Hey everyone! 👋

AI is only as smart as the way we feed it information. In AI Education, we dive into context windows and chunking so long texts don’t overwhelm your prompts. This week’s 10-Minute Win turns that same discipline toward your finances with a personalized emergency fund target and plan. And in Founder’s Corner, we zoom out to the bigger picture of AI at work, and what the latest adoption data really means for your career and how to be the person who’s ready for what’s coming.

1) GPT-5.1: A smarter, more conversational ChatGPT

Summary: OpenAI rolled out GPT-5.1 with two main flavors: Instant (warmer, faster, better at following instructions) and Thinking (more adaptive reasoning with clearer, less jargony explanations). It also adds built-in controls so users can tune tone and style without prompt hackery.

Why it matters: This is a baseline upgrade to what “default ChatGPT” can do. Smarter + easier to steer means more people will quietly push real work—planning, research, money decisions—through AI every day.

2) Piloting group chats in ChatGPT

Summary: ChatGPT now supports group chats (pilot in a handful of countries), letting up to 20 people plus the model collaborate in the same thread. Group chats live in their own space, don’t share personal memory, and are powered by GPT-5.1 Auto with search, files, images, and voice.

Why it matters: This pushes AI from solo assistant to social workspace—think trip planning, study groups, small teams. Expect a wave of “AI in the group chat” workflows and products.

3) Anthropic to invest $50 billion to build data centers in the U.S.

Summary: Anthropic (Claude) announced a $50B program to build custom AI data centers in Texas and New York with partner Fluidstack, with the first sites coming online in 2026 and thousands of jobs attached.

Why it matters: Another monster capex signal that AI infrastructure is the new critical utility. Chips, power, cooling, and land are the real choke points—and Anthropic is moving to own more of that stack instead of just renting from clouds.

4) Poll: People worry about AI’s impact, but not their jobs

Summary: A new multi-country poll finds most adults are worried about AI’s impact on the economy and society, but far fewer think their own job is at serious risk. Many expect disruption in the abstract while assuming they’ll personally dodge it.

Why it matters: Classic “it’ll hit someone else” psychology. For you, that’s the gap to exploit: calmly mapping where AI is actually biting first (tasks, not titles) and adjusting skills and roles accordingly.

5) AI boom fuels fresh wave of legal tech investments

Summary: Investors have pushed $750M+ into AI-driven legal startups in recent weeks alone—GC AI, Clio, Legora, DeepJudge, SpellBook, EvenUp, Eve, and others—backing tools for drafting, research, contract analysis, and litigation support.

Why it matters: Legal is a great case study for AI as margin expansion: high billable rates, repetitive knowledge work, and lots of text. If AI can reliably take chunks of that workflow, you get a blueprint for what might happen next in accounting, compliance, banking ops, and other white-collar domains.

Last time, you learned how to design your ask: name the goal, share only the right facts, choose a format, set limits, and (optionally) show a tiny example. That is prompt engineering in plain English—shaping your words so the model can follow a clear pattern.

This week adds the constraint behind every good prompt: the model reads on a fixed-size page. If your message is longer than that page, the overflow is not read. Even a well-written prompt can miss key details when the input is too long.

The fix is chunking. You split long text into small, self-contained pieces, give each piece a short title, add a 1–3 line summary, and order the pieces so the most important ones appear first. You’re still doing prompt engineering—now with better content design. Clear prompts plus well-designed chunks help the model see what matters, stay on topic, and answer cleanly.

Core lesson — concepts first

Context window (the page size): A model reads a fixed amount of text at once. Anything past the limit is ignored.

Chunking (split): Break long content into small units that make sense on their own. Each chunk should answer: “What is this, and why does it matter?”

Titles (label): Give each chunk a short, direct title so the model understands the input. Examples: “Groceries — Outliers,” “Restaurants — Trend,” “Transport — New Charges.”

Summaries (condense): Start every chunk with 1–3 lines that highlight decisions and exceptions. Keep only the facts that affect the answer.

Ordering (prioritize): Put chunks in the order the model should read them. Most relevant first. If the page fills, lower-priority chunks are the ones that fall off.

De-duplication (remove repeats): Repeated email quotes and boilerplate waste space. Keep one clear recap and the newest message.

