The State AI Readiness Divide: Which States Are Actually Helping Their People Learn and Use AI?

, , , , , , ,

The State AI Readiness Divide: Which States Are Actually Helping Their People Learn and Use AI?

An Underwood Partners Collaboration Evangelist White Paper

By Craig Underwood in collaboration with ChatGPT

Artificial intelligence is often discussed as if it were a single national wave: ChatGPT appears, companies scramble, schools debate cheating, workers worry, and everyone tries to keep up.

But that is not what is happening.

AI adoption is not spreading evenly across the United States. It is being shaped by public choices — by whether states are giving government employees secure AI tools, helping educators understand how to teach with AI, investing in public higher education, and making AI training available to residents.

In other words, the next AI divide may not be just between people who use AI and people who do not.

It may be between states that help their people learn AI — and states that leave them to figure it out alone.

That was the hypothesis behind this research. I wanted to know whether states differ meaningfully in their investment in and support for AI, and whether Massachusetts — my home commonwealth — is at or near the top.

The short answer: yes.

The longer answer is more interesting.

State AI readiness evaluation criteria.

For this analysis, we ranked all 50 states plus Washington, D.C. on a 100-point AI readiness index. The index was not designed to measure private-sector AI activity alone. Silicon Valley venture capital, university research labs, and AI startups matter, but they were not the core question.

 

The question was civic and public:

Which states are helping their residents, students, educators, public employees, and institutions actually learn and use AI?

The ranking used four categories:

 

We gave the most credit for concrete implementation: secure AI tools, training programs, statewide public AI portals, public higher-ed credentials, K–12 AI guidance, workforce-development programs, and published use cases.

We gave much less credit for task forces, commissions, executive orders, or legislation unless they were tied to real funding, training, tool access, implementation, or public-facing programs.

We also distinguished among three policy postures:

Policy posture Meaning
Responsible AI enablement The state encourages AI adoption while providing guardrails, training, privacy protections, and human oversight.
Neutral risk management The state is studying or regulating AI but has limited public evidence of broad adoption support.
Restrictive or chilling policy The state’s actions appear more likely to discourage useful AI adoption than to enable it.

That distinction matters. AI governance is not automatically anti-AI. In fact, the most impressive states are not choosing between innovation and responsibility. They are trying to do both.

The top 10 states and jurisdictions

The highest-ranked states are not merely “interested” in AI. They are putting AI into public systems.

Rank State / Jurisdiction Score Tier
1 Massachusetts 91 AI Readiness Leader
2 Utah 88 AI Readiness Leader
3 New York 86 AI Readiness Leader
4 New Jersey 84 AI Readiness Leader
5 California 82 AI Readiness Leader
6 District of Columbia 80 AI Readiness Leader
7 Maryland 78 AI Readiness Leader
8 Colorado 77 AI Readiness Leader
9 Ohio 76 AI Readiness Leader
10 Georgia 75 AI Readiness Leader

Massachusetts ranks first because it appears to have the most complete public AI readiness strategy. The Commonwealth has launched a secure enterprise AI assistant for executive-branch employees, partnered with Google and the Massachusetts AI Hub to offer no-cost AI training to residents, and developed K–12 AI guidance and educator resources through the Department of Elementary and Secondary Education.[1]

Utah ranks second and may be the national leader in education-focused AI readiness. Utah has announced a statewide AI Workforce Credential for public college and university graduates and a statewide partnership to bring Gemini for Education to K–12 schools.[2]

New York ranks third because of the scale of its state workforce rollout. The state has expanded AI training and a secure AI tool, AI Pro, to more than 100,000 state employees, while also investing in Empire AI, a major public-interest AI research consortium.[3]

New Jersey ranks fourth because it has one of the clearest examples of practical public-sector AI implementation. Its NJ AI Assistant has been used by about 20,000 state employees, with more than 300,000 sessions and 1 million prompts as of February 2026.[4]

California ranks fifth because it combines a massive AI economy with state and local government training, K–12 guidance, higher-ed activity, and workforce-disruption planning. But California’s posture is more complex than the top four because it is also heavily focused on regulation and labor-market disruption.

The bottom 10 states

The bottom states should not be described as “anti-AI.” That would overstate the evidence. In many cases, they are better described as states with limited public evidence of broad statewide AI enablement.

Rank State Score Tier
42 West Virginia 42 Limited / Early
43 Wyoming 41 Limited / Early
44 New Hampshire 40 Limited / Early
45 Alaska 39 Limited / Early
46 North Dakota 38 Limited / Early
47 Louisiana 37 Limited / Early
48 Nebraska 36 Limited Public Evidence
49 Kansas 34 Limited Public Evidence
50 South Dakota 32 Limited Public Evidence
51 Arkansas 30 Limited Public Evidence

This is an important methodological point. A low ranking does not prove that nothing is happening. It means I found less recent public evidence that the state is combining AI tool access, training, education strategy, workforce development, and responsible-use infrastructure into a coherent statewide approach.

That distinction matters because the AI landscape is changing fast. A state could move from the bottom tier to the middle quickly if it launched a statewide AI training portal, gave public employees secure AI access, published K–12 guidance, and used community colleges to scale workforce credentials.

