Happy New Year! Kicking off 2026 with AI is one of the smartest ways to start the year, because small, consistent reps compound fast.This issue is designed to help you set clear intentions and turn them into systems you can actually execute. 

🧭 Founder’s Corner: My 5 AI resolutions for 2026, focused on distribution, automation, data architecture, social growth, and creativity.

🧠 AI Education: ChatGPT Part 1, a beginner-friendly tour of the Free tier so you stop guessing and start using models, modes, and limits with intention. 

🧺 10-Minute Win: Thematic Basket Builder, a fast workflow to build a trend-based watchlist in minutes and validate tickers in Google Sheets. 

Let’s dive in. 

1) Exclusive: Groq investor sounds alarm on data centers

Summary: A Groq investor warns that too many data centers are being built without committed tenants, calling the “build it and they will come” strategy a trap and predicting a financing crisis in 2027–2028 for speculative landlords.

Why it matters: The AI boom is increasingly an infrastructure + financing story (power, land, capex). If funding tightens, it hits chips, cloud capacity, and software multiples fast.

2) Apple faces high stakes in 2026 as AI-powered Siri launch slips 

Summary: Apple is aiming to launch its long-delayed AI-upgraded Siri in 2026, a pivotal part of “Apple Intelligence.” Analysts frame it as one of Apple’s most important software releases in years, tied to sparking another iPhone upgrade cycle (with newer devices required for on-device AI).

Why it matters: If Apple nails Siri, it can reignite the upgrade engine and defend its ecosystem. If it misses again, the market narrative shifts from “late but fine” to “structurally behind” in AI.

3) Bernie Sanders slams Amazon for replacing workers with automation: “Maybe it’s time to tax robots”

Summary: Sanders argues companies like Amazon are incentivized to replace workers because robots don’t require wages or benefits, and he calls for taxing automation (“robots”) and using the revenue to support working families impacted by job displacement.

Why it matters: This is a clean signal that AI/automation is becoming a tax + labor policy fight, not just a tech story. Expect more pressure on hyperscalers and data-center buildouts as “who wins, who loses” becomes political. 

4) Sam Altman is hiring someone to worry about the dangers of AI

Summary: OpenAI is hiring a Head of Preparedness to forecast and mitigate severe-risk scenarios (mental health harms, cybersecurity misuse, and frontier capability risks), building a more operational safety pipeline as models accelerate. 

Why it matters: Agents are moving into real systems (browsers, files, workflows). Safety/evals are becoming org chart + budget items, and that’s going to shape what enterprises will (and won’t) deploy. 

5) CES 2026 Preview: 6 biggest trends to watch 

Summary: A practical preview of CES 2026’s biggest themes—AI baked into laptops, TVs, wearables/health tech, smart-home gear, and plenty of robotics—plus what to watch from the major players.

Why it matters: CES is the “AI distribution” show: it’s where AI becomes defaults in devices people already buy. The winners won’t be the flashiest demos; they’ll be the products that make AI invisible and useful.

2025 was a year of nonstop AI change, and it forced me to learn through the noise instead of waiting for clarity. I used the past 12 months to build a foundation so I can move faster and build smarter systems in 2026. There are still concepts I want to understand, tools I want to pressure-test, and workflows I want to automate. The start of the new year is the perfect time to set intentions and turn them into goals I can actually track. Building in public keeps me accountable to this community and to myself. In 2026, my focus is simple: build better systems, teach what I’m learning, and grow this platform with intention. Here are my five AI resolutions for 2026.

Resolution 1: Upgrade the website into a true distribution hub

I’ve discovered several bugs and gaps while navigating MindOverMoney.ai that make the site harder to use and harder to find. One issue is discoverability in search engines and AI scrapers, which I’ve started to address with custom meta tags on each post. In 2026, SEO and AI searchability will be a core part of my distribution strategy. My plan is to test a few specific ways to drive organic traffic, with the goal of turning that traffic into new subscribers. I’ll use Ghost analytics to track what’s working and make adjustments in real time. This focus sets the site up for long-term growth, but it will require constant iteration. I also want the site to guide people to the best content fast, so new visitors understand what Neural Gains Weekly is and why it’s worth subscribing.