When to split by topic vs. time:

• Topic (groceries, restaurants, transport): best for category comparison and spotting outliers.
• Time (week 1, week 2, week 3): best for trends and changes across the month.

For today, we will use topic-based splitting because it is easy to compare and scan.

Why this works: Short, labeled chunks fit the page. The model sees what matters and can answer cleanly.

Contrast & clarity — common mistakes vs. better habits

• Mistake: Paste a long prompt with no structure. Better: Split into small, titled chunks with 1–3 line summaries.
• Mistake: Hide the goal at the end. Better: State your goal first, then place the most relevant chunk right below it.
• Mistake: Keep every quoted email and header. Better: Keep one recap and the newest message only.
• Mistake: Vague titles like “Notes.” Better: Specific, scannable titles: “Groceries — Outliers,” “Transport — New Merchants.”

Examples that land

Example 1 — Bank statement: find category outliers:

Scenario: You have a long bank statement. You want the AI to tell you which categories look unusual this month.

Your goal: Find category outliers and say why.

The prompt: Goal: Find category outliers this month.Return three bullets: category, short reason, biggest driver. Use only the chunks below.

The chunks you paste:

Chunk — Groceries Summary (Outliers): Two charges over 100; new store “OrganicCo”; total higher than usual.Examples: OrganicCo — 10/12 — 128.42; FreshMart — 10/21 — 112.09.

Chunk — Transport Summary (Outliers): New ride service started; three weekend trips; total up vs last month.Examples: RideCo — 10/05 — 31.20; RideCo — 10/12 — 28.70.

Chunk — Restaurants Summary (Trend): Down 75 vs last month; fewer visits; one 60 dinner.Example: Bistro — 10/14 — 60.00.

What you should get back:

• Groceries — higher total from two large trips — OrganicCo 128.42

• Transport — new ride habit increased total — three weekend trips

• Restaurants — down overall — fewer visits

Example 2 — Budget planning: explain a rise and suggest one fix

Scenario: Your grocery spending went up. You want the AI to explain the cause and propose one practical fix for next month.

Your goal: Explain the rise in plain English and suggest one specific action.

The prompt: Goal: Explain why groceries rose and propose one practical fix for next month. Answer in two short paragraphs. Use only the chunks below.

The chunks you paste:

Chunk — Groceries Summary (Headline): Up 120 vs last month; one-time party 95; two large trips this month. Examples: OrganicCo — 10/12 — 128.42; FreshMart — 10/21 — 112.09.

Chunk — Restaurants Summary (Context): Down 75 vs last month; two skipped outings; small takeout only.

What you should get back:

Paragraph 1: A plain explanation (one-time party + two big trips drove the increase).

Paragraph 2: One concrete fix (for example, plan one home-cooked dinner with leftovers to replace a restaurant meal).

One-screen recap

• The model reads a page of text at a time. Overflow is cut off.
• Chunk long text into small, titled pieces with 1–3 line summaries.
• Order chunks so the key part comes first.
• Remove repeats and filler. Clear beats long.
• Use topic-based chunks for category comparisons; use time-based chunks for trend questions.

🛟 Emergency Fund Sizing

Why this matters: Emergencies aren’t “if,” they’re “when.” In about 10 minutes, you’ll use AI + a simple Google Sheet to calculate how many months of expenses you personally need, turn that into a clear dollar target, and set up automatic reminders so your emergency fund grows in the background.

Step 1 — Build your essentials sheet (3 minutes)

Open Google Sheets and set up a simple table like this:

How to set it up:

1. In A1, type Category. In B1, type Monthly Essential ($).
2. Fill A2–A9 with only “keep-the-lights-on” categories (no dining out, no shopping, no streaming).
3. Enter your actual monthly amounts in B2–B9.
4. In B10, enter: =SUM(B2:B9)  
5. In C10, label that result Essential_Monthly_Spend.

🧠 If you completed Volume 1 (AI Budget Analyzer): Don’t start from scratch. Open your Vol. 1 budget sheet, filter it down to essential categories only (Housing, Utilities, Groceries, Transportation, Insurance, Minimum Debt Payments, Phone/Internet, Childcare). Use a simple SUM on those rows to get your Essential_Monthly_Spend and plug that total directly into this new sheet. You’re reusing the hard work you already did.