The map tells the story.

The lead map shows a striking pattern: AI readiness is not evenly distributed.

Some states are already treating AI as a public-capacity project. They help people learn, practice, and use AI responsibly. Others are still closer to the study-and-watch stage.

National research supports this uneven picture. Code for America’s 2026 Government AI Landscape Assessment concluded that states are at different stages of AI adoption, ranging from readiness and piloting to implementation and impact.[5] The National Conference of State Legislatures has also tracked a surge in state AI activity, reporting that all 50 states, Puerto Rico, the Virgin Islands, and D.C. introduced AI-related legislation in 2025, while 38 states adopted or enacted around 100 measures.[6]

But legislation alone is not the same as readiness.

The leading states are doing more than passing bills. They are creating tools, training people, building public infrastructure, and giving educators and public employees permission to experiment within guardrails.

What the K–12 comparison adds

We also compared each state’s AI readiness rank with its K–12 education rank, using WalletHub’s 2026 public school system ranking as the K–12 comparison source.[7]

The result is one of the most interesting parts of the analysis.

Good schools help, but they do not automatically produce AI readiness.

Some states have both strong K–12 systems and strong AI readiness. Massachusetts is the clearest example: it ranked first in both our AI readiness index and the WalletHub K–12 ranking. New Jersey, New York, Maryland, Virginia, Connecticut, Rhode Island, Wisconsin, Pennsylvania, and Utah also appear in the high-education / high-AI zone.

These states are in the strongest strategic position. They have educational capacity and are beginning to convert that capacity into AI readiness.

But the scatterplot also reveals two other important patterns.

First, some states appear to be using AI as a leapfrog strategy. California, Georgia, Ohio, North Carolina, Colorado, Oregon, Arizona, Texas, and New Mexico do not rank at the very top on K–12 education, but they show stronger public evidence of AI support and investment than their education rankings alone would predict.

That may be strategically wise. AI training can scale faster than traditional education reform. Community colleges, public universities, workforce boards, libraries, and state agencies can help adults, small businesses, and public employees gain practical AI skills quickly.

Second, some states with relatively strong K–12 systems appear to be underleveraging that advantage. New Hampshire, Nebraska, North Dakota, Montana, Wyoming, Indiana, and Vermont all have educational strengths, but weaker public evidence of broad AI readiness.

That is the danger for high-education states: yesterday’s education advantage does not automatically become tomorrow’s AI advantage.

Four strategic positions for states

The AI readiness vs. K–12 education scatterplot creates four useful strategic categories.

  1. High K–12 / High AI: Compounding advantage

These states have the strongest starting position. They have relatively strong education systems and are also investing in public AI readiness.

Strategic imperative: move from AI readiness to AI advantage.

These states should publish public AI use-case libraries, scale AI literacy expectations, train public employees, support educator professional development, fund public higher-ed credentials, and measure productivity and learning gains.

The risk is complacency. Strong schools and strong universities are not enough. AI advantage requires structured implementation.

  1. Lower K–12 / High AI: Leapfrog opportunity

These states may be using AI as a way to accelerate workforce development and public-sector modernization.

Strategic imperative: use AI as a human-capital accelerator.

These states should focus on community colleges, adult learners, small-business training, teacher productivity, student supports, and accessible AI credentials.

The opportunity is speed. A state does not need to wait 20 years to improve every K–12 metric before helping people build useful AI skills.

The risk is inequity. If AI programs primarily benefit already-advantaged residents, AI may widen internal gaps.

  1. High K–12 / Lower AI: Underleveraged education advantage

These states may have strong educational foundations but less visible statewide AI implementation.

Strategic imperative: convert education strength into AI readiness.

The playbook is straightforward: issue practical K–12 AI guidance, train educators, give public employees secure tools, create public AI learning portals, and use public universities and community colleges as AI training engines.

The risk is losing ground to states with weaker traditional education systems but stronger AI implementation strategies.

  1. Lower K–12 / Lower AI: Double catch-up challenge

These states face the hardest strategic position.

Strategic imperative: start with practical, low-cost, high-leverage AI readiness moves.

They should create a statewide AI learning portal, train state employees in responsible AI use, publish K–12 AI guidance, partner with community colleges on short credentials, and build a public use-case library.

The risk is that AI becomes another driver of regional inequality.

The deeper point: AI readiness is public infrastructure.

The best states are not treating AI as a gadget. They are treating AI readiness as a form of civic infrastructure.

That means giving people:

  • access,
  • training,
  • permission,
  • examples,
  • guardrails, and
  • time to practice.

A state employee who has access to a secure AI assistant can learn by doing. A teacher with practical guidance can move beyond fear and confusion. A community college student with an AI credential can enter the labor market with greater confidence. A small business owner with free training can experiment without hiring a consultant. A resident with access to a public AI learning portal can begin without needing to be part of the technology sector.

That is the difference between AI as a private advantage and AI as a public capability.

Why Massachusetts matters

Massachusetts is a particularly useful case because it already had advantages: great universities, strong K–12 performance, a dense innovation economy, world-class hospitals, and a highly educated workforce.