I won’t grow subscribers if the site is difficult to navigate. There’s a glaring issue I need to solve: the signup process is fragmented, and the dreaded spam problem keeps getting in the way. Ghost requires email confirmation to complete a subscription, but the base code that drives the experience does a poor job of making that clear. Even worse, the confirmation email often lands in spam, which lowers the odds that someone completes their subscription. In 2026, one of my first projects is building a landing page that makes subscribing simple and sets expectations from the start. It will walk people through completing registration and show them how to keep future emails out of spam. This should stabilize onboarding and dramatically increase the number of people who complete the signup process.

Resolution 2: Automate workflows behind the newsletter

Creating content for Neural Gains Weekly is time-consuming, and it can be hard to carve out time for other projects. I need to make time to focus on these resolutions and execute the projects that align with my 2026 goals. How do I free up time while building new skills? Automation. There are already tools that can automate pieces of my workflow. And as AI agents improve, I expect even more opportunities to open up. My initial step is figuring out where to start, which really means choosing the single easiest task to automate first. The goal is not to build a full one-button newsroom overnight. It’s to automate one small piece, prove it works, then stack wins from there. Next, I’ll pick the simplest tech stack that gets the job done, ideally without adding more subscriptions. The ultimate goal is to automate as much of the manual work as possible, so I can spend more time building, writing, and experimenting.

Resolution 3: Learn data architecture so I can build smarter with AI

Education has always been the foundation of how I grow. I’m naturally curious, and I like to understand the why and the how. AI has introduced a new set of concepts that will matter for my growth, both personally and professionally. My first challenge in 2026 is learning data architecture, especially as agentic AI starts to reshape the infrastructure underneath modern systems. I’m fortunate to work on IT and AI projects in my career, so I’ve picked up the basics of data architecture. It’s a concept I naturally gravitate toward, but I still lack depth. A focused learning plan will be crucial if I want to build and deploy agentic systems. It will also matter as I build automation, because I need to understand how tools connect, whether that’s through APIs or MCP. I view this as a mandatory curriculum, and it will only become more relevant as the AI ecosystem evolves in 2026.

Resolution 4: Build a social media strategy that drives growth

This is the simplest resolution, and also the most challenging. I’ve grown to 106 subscribers through direct engagement with friends, family and coworkers. I’ve picked up a handful through LinkedIn, but that channel is only as good as the effort I put into it. 2026 will be the year I commit to a social media strategy that drives traffic to the site and newsletter subscriptions.

My main focus is LinkedIn, where I’ll consistently share how I’m thinking about AI, engage with other AI leaders, and market my content. I also plan to research how the algorithm works so I can expand reach and drive engagement beyond my current network.

X (formerly Twitter) will be my secondary focus. Admittedly, I’ve put very little effort into this platform. I have a big decision to make: do I keep posting from the MindOverMoney.ai account, or switch to my personal account? There are pros and cons either way, but the bigger issue is consistency. I need to pick a lane and stick with it. I’m treating this like training: small reps every week, tracked over time, until consistency becomes automatic.

Resolution 5: Let creativity drive the build

I’ve written about this before, but AI has unlocked a creative side to me that’s been dormant. Creative projects keep me balanced and motivated to show up each week. I have a list of projects that could supplement the newsletter and add new formats to the ecosystem I’m building. I need to prioritize that list and choose projects that genuinely add to my growth. But I also want to enjoy the process and make space for creativity. This resolution is for me. It’s how I plan to keep my energy and commitment high all year. Balance will be important if I want to stay grounded and focused as the pace of AI keeps accelerating.

If you have only used ChatGPT as a “question box,” you are seeing maybe 20 percent of what it can do. Up to now, we have been building the foundation, including the full large language model series. You now understand the building blocks behind tools like ChatGPT. Now we pivot. In 2026, we are going to learn the platforms themselves, starting with ChatGPT’s Free tier. This 4-part series is a guided tour of what is possible, what is limited, and what is worth trying first. The goal is simple: stop guessing, and start using the product with intention.

Feature Index (Part 1)

The model story (what is answering you)

GPT-5.2 on Free

Free tier includes access to GPT-5.2, but it is capped. When you hit the GPT-5.2 message limit, ChatGPT will switch you to a smaller model until your limit resets.

Do Free users get to pick models?

In general, manual model selection is a paid-plan feature. Free is primarily the default experience.

Why models relate to cost

Better models cost more to run. More capability usually means more compute per answer. That is why the Free tier is generous, but capped.

What “Thinking” means 

“Thinking” is not a separate product. It is a deeper reasoning behavior. It is used when a task is complex or when a user selects a deeper option in a model picker (more common on paid plans). The important thing to understand is this: deeper thinking usually means slower, more careful answers.

How to “trigger” deeper thinking with your prompts

You cannot always manually turn on a deeper thinking mode on the Free tier. But you can nudge the system to respond more carefully by making the task feel like it requires deeper work.

Here are prompt signals that usually lead to a more thoughtful answer:

• Ask for a plan before the answer.First, outline the steps you will follow. Then answer.
• Force tradeoffs.Give me two options. Compare them. Then recommend one and explain why.
• Add real constraints.Keep it under 150 words. Use a table. Do not make assumptions. If you are missing info, ask me questions first.
• Ask for verification behavior.List what you are unsure about, and what you would need to confirm it.
• Ask for an error check.After you write the answer, review it for mistakes and rewrite anything unclear.

The modes story (ChatGPT is not one thing)

A beginner-friendly way to understand ChatGPT is: one app, multiple modes.

Mode 1: Standard chat

Fast back-and-forth. Great for writing, planning, learning, and brainstorming.

Mode 2: Search

Search is for up-to-date information with links. It is the mode you use when you want the model to look things up instead of guessing.

Mode 3: Shopping research

Shopping research is a deeper buyer’s-guide experience. It asks follow-up questions, searches products, and summarizes tradeoffs. It is available on Free, but limits can change.

Mode 4: Deep Research

Deep Research is designed for multi-step, in-depth research that produces a documented report with citations. This is not positioned as a Free feature today.

Mode 5: Agent mode (paid)

Agent mode is the “take actions for me” mode. It can browse websites and take steps on your behalf, while pausing for clarification or confirmation. This is paid-only today.

Free tier limits and upgrade triggers

What “Free with limits” usually looks like

• You can use the feature, but only a certain amount per time window.
• Tool limits are separate from text limits.
• Limits can tighten temporarily during peak demand.

What triggers upgrade prompts most often

• Heavy GPT-5.2 usage.
• Heavy tool usage (files, data analysis, images).
• Wanting paid-only modes like Agent mode.

One-screen recap

• ChatGPT has multiple modes. Chat is only one of them.
• Free includes GPT-5.2, but it is capped and can fall back to a smaller model.
• “Thinking” means deeper reasoning. It is slower but more careful.
• You can nudge deeper thinking by asking for steps, tradeoffs, constraints, and self-checks.
• Search is how you get current info with links and reduce guessing.
• Shopping research is a deeper buyer’s-guide mode and is available on Free with limits.
• Deep Research and Agent mode are real, but they are not Free features today.

🧺 Thematic Basket Builder 

Most investors don’t need “one stock pick.” They need a clean way to express a thesis without betting everything on a single name. A thematic basket does that: you pick a trend (AI chips, GLP-1 drugs, clean power, cybersecurity), then build a small set of companies that benefit in different ways. In 10 minutes, you’ll use ChatGPT to generate a basket and Google Sheets to auto-validate tickers so you can paste the result straight into a watchlist.

