning.

Do I need math or coding before learning AI?

No for everyday use. You only need deeper math and code later if you want to build models or move into ML engineering.

Which free AI course should I start with?

Elements of AI and Google's ML Crash Course are accessible beginner options. Karpathy's Zero to Hero and fast.ai are better if you want to build things.