But those advantages alone are not what made Massachusetts rank first.

What matters is that Massachusetts appears to be converting its existing advantages into public AI readiness.

The Commonwealth is giving executive-branch employees secure AI access. It is offering residents no-cost AI training. It is building a public AI Hub. It is supporting K–12 educators with guidance and resources.

That combination is the key.

Massachusetts is not simply an AI-rich state. It is investing to become an AI-ready Commonwealth.

The policy lesson

The policy lesson is not that every state should copy Massachusetts exactly. States differ in size, economy, demographics, politics, education systems, and fiscal capacity.

But every state can do five things:

  1. Give public employees safe AI access and training.
  2. Publish practical K–12 AI guidance.
  3. Use public higher education and community colleges to scale AI credentials.
  4. Create a public AI learning portal for residents and small businesses.
  5. Measure outcomes and publish use cases.

The first states to do these things well will have an advantage.

Not just a technology advantage. A civic advantage.

Final thought

AI adoption is often framed as an individual responsibility: learn the tools, keep up, do not fall behind.

But this research suggests that public leadership matters enormously.

Some states are making AI learning easier, safer, and more accessible. Others are leaving residents, teachers, workers, and public employees to navigate the AI era largely on their own.

That is the state AI readiness divide.

And it may become one of the most important competitiveness and equity issues of the next decade.

 

 

Endnotes

[1] Massachusetts state AI assistant materials; Massachusetts-Google free AI training announcement; Massachusetts AI Hub; Massachusetts Department of Elementary and Secondary Education AI resources.

[2] Utah System of Higher Education announcement of statewide AI Workforce Credential; Google announcement of Gemini for Education partnership with Utah K–12 schools.

[3] New York Governor’s Office announcement of AI Pro and AI training expansion to state workforce; Empire AI materials.

[4] New Jersey Innovation Authority, NJ AI Assistant project page and 2026 update.

[5] Code for America, 2026 Government AI Landscape Assessment.

[6] National Conference of State Legislatures, 2025 AI legislation summary and AI legislation database.

[7] WalletHub, 2026 public school system ranking.

Source links for the article endnotes

Massachusetts has an official enterprise AI assistant page, and Google announced no-cost AI training for Massachusetts residents through the Massachusetts AI Hub and Grow with Google. (Massachusetts Government)

Utah’s higher-ed system announced a no-cost AI Workforce Credential for more than 50,000 public college and university graduates, and Google announced Gemini for Education access for every Utah K–12 school beginning in the 2026–2027 school year. (Utah System of Higher Education)

New York announced expansion of AI education and AI Pro to more than 100,000 state employees, while New Jersey reported about 20,000 state employees using its AI Assistant, with more than 300,000 sessions and 1 million prompts as of February 2026. (Governor Kathy Hochul)

Code for America’s 2026 Government AI Landscape Assessment evaluates state AI adoption across readiness, piloting, implementation, and impact, while NCSL tracks AI legislation across states and reported that 38 states adopted or enacted around 100 AI-related measures in 2025. (Code for America)

For education context, Playlab reports that 36 states plus Puerto Rico had official K–12 AI guidance or policy frameworks as of May 2026, and FutureEd tracked 71 AI-in-education bills across 27 states during the 2026 legislative session. (learn.playlab.ai

Want to Get Started with AI? Learn How to Ask Better Questions. Three free or publicly accessible ways for students, educators, nonprofit leaders, and executives to learn the basics of AI prompt engineering.

, , , , , , , ,

One of the most common questions I hear from nonprofit leaders, educators, students, and executives is deceptively simple:

How do I get started with AI?

My answer: start with prompt engineering.

That phrase may sound technical, but the basic idea is very practical. Prompt engineering is simply the skill of designing and refining instructions so AI tools can produce more useful responses.1

For students, it can mean learning how to use AI to prepare for interviews, improve writing, explore career paths, or summarize complex material.

For executives, it can mean using AI to draft memos, pressure-test strategy, analyze donor or customer segments, prepare for meetings, test a hypothesis, or turn rough ideas into clearer plans.

The good news is that no one needs to spend thousands of dollars to learn the basics. Here are three online resources I recommend; IBM SkillsBuild and Microsoft Learn provide free/publicly available learning resources, while the OpenAI Academy collection is publicly accessible and designed around practical workplace AI skills.2

You can also start by reviewing my article GOOGLE’S 9 HOUR AI PROMPT ENGINEERING COURSE – IN 1 CARTOON!, summarized in this cartoon:

Use the right resource for the right audience: students, executives, or time-constrained workplace learners.

1. IBM SkillsBuild: Prompt Engineering – Shaping Better AI Responses

This is my top recommendation for students and workforce-development programs.

IBM SkillsBuild offers a structured prompt-engineering module. The course page says learners identify applications of different prompt types used in generative AI, explore prompt-engineering techniques, select techniques suited to specific tasks, and review best practices for effective prompts.3

IBM SkillsBuild also describes its broader AI learning resources for adult learners as offering free access, practical AI courses, and opportunities to earn credentials.4

For youth-serving organizations, career programs, and adult-learning settings, this is probably the strongest standalone option.