Step 2 — Let AI choose your Target Months (3 minutes)

Open ChatGPT (Free) and paste this prompt. Fill in your answers right in the prompt before you click send:

You are my household risk scorer. Based only on my answers, recommend a months-of-expenses target for an emergency fund and a stepwise plan. Answers:

• Employment type & stability (stable salary / variable / contractor): ___
• Number of dependents: ___
• Are you the sole earner? (yes/no): ___
• Income variability (commissions, tips, gig): low/med/high: ___
• Access to other cash buffers (HSA, savings, etc.): good/limited/none: ___
• Fixed expenses share (>60% of take-home?): low/med/high: ___
• Industry layoff risk next 12 months: low/med/high: ___

Rules:

• Map to Target Months: very low ≈ 3; low ≈ 4–5; moderate ≈ 6; elevated ≈ 7–9; high ≈ 10–12.
• Output: Target Months, Rationale (3 bullets), and Milestones (1 month → 3 months → full).
• Do not invent dollar amounts; I’ll do the math separately.

You should get something like:

• Target Months: 6
• Rationale: 3 bullets
• Milestones: 1 month → 3 months → full target

Step 3 — Turn months into dollars (and connect to Vol. 2) (2–3 minutes)

Back in Google Sheets:

1. In A12, type Target_Months. In B12, enter the number from ChatGPT (e.g., 6). In A13, type EF_Target_$. In B13, enter: =ROUND(B10 * B12, 0)  
2. That’s your emergency fund dollar target.
3. Add milestones:

🔁 If you completed Volume 2 (Net Worth Tracker): Open your Vol. 2 Net Worth sheet and:

• Find your current cash/emergency savings line.
• Add a new row under it: “Emergency Fund Target (from this 10-Minute Win)” and enter your EF_Target_$.
• Add another row: “Gap to Target” = Target – Current.Now your Net Worth tracker shows, at a glance, how far you are from a fully funded emergency fund — and you’ll see that gap shrink over time.

Step 4 — Make it automatic (2 minutes, desktop)

A) Save your plan (Docs/Notion)

• Copy ChatGPT’s Target Months, Rationale, and Milestones into Google Docs or Notion.
• At the top, add your EF_Target_$ and Monthly_Transfer from Sheets.

B) Set a recurring reminder (Google Calendar, desktop)

1. Go to calendar.google.com.
2. Click your next pay date → “More options.”
3. Title the event: Move $<Monthly_Transfer> → Emergency Fund.
4. Set Repeat: every month (or each pay period).
5. Paste your Docs/Notion link into the Description.
6. Add notifications (e.g., 1 day before, 1 hour before) → Save.

Now your emergency fund is a scheduled behavior, not a vague intention.

The Payoff

You’re no longer guessing at “3–6 months” because a blog said so. You have a personalized Target Months, a clear dollar number, milestones, and a recurring reminder that nudges you toward it every month. If your life changes (job, income, dependents), you can re-run the same risk prompt and tweak your target in minutes.

🧩 For returning readers: Volume 1 gave you clean spending data. Volume 2 gave you your Net Worth. This workflow plugs into both: you’re now connecting monthly essentials (Vol. 1) and cash on hand (Vol. 2) into a single, concrete emergency fund plan.

Transparency & Notes for Readers

• All tools are free: Google Sheets, ChatGPT Free, Google Docs/Notion, Google Calendar.
• Be conservative: If you’re torn between two targets (e.g., 5 vs. 6 months), choose the higher buffer.
• Privacy: Don’t paste account numbers or personally identifying info into AI; just categories and amounts.
• Educational workflow — not financial advice.

AI adoption in the workplace is accelerating as more companies weave it into their daily business operations. Recently, Wharton and McKinsey released comprehensive studies on the current state of AI in the enterprise. The theme is consistent: AI deployment has moved out of the exploration and experimentation phase. We’re entering an era of acceleration and accountability. Here are a handful of stats that I found eye-opening from the studies:

• 82% use Gen AI at least weekly (+10pp YoY), and 46% (+17pp YoY) daily. (Wharton)
• 62% of survey respondents say their organizations are at least experimenting with AI agents. (McKinsey)
• 88% say their organizations are regularly using AI in at least one business function. (McKinsey)
• 23% of respondents report their organizations are already scaling an agentic AI system somewhere in their enterprises. (McKinsey)
• 72% of business leaders report tracking formal, structured Return on Investment (ROI) metrics for their Gen AI technology investments. (Wharton)