Step 1 — Define your theme and constraints (2 minutes)

Pick one theme and keep it tight:

• “AI Infrastructure” (chips, networking, cloud, power/cooling)
• “Cybersecurity”
• “Weight loss / GLP-1 ecosystem”
• “Uranium + nuclear”
• “Data centers”
• “Defense tech”

Then decide your constraints (quick gut calls are fine):

• Basket size: 8–15 names
• Geography: U.S. only (recommended for beginners)
• Risk: Lower / Medium / Higher
• What to avoid: “no microcaps,” “no pre-revenue,” “avoid penny stocks,” etc.

Step 2 — Have ChatGPT build the basket + copy/paste output (4 minutes)

Paste this prompt into ChatGPT:

Role: You are my thematic basket analyst.Goal: Build a thematic stock basket I can paste into Google Sheets and validate automatically.

My theme: [PASTE THEME]

Constraints: [PASTE CONSTRAINTS]

• Basket size: 
• Geography: 
• Risk level: 
• Exclusions: 

Tasks:

1. Propose a basket with 4 groups:
   • Core Leaders (3–4)
   • Enablers/Picks-and-Shovels (2–4)
   • Challengers/Up-and-Comers (2–4)
   • Wildcard (1–2) (higher risk, optional)
2. For each company, output:
   • Ticker in GOOGLEFINANCE format (e.g., NASDAQ:NVDA, NYSE:V)
   • Company name
   • Role in the theme (Leader/Enabler/Challenger/Wildcard)
   • 1-sentence rationale (plain English)
   • 1 key risk (plain English)
3. Output the basket twice:
   • A clean table for reading
   • Then CSV ONLY (no commentary) with headers: Ticker,Company,Role,Rationale,Key Risk

Rules:

• If you are not confident a ticker is correct, mark it as CHECK instead of guessing.
• Keep it beginner-friendly.
• No buy/sell recommendations.

You should end with a CSV block that looks ready to paste.

Step 3 — Paste into Sheets and auto-validate tickers (3 minutes)

1. Open Google Sheets → New blank sheet → click A1.
2. Paste the CSV from ChatGPT (it should split into columns automatically).
3. Add two new columns:

• Column F header: Price Check
• Column G header: Status

In F2, paste this formula and fill down:

=IFERROR(GOOGLEFINANCE(A2,"price"),"⚠️ NOT FOUND")

In G2, paste this formula and fill down:

=IF(F2="⚠️ NOT FOUND","Replace / Verify","OK")

Now your sheet flags bad tickers automatically. No manual hunting required.

Step 4 — Fix anything flagged and finalize your copy-paste basket (1 minute)

If any rows show “⚠️ NOT FOUND”, copy those tickers and paste this into ChatGPT:

These tickers failed GOOGLEFINANCE() validation in Google Sheets: [PASTE FAILED ROWS]

Replace each failed ticker with the closest alternative that fits the same Role in the theme.Output CSV ONLY with the same headers: Ticker,Company,Role,Rationale,Key Risk.Use GOOGLEFINANCE-friendly tickers (e.g., NASDAQ:___, NYSE:___).If you can’t confidently validate, write CHECK.

Paste the corrected CSV into your sheet.

The Payoff

You now have a theme basket you can reuse: a structured set of companies mapped to a thesis, with risks listed, and a Sheet that flags ticker issues automatically. This is a better way to learn sectors than doomscrolling headlines—because you’re forcing yourself to answer: “Who benefits, how, and what could break?”

Transparency & Notes for Readers

• All tools are free: ChatGPT + Google Sheets. GPT-5.2 is available on Free with usage limits.
• Ticker validation: GOOGLEFINANCE() doesn’t support every ticker/region; “NOT FOUND” can mean “unsupported,” not “fake.”
• This is a watchlist workflow: it helps you organize research; it does not tell you what to buy.
• Educational workflow — not financial advice.