2. Microsoft Learn: Create Effective Prompts for Generative AI Training Tools

This is my top recommendation for busy executives, educators, and organizational leaders who want a fast, accessible introduction.

Microsoft Learn lists this as a beginner module with seven units, and identifies relevant audiences that include business users, K-12 educators, higher education educators, and school leaders. The module covers basic prompt-engineering concepts, elements of an effective prompt, and prompting best practices.5

For executives, I would use this as pre-work before a 60- to 90-minute AI workshop. It gives everyone a shared vocabulary without overwhelming them.

3. OpenAI Academy: ChatGPT at Work

OpenAI Academy offers a ChatGPT at Work collection that includes workplace-focused videos such as Introduction to Prompt Engineering, Advanced Prompt Engineering, ChatGPT for Data Analysis, Deep Research, ChatGPT Search, and other work-related ChatGPT topics.6

This is not a full course in the traditional sense, but it is a terrific executive appetizer. The videos are short, practical, and focused on real work: writing, analysis, research, data, reasoning, and productivity.

For leaders who are curious but time-constrained, this may be the easiest entry point.

How I Would Use These Resources

For a student program, I would assign the IBM SkillsBuild course and then run a hands-on session where students use AI to improve a resume, prepare for an interview, research a career, or summarize a reading.

For executives, I would assign the Microsoft Learn module and a few OpenAI Academy videos before a live workshop focused on real organizational use cases: preparing a board memo, drafting a strategy document, analyzing donor prospects, or creating a first version of an implementation plan.

For both audiences, I would add one essential lesson: AI is powerful, but it is not magic. Users still need to check facts, protect sensitive information, and apply human judgment.

Additional CHU Tips

A few tips I have learned recently through other courses I have taken:

  1. Before asking the AI engine to execute on your prompt, ask it to “Please ask me any clarifying questions to help you answer this prompt/ conduct this research/ create this image/ etc.” I predict you will be surprised at how good the questions are.  They almost always make me think deeper and more clearly about what I want from the prompt.
  2. If appropriate – and especially for larger research projects – you can ask the AI engine to “Convene a panel of experts to ask me questions about the prompt from different points of view.”
  3. To cut down on errors, mistakes, and “hallucinations,” include the following:
    1. “Do not make any thing up.”
    2. “Do not guess.”
    3. “Do not use any sources or research material older than [ __ ] months. (For AI related research, I never allow AI to go back any further than 6-12 months.)
    4. “Please footnote or endnote all references and facts and include a bibliography of sources used in your research, whether or not you actually used them in your response.”
    5. For all footnotes, endnotes and the bibliography, please make sure you include the date that source was published and not just the date you accessed the information.
    6. Please include links to the source documents whenever possible.
  4. Jeff Bussgang, who is the smartest person I know in this space, also recommends using “Please” and “Thank you” in our prompts, so I figure this can’t hurt!

Please let me know your best prompting insights as we all travel this journey of discovery together.

The Bigger Point

Prompt engineering is not just a technical skill. It is quickly becoming a basic workplace and learning skill.

The people who learn how to collaborate effectively with AI will have a major advantage. They will be able to think faster, draft faster, test ideas faster, and learn faster.

That does not mean AI replaces human judgment. In fact, the opposite is true. The better the human judgment, the better the AI collaboration.

So if you are a student, educator, nonprofit leader, or executive wondering where to begin, start here:

Learn how to ask better questions.

That is the first step toward getting better answers.

Endnotes

Suggested for web publishing: convert the superscript numbers in the article to bracketed links that jump to these notes, or keep the notes as a simple source list at the bottom of the post.

  1. IBM defines prompt engineering as writing, refining, and optimizing inputs to encourage generative AI systems to create specific, high-quality outputs. IBM also emphasizes that better prompts directly influence the quality, relevance, and accuracy of generative AI outputs. Source: IBM Think, “What Is Prompt Engineering?”
  2. IBM SkillsBuild describes its adult-learner AI resources as offering free access and courses in generative AI and related areas. Microsoft Learn makes the cited module publicly available. OpenAI Academy makes the cited ChatGPT at Work collection publicly accessible. Source: IBM SkillsBuild, “Artificial Intelligence”; Microsoft Learn, “Create Effective Prompts for Generative AI Training Tools”; OpenAI Academy, “ChatGPT at Work.”
  3. The IBM SkillsBuild course page for “Prompt Engineering: Shaping Better AI Responses” lists the course as eligible for registered learners, available in English, Arabic, Brazilian Portuguese, and Spanish, and four hours in duration. It says learners identify applications of different prompt types, explore prompt-engineering techniques, choose techniques suited to tasks, and review best practices for writing effective prompts. Source: IBM SkillsBuild, “Prompt Engineering: Shaping Better AI Responses.”
  4. IBM SkillsBuild describes its AI learning resources for adult learners as offering free access, AI courses, and opportunities to earn industry-recognized credentials. Source: IBM SkillsBuild, “Artificial Intelligence.”
  5. Microsoft Learn lists “Create Effective Prompts for Generative AI Training Tools” as a beginner module with seven units for audiences including K-12 educators, business users, higher education educators, and school leaders. Microsoft says the module covers the basic concepts of prompt engineering, the elements of an effective prompt, and prompting best practices. Source: Microsoft Learn, “Create Effective Prompts for Generative AI Training Tools.”
  6. The OpenAI Academy “ChatGPT at Work” collection is categorized under Workplace & Business, ChatGPT, Advanced & Builder Skills, and Work. The page lists videos including “Introduction to Prompt Engineering” (5:52), “Advanced Prompt Engineering” (8:50), “ChatGPT for Data Analysis” (4:46), “ChatGPT Search” (5:43), and “Deep Research” (7:06). Source: OpenAI Academy, “ChatGPT at Work.”
  7. All graphics in this draft are original graphics created for this article; no third-party images were used. Source: Original graphics created for Collaboration Evangelist draft, June 5, 2026.