I highly recommend reading through each of these studies, but the trend is clear. Leaders across the country are successfully using AI in their everyday work, while building AI systems into their current architecture. Many of you work in the corporate world, and the larger the organization, the harder it can be to understand where your company is on its AI journey. It can be challenging to experiment, as you might not have access to the right tools. A lack of visibility into other departments might hinder your ability to collaborate to solve problems where AI could be part of the solution. I’m sure you’ve experienced at least one of these challenges and can list additional scenarios that slow down AI adoption within your organization. Luckily, there is time to act and positively impact the trajectory of your department, and hopefully your company. Pilot mode is still the norm across industries, despite increases in overall AI usage. According to the McKinsey study, nearly two-thirds of respondents say their organizations have not yet begun scaling AI across the enterprise. The time to act is now, and I want to share 5 practical tips to help you be ready for the next phase of AI within your company.

1. Find ways to use AI daily

Seek any and all opportunities to build your AI skills in the workplace. This could be as simple as using Copilot to summarize emails, write or edit executive summaries, or build a talk track for an upcoming leadership readout. Start small, build a portfolio, and be creative. Your role might not have obvious AI use cases, and it will be up to you to find scenarios to test your skills and build confidence. 

2. Workflow first

AI adoption requires a mindset shift that prioritizes problem solving. Pilots will often fail when AI capabilities are forced onto an existing process. You can get ahead by building out detailed problem statements to rewire legacy workflows. This framework will be a foundation for any company that expects to see positive outcomes from AI initiatives. The McKinsey survey highlights this concept emphatically: ‘AI high performers—organizations seeing the greatest financial impact from AI—are nearly three times as likely as other organizations to report that they have fundamentally redesigned individual workflows in their deployment of AI (55% vs. 20%).’ 

3. Rethink ROI

Every company’s leadership will expect ROI from AI initiatives. That’s how the world works. And the ROI story will be challenging if decision makers and executives are not keeping up with the world of AI. It’s up to you, as an AI leader, to help bridge those gaps by evolving how you position ROI. Adopting a transformative mindset that pushes for growth and innovation, not just cost savings, will win in the long term. This simple reframe can help evolve how your organization approaches ROI and drive better alignment for AI infused projects. Here are two stats from the McKinsey study that hammer home this point:

• While cost efficiency is often the objective of AI efforts (reported by 82% of organizations), organizations achieving the greatest value (AI high performers) are more likely to also set growth (80%) and innovation (79%) as objectives.
• Organizations intending to use AI to bring about transformative change to their businesses are 3.6 times more likely to be AI high performers

4. Knowledge share

Not everyone around you will have the same level of AI knowledge and skill in the workplace. And that’s okay. It will be important to help others and seek mentorship as AI’s influence at work continues to evolve. I’m working through this with my team at work by building mechanisms to share AI use cases and best practices. I make it a point to talk about AI during 1x1s and gather ideas for the team. I’m not afraid to ask my peers for their opinions and thoughts on AI topics to broaden my viewpoints and understand how others are thinking about the future. There is no right or wrong way to share insights with others. A ‘rising tide lifts all boats’ mentality will ensure everyone is set up for success and can influence the future of work within their organization. 

5. Don’t forget about governance 

View AI as a powerful companion that requires human expertise, judgment, and validation, rather than a fully autonomous replacement. You cannot blindly trust that AI’s output is accurate and ready for mass consumption within your company. Prioritize auditing outputs and building a personal governance policy for AI use in your workflows. This will help build a framework that aligns (or will align) with broader governance strategies implemented at the corporate level. Prioritizing quality over speed is what actually accelerates AI usage. This is particularly important because inaccurate results remain one of the top three concerns leaders cite when using Gen AI (Wharton). 

Bringing it all together

AI is no longer a side project, it is becoming part of how work gets done. The good news is you do not need a new title or a massive budget to have an impact. Use AI daily, redesign workflows instead of bolting tools onto old processes, evolve how you talk about ROI, share what you learn, and keep quality and governance front and center. If you do that consistently, you will be ready for whatever phase of AI your company enters next, and you will have real examples to show the value you are creating along the way.

Goals & Milestones: