January 15th, 2026
Happy Thursday!
And… Happy New Year! Welcome to 2026 and another issue of HOW TO // AI! We’re glad you made it! We’ve taken a bit of time off over the holidays so we have a lot to cover in today’s issue!
Here’s what we’ve got for ya:
🎯 The 2026 AI Starter Stack: Learn Faster, Work Smarter
🎓 New ChatGPT Prompt Packs from OpenAI Academy
🩺 ChatGPT Health - Coming Soon!
💻 Google Antigravity vs Cursor vs Codex
🧠 New Guided Learning Mode in Gemini
Let’s get started!
AI-native CRM
“When I first opened Attio, I instantly got the feeling this was the next generation of CRM.”
— Margaret Shen, Head of GTM at Modal
Attio is the AI-native CRM for modern teams. With automatic enrichment, call intelligence, AI agents, flexible workflows and more, Attio works for any business and only takes minutes to set up.
Join industry leaders like Granola, Taskrabbit, Flatfile and more.
🎯 The 2026 AI Starter Stack: Learn Faster, Work Smarter
If you’re new to AI, your goal in 2026 isn’t to learn everything—it’s to build a repeatable workflow you can use weekly. Start with one main chatbot, one place to capture notes, and one real-life use case you’ll practice on (work emails, a side project, a spreadsheet, school, or job searching).
A simple rhythm that works: learn (20 min) → practice (20 min) → ship (20 min). “Ship” can be as small as a cleaner email, a one-page plan, or a better outline. Consistency beats intensity.
Learn the fundamentals that actually matter
You don’t need to code to benefit from AI, but you do need three basics:
Prompting: Give context, constraints, and an example of what “good” looks like.
Verification: Ask for sources, cross-check claims, and have the model flag assumptions.
Workflow: Use AI like a process: idea → outline → draft → revise → final.
Your highest-ROI beginner skills: summarizing, rewriting, brainstorming, planning, and creating reusable templates (checklists, SOPs, scripts).
Tools to explore (without getting overwhelmed)
Think of your tools in layers: (A) a main LLM, (B) specialist helpers, (C) automation once you’re consistent.
A) Main LLMs (your daily drivers)
ChatGPT — “all-around” assistant for writing, planning, tutoring, and structured outputs.
Claude — great for clean writing, long documents, and thoughtful reasoning.
Gemini — strong for Google ecosystem workflows (Docs, Gmail, Sheets) and general help.
Grok — useful for unbiased and fast answers and current-events insight - cool graphics too!
B) Specialist tools (pick 1–2, not 12)
Perplexity — “answer + sources” research tool; great for quick learning and citations.
Grammarly or LanguageTool — fast grammar + tone cleanup for everyday writing.
Canva — quick designs with AI help (thumbnails, social posts, simple visuals).
Pro tip: Choose one “home base” (Google Docs/Notion) and make AI feed into it.
Best Prompts for Learning AI in 2026
Let’s not wrap up without pointing out the importance of learning how to properly prompt. Learning how to prompt is the difference between AI being a fun toy and a real productivity tool. A good prompt gives the model the context, constraints, and “definition of done” it needs—so you get useful, accurate results faster with fewer back-and-forth messages. So when you ask your favorite AI friend for something, think for a second about what you want, for whom, in what format, and how you’ll verify it
Here are some great prompts to help you learn AI in 2026:
Write me a 7-day AI learning plan for a beginner with 30 minutes per day, including daily mini-projects.Explain [AI term] like I’m 12, then like I’m a working professional, with one example for each.Turn these messy notes into a clear checklist with steps, time estimates, and a “definition of done”: [paste notes]Summarize this article in 5 bullets, include 3 key quotes, and give me 3 practical ways to use this today: [paste link/text]Rewrite this to be clearer and more friendly, keeping the meaning the same. Give me 3 tone options: [paste text]Ask me 10 questions to clarify my goal, then propose 3 approaches and recommend the best one.Create a reusable template for [task] with inputs, steps, output, common mistakes, and a final quality check.And final tip, use AI daily and your skills will get better every day!