 

Bibliography / Links

Accessed June 5, 2026. All graphics in this draft are original graphics created for this article; no third-party images were used.

  1. IBM Think. “What Is Prompt Engineering?” https://www.ibm.com/think/topics/prompt-engineering
  2. IBM SkillsBuild. “Artificial Intelligence.” https://skillsbuild.org/adult-learners/explore-learning/artificial-intelligence
  3. IBM SkillsBuild. “Prompt Engineering: Shaping Better AI Responses.” https://skillsbuild.org/college-students/course-catalog/prompt-engineering-shaping-better-ai-responses
  4. Microsoft Learn. “Create Effective Prompts for Generative AI Training Tools.” https://learn.microsoft.com/en-us/training/modules/create-prompts-for-generative-ai-training-tools/
  5. OpenAI Academy. “ChatGPT at Work.” https://academy.openai.com/public/collections/chatgpt-at-work-2025-02-14
  6. OpenAI Academy. “Introduction to Prompt Engineering.” https://academy.openai.com/public/videos/introduction-to-prompt-engineering-2025-02-13
  7. OpenAI Academy. “Advanced Prompt Engineering.” https://academy.openai.com/public/videos/advanced-prompt-engineering-2025-02-13

PMC 2025: The Spirit of Massachusetts, the Best of America, & Collaboration on Steroids!

, , ,

PMC 2025: The Spirit of Massachusetts, the Best of America, & Collaboration on Steroids!

By Craig Hall Underwood, August 22, 2025

On Sunday morning, August 3rd, at 3:45 a.m., I stood in the pre-dawn glow watching hundreds of volunteers serve breakfast to thousands of riders before the final leg of the Pan-Mass Challenge (PMC). I had an epiphany. This scene — and the entire weekend’s event, where I completed my 11th PMC riding 170 miles from Wellesley to Bourne on Saturday and then on to Provincetown — revealed something powerful: The PMC embodies the Spirit of Massachusetts and the very Best of America, at a time when we desperately need both.

Read more

CHU RECOMMENDS: 4 Books, 1 Novel, & 3 Podcasts to Learn About & Keep Up with AI

, , , , , , , ,

CHU RECOMMENDS: 4 Books, 1 Novel, & 3 Podcasts to Learn About & Keep Up with AI

Net/ Summary:

The Books:

  • The AI Driven Leader by Geoff Woods
  • AI First by Adam Brotman and Andy Sack
  • The Experimentation Machine by Jeffrey Bussgang
  • Human + Machine by Paul R. Daugherty & H. James Wilson

The Novel:

  • Coded Justice, by Stacey Abrams

The Podcasts:

  • The Most Interesting Thing in AI
  • Beyond the Prompt – How To Use AI In Your Company
  • Practical AI

More Details:

The Books:

The AI Driven Leader by Geoff Woods
This book blends strategic frameworks with real-world applications to help leaders understand and implement AI initiatives. Woods distills complex concepts into approachable tools for decision-makers.
Best for those wanting to translate AI potential into actionable leadership strategies.

AI First by Adam Brotman and Andy Sack
A practical playbook for leaders in any industry, AI First lays out the mindset and operating model needed to integrate AI into every level of an organization. The authors focus on customer experience, organizational design, and cultural transformation.
Best for those wanting to shift their business mindset to lead with AI from the ground up.

The Experimentation Machine by Jeffrey Bussgang
Bussgang, a venture capitalist and Harvard professor, explores how top startups use experimentation and data—especially with AI—to drive innovation. The book includes case studies from the tech world and frameworks for iterative learning.
Best for those wanting to understand how AI powers experimentation, especially in fast-moving startups and product teams.

Human + Machine by Paul R. Daugherty & H. James Wilson
This foundational book from two Accenture executives lays out how companies can combine human ingenuity with AI’s capabilities to reinvent business processes. It introduces the concept of “missing middle” jobs and augmented intelligence.
Best for those wanting to explore how AI augments—not replaces—human work across industries.

 

 The Novel:

Coded Justice by Stacey Abrams
Abrams weaves a fast-paced legal thriller around a deepfake video and a high-stakes courtroom battle. Beneath the drama lies a thought-provoking look at AI, surveillance, and ethics.
Best for those wanting to grasp the real-world potential in health care and the risks of AI through a gripping, fictional lens.

The Podcasts:

The Most Interesting Thing in AI
Hosted by Nathan Benaich, this podcast offers sharp insights into the latest developments in AI, from research to policy. With guests from DeepMind, OpenAI, and VC firms, it’s both accessible and rigorous.
Best for those wanting to stay current on breakthroughs and big-picture trends in the AI ecosystem.

Beyond the Prompt – How To Use AI In Your Company
This podcast focuses on real-world implementations of AI tools across functions like sales, marketing, operations, and customer service. It’s geared toward business leaders and internal innovators.
Best for those wanting to learn practical, tactical ways to embed AI in daily business workflows.

Practical AI
Produced by Changelog, this long-running podcast brings together developers, data scientists, and entrepreneurs to discuss AI applications, tools, and ethics. Each episode balances technical depth with big-picture thinking.
Best for those wanting to understand how AI actually works—and how to build with it.

Please let me know your recommendations! 

PS/ Disclaimer: One of the major debates about AI is “Does it/ can it/ or will AI make people lazy?”  FYI, I have actually read all 5 books (and others I have not recommended) and listen to these podcasts every week, but after outlining the article, I asked ChatGPT to write the short summaries and “best for” recommendations for each bullet from the summary.  I did not feel the need to edit what she wrote. Am I lazy or productive????

GOOGLE’S 9 HOUR AI PROMPT ENGINEERING COURSE – IN 1 CARTOON!

, , , , , , ,

Summary:

I recently completed Google’s Coursera course on prompt engineering to continue sharpening my use of AI—for both work and play. Like many of you, I want to get better, faster, and more reliable outputs from these powerful tools. To help me remember the core principles, I’ve adopted and modified a fun and memorable mnemonic (more on that in a second). But first, here’s the big idea:

To write effective prompts, you need to include seven core sections:

Teeny Tiny Crabs Riding Fat Enormous Iguanas
T = Task  T = Tone  C = Context  R = References  F = Format  E = Engage  I = Iterate

Include each of these elements in your prompt, and you’ll dramatically improve your results.

Where This Came From

Much of this is based on Tina Huang’s excellent YouTube summary, “Google’s 9-Hour Prompt Engineering Course in 20 Minutes.” (As someone once said: “If you steal from one person, it’s plagiarism. If you steal from many, it’s research.” I do a lot of research!)

The Google course itself is built around five inputs for effective prompt design:

Task, Context, References, Engage, Iterate

Their trainers introduced the mnemonic:

Thoughtfully Create Really Excellent Inputs

Tina found it hard to remember—and I agreed. She came up with a more vivid version:

Tiny Crabs Riding Really Enormous Iguanas

To that, I added two key concepts from my own experience—Tone and Format—to make it more complete.

The final version?

Teeny Tiny Crabs Riding Fat Enormous Iguanas

Why Add “Tone” and “Format”?

I’ve written hundreds of prompts across multiple platforms while using AI in professional consulting and nonprofit fundraising. Here’s why those two extra sections matter:

  • Tone is critical for anything involving writing. For example, when I want a professional yet conversational article (like this one), I say so in the prompt. Tone is often influenced by the Context section, especially when I open with something like:

“I am the head of a boutique consulting firm that specializes in…”
This helps the AI match voice, audience, and style.

  • Format becomes essential for large-scale research tasks. I’m currently using AI to research the charitable giving practices of 200+ companies to prioritize potential donors for several nonprofit clients. In this kind of work, it’s not enough to just get raw information—I need it delivered in tables, bulleted lists, or other clean formats that are easy to review and analyze.
    Gemini (Google’s AI) is particularly strong at formatting outputs into tables, and with one click, those can be exported into Google Sheets—saving tons of time.

A Quick Walkthrough of Each Section

  • #Task#: Be as specific as possible. Clearly define what you want the AI to do. I often assign the AI a role or persona, like:

“You are an expert content strategist writing a blog post on how to use AI in marketing research.”
This helps focus the AI’s responses.

  • #Tone#: Set the mood—professional, playful, technical, academic, or conversational. If your writing has a target voice, state it upfront.
  • #Context#: Describe who you are, why you’re asking the question, and who the audience is. The better the AI understands your intent, the better its output.
  • #References#: Upload or include everything relevant: job descriptions, resumes, bios, organizational goals, website URLs—whatever provides grounding for the AI’s output. The more context it has, the smarter it behaves.
  • #Format#: Be clear about how you want the information delivered. Want a comparison table? A checklist? A three-paragraph summary? Tell it.
  • #Engage#: Don’t copy and paste the output blindly. Engage with it. Review it. Challenge it. AI tools still make plenty of hallucinations (mistakes), so check citations, dates, and factual claims before you use them. Ask for sources. Fact-check relentlessly.
  • #Iterate#: Rarely do you get the perfect response on the first try. In my experience, it takes 2–3 iterations to get what I really want.

Here’s a fun example:
I used ChatGPT to create a cartoon for a party invitation featuring my American Bully, Izzy. I uploaded a photo and asked for a cartoon of her in a chef’s hat next to a grill.