The Rundown: OpenAI offers Prompt Packs with pre-built prompts for ChatGPT, tailored to roles like sales, engineering, and management to boost workflows.
Details:
Packs span sales outreach, customer success, product strategy, engineering debugging, HR recruiting, IT scripting, team management, and executive decisions.
They include prompts for research, data analysis, visualizations, and automation with built-in safeguards.
Designed for quick integration into daily tasks across industries.
Takeaway: Prompt Packs simplify advanced AI use by offering ready-to-go queries. Try one matching your role to cut down on setup time. Adopting them now sharpens your edge in AI-driven productivity.
The Rundown: OpenAI launched ChatGPT Health, a dedicated space in ChatGPT that securely connects your personal health data from apps and records to deliver tailored insights and support.
Details:
Integrates with Apple Health, MyFitnessPal, medical records via b.well, and more, grounding responses in your real data for tasks like decoding lab results or prepping for doctor visits.
Keeps health info in an isolated, encrypted compartment with strict privacy rules—no training on your data, easy disconnects, and deletion controls.
Available now via waitlist for ChatGPT users (Free through Pro), starting in the US with physician-reviewed safety features.
Takeaway: ChatGPT Health brings personalized AI to everyday wellness without replacing doctors. Join the waitlist if you track health metrics to start experimenting with data-driven queries. As AI steps deeper into health, it’s a smart moment to test these tools responsibly and stay ahead in personal health management.
The Rundown: Google launched Antigravity, a free AI IDE core for agent-first development, emphasizing trust and abstractions, while rivals like Cursor offer AI-assisted editing in a VS Code fork, and OpenAI's Codex provides agentic coding via CLI and extensions for task delegation.
Details:
Antigravity focuses on browser-in-the-loop agents for UX, production apps with verification, and enterprise orchestration, available as a no-cost Mac download, differing from Cursor's autocomplete and chat in a familiar editor, or Codex's repo navigation for bug fixes and tests.
While Cursor boosts productivity with targeted edits and full autonomy modes, Antigravity prioritizes cross-surface agents and feedback; Codex stands out with local/cloud hybrids for secure, reasoning-based assistance in IDEs like VS Code.
All three tools accelerate coding, but Antigravity targets agent orchestration for pros, Cursor suits beginners with intuitive AI, and Codex excels in enterprise with security-aware models like GPT-5.2.
Takeaway: Google Antigravity enters the AI coding arena with a focus on agent-driven workflows, offering a fresh alternative to Cursor's editor-centric approach and Codex's comprehensive agents. Developers should test Antigravity for complex projects needing orchestration, while sticking with Cursor for quick edits or Codex for secure enterprise tasks. As AI tools evolve, blending these could redefine productivity, so experiment across them to find your optimal stack.
The Rundown: Google Gemini’s new Guided Learning mode turns Gemini into a step-by-step tutor—focused on helping you understand how and why, not just spitting out answers.
Details:
Guided Learning breaks topics into clear steps and uses questions to keep you actively learning (more “teach me” than “tell me”).
It can deliver richer study experiences (like visuals/interactive elements) to help you retain concepts, not just finish a task.
How to try it (as shared in the post): open Gemini on the web, start a new chat, pick Guided learning from the mode list, then ask a question or upload a document to study.
Takeaway: This is a real shift from “chatbot answers” to “AI tutoring workflows,” which is why it’s being positioned as a study/learning upgrade inside Gemini.
If you’re studying, skilling up, or trying to learn from documents, it’s designed to keep you moving in structured steps instead of getting lost in random follow-ups. It also lands in the middle of an AI learning arms race—Google is explicitly pushing learning modes as competitors roll out their own study-focused experiences.
🤔 Got Questions?
Have a question you want answered? What do you want to learn about in the next newsletter? What are your 2026 learning goals?? Email [email protected] and you just might be featured in an upcoming issue!
This newsletter is your starter kit for mastering AI with confidence. We keep things simple, show real examples, and focus on quick wins you can repeat.
Stay curious, stick with it, and watch your skills grow week by week!
Until next time :)