  • Version 1: Looked like Izzy, but she was wearing her leather biker jacket from the photo.
  • Version 2: I asked for a chef’s jacket—got that—but the dog didn’t look like Izzy anymore.
  • Version 3: I asked to combine the head from version 1 and the jacket from version 2—and nailed it!

Iterating means providing clear, specific feedback. When the AI gets something wrong, tell it what and why. That’s how you teach it—and improve your results over time.

One Final Tip: Use Section Headers in Your Prompts

I have read several books and articles about AI and cannot remember where I learned this, but some recommend separating your prompts into the TTCRFEI sections and placing a “#” before and after the section title, e.g. #Task#.  Experts suggest this will improve the quality of your outputs by adding clarity and structure, focus, and reducing ambiguity.  By segmenting the prompt, you can make sure the AI knows what it is reading and how to use the information. For example, after reading # Task #, the AI knows the following text defines what it needs to do. After # Reference #, it understands that the text is supplementary information that it should consider. It also avoids or at least reduces misinterpretation.

Let’s Keep Learning Together

I hope this is helpful. If you’ve developed your own strategies for writing better prompts—or if you’ve tested this framework—I’d love to hear your thoughts. The more we share, the better we all get.

And Yes, I still recommend investing in the Coursera Google course.

One of my mantras as CEO of The Loyalty Group was “training without testing is little more than orientation.”  My (and Tina’s) shortened version/cartoons notwithstanding, if you are relatively new to AI, I would recommend taking the Coursera course.  It’s free, and if you have “fast ears” and can listen to the video portions at 1.5–2.0X speed, it takes far less than 9 hours.  The two things I like best about it are:

  1. Completing the course requires you to show actual prompts you have written, along with improvements you have made.
  2. The course has an exam at the end of each of the four major sections. If you score at least 80% on the exams, you receive this certificate:

CHU Recommends – The Experimentation Machine by Jeff Bussgang

, , , , , , ,

CHU Recommends: The Experimentation Machine, by Jeff Bussgang

 

I am fortunate to have known Jeff Bussgang for over 25 years and can honestly say that he is the perfect person to have written this book (and perfect is not a word I use often!).

Jeff studied AI in the late 80’s as a computer science student at Harvard (And before you call BS on that statement, realize that the term Artificial Intelligence was coined by in 1955 by computer pioneer John McCarthy!); was a very early stage “joiner” at Open Markets, an internet commerce infrastructure leader in the early Web 1.0 days; the Co-Founder UPromise; and currently serves as a Co-Founder and Partner at Flybridge Venture Capital and a Senior Lecturer/ Professor at Harvard Business School, where he formerly was an (maybe the first?) Entrepreneur in Residence.  Flybridge is currently investing a fund 100% dedicated to AI start-ups and HBS now requires all students to use AI in most, if not all, classes.   Jeff is also one of the nicest (not a word often associated with VC’s),charitable and by far the most connected person I know.  Despite how incredibly busy he is, Jeff somehow always seems available to take my call!

Leaders far more qualified than I have written glowing reviews of The Experimentation Machine, including:

Eric Ries, author of The Lean Startup: “A Perfect Guide for AI-era founders”

and Reid Hoffman, Co-Founder of LinkedIn and Inflection AI “A valuable, empowering read for entrepreneurs everywhere.”

I would take Hoffman’s praise a step further and recommend The Experimentation Machine as an important read, not only for AI Founders, but also for AI startup joiners/ followers and the general public, as it has become clear to me that everyone will either be affected by AI or benefit from using it.  In addition to learning a lot about AI and how “10X” founders are using AI to create next-generation businesses, you will learn about building a business plan, finding a profitable product market fit, and valuing early-stage enterprises.

If you want to immediately add some “AI Intellectual Capital,”  I recommend first reading Bussgang’s Appendix A “Prompt Tips and Examples,”  which contains a mini-course in prompt engineering, and D “AI Apps Unwrapped,”  which contains a list of dozens of AI tools by usage category (e.g. “General Assistance,” “Get Creative,” and “Have Fun.”).

I have recommended this book to hundreds of friends, followers, clients, and colleagues and am confident that you will benefit from reading (or listening to the audiobook in Jeff’s own – not his AI – voice).  My only suggestion for Jeff is to create a companion PDF for the audiobook, which could easily be done by using the PowerPoint slides he shared at the recent HBS book launch reception for The Experimentation Machine.

CHU

OVERWHELMED BY AI UBIQUITY? YOU’RE NOT TOO LATE TO GET STARTED – HERE’S HOW.

, , , , , , , ,

Net/ Summary:

If it feels like AI came out of nowhere, is suddenly everywhere, and you somehow missed the bus—you’re right about the first two, and dead wrong about the third. While AI adoption has skyrocketed—faster than any technology in modern history—actual usage still varies widely by industry, age, and education. I’ve recently taken a deep dive into AI’s capabilities, and I’m here to tell you: it’s not too late to catch up. You can start using these powerful tools today.

The best way to learn AI is to use it. Tools like ChatGPT, Microsoft Copilot, and my favorite, Google Gemini, offer free versions and premium options (around $20/month) with expanded capabilities. The key is not to treat them like Google search boxes, but like smart collaborators.

Give AI context. Tell it who you are (e.g., “a nonprofit strategist”), how you’ll use the output (e.g., “to draft a fundraising plan”), and what tone you prefer (e.g., “professional but conversational”). Upload background documents and relevant data. Ask it to help you brainstorm, reword emails, rewrite resumes, analyze data, or even generate cartoons from photos.

Don’t know where to begin? Try:

  • Editing a recent cover letter and resume (include the job description in your uploads)
  • Rewriting an important email in a friendlier tone
  • Summarizing a report
  • Asking how AI could help with a work or personal challenge

I highly recommend reading the books The AI Driven Leader, by Geoff Woods and The Experimentation Machine, by my friend Jeff Bussgang.  It you want to jump start your learning from these books, consider reading Wood’s last Chapter 14 “”Go from 0 to 1: The Simple Path to Deliver Value with AI” and the Appendix “Prompt Library”   and Bussgang’s Appendix A “Prompt Tips and Examples” and D “AI Apps Unwrapped”  which contains a list of dozens of AI tools by usage category (e.g. “General Assistance,” “Get Creative,” and “Have Fun.” ) first.  There are also dozens of online courses on how to effectively and efficiently use AI, including Google’s AI Prompt Engineering Course on Coursera or – if you are as impatient as I – first watch the YouTube video “Google’s 9 Hour AI Prompt engineering Course in 20 Minutes.”

Read more

LEADING BY LISTENING – 8 QUESTIONS FOR NEW TEAM MEMBERS  THAT HAVE NOTHING TO DO WITH “THE WORK”

, , , , , ,

Net/ Summary:

When on-boarding new team members or meeting new peers during my second tenure at Year Up leading national accounts like GE and Nielsen, I always scheduled two sessions with my colleagues. The first meeting had nothing to do with strategies, performance metrics, or deliverables. It was about connection.  These are the 8 questions I asked during the initial session:

  1. Tell me about your family and where you grew up.
  2. Tell me about your family today.
  3. What are your other non-work priorities?
  4. Tell me about your work style. What hours do you like to work? When are you most productive?
  5. Tell me about your communications style. How do you like to interact? Email, text, PowerPoint, IM, calls, etc.?
  6. What do you want to get better at? If we had an unlimited training budget (which we don’t) what training would you like us to invest in you?
  7. Do you have any pet peeves?
  8. What is your “Why?”

Read more

Waffle House Logic – My Thoughts on the 2024 Election and Actions to Fight Back Now

,

Waffle House Logic – My Thoughts on the 2024 Election and Actions to Fight Back Now

Net: I spent the last 5 days of the 2024 presidential campaign doing GOTV work in Allentown, PA for the Harris/ Walz campaign.  I believe Trump won because – as the manager of the Allentown Waffle House explained to me at 6 AM on my first morning in town,

“Trump gave us checks.  When he was president, there was more money in the economy and prices were lower.  And if he doesn’t win, we will be fighting World War III and they will draft not only my sons, but my daughter as well.  I’ll go to prison before I let that happen.”

While I could have countered her points with a few sentences each, I didn’t have the 6-word “bumper sticker/ yard sign” to combat her logic.  Although I have had this article outlined in my head since the disastrous evening of the election, I couldn’t break through my anxiety paralysis until now to commit fingers to keyboard.  One reason for this was I felt the need to also publish a “5 things you can do right now to fight back” article along with the Waffle House Logic, but have struggled to come up with the right action steps.  Fortunately, I received this email from Indivisible.org yesterday and am unabashedly plagiarizing their recommended actions for this week:

  1. Join an April 5 “Hands Off!” protest near you. 
  2. Register for the “Hands Off!” mass call on Tuesday, April 1 at 8pm ET, 5pm PT to prepare for this weekend.
  3. Call Your representative: Help block Republicans from disenfranchising millions of voters with their “Silencing Americans Act.”
  4. Hold the Trump administration accountable for its jaw-dropping national security meltdown over Signal.
  5. I’ll add a fifth action.  If you haven’t done so already, sign up to receive INDIVISIBLE’S email action alerts.

Read more

DEMOCRATS (USED TO) SUCK AT MARKETING. 3 IDEAS THAT DON’T.

, , , , , ,

DEMOCRATS (USED TO) SUCK AT MARKETING. 3 IDEAS THAT DON’T.

 Net/ Summary:

 I absolutely LOVE the Harris Walz positioning “Freedom, Opportunity, and The Promise of  America.” It’s tight, brilliant, positive, and even better, in my opinion, than Jon OssofF’s “Jobs, Justice, Healthcare.” The team’s consistent use of the 3-point slogan also relieves years of personal frustration at Democrats’ inability to not suck at marketing.  In 2016, I wrote about this regarding the weak selling of the Affordable Care Act in my article What Obama Can Learn About Marketing from Ross Perot, Cecil Underwood, and Coalition Loyalty Program Marketing.

A few ideas for the Harris-Walz team:

  1. Every time Trump loses his mind in a social media post, press conference, speech, etc., refer to this as “WEIRD RIDICULOUS RANTS.” That could stick.
  2. Positioning the success of the Biden-Harris administration. Whenever Trump goes off on one of this “Biden is the worst president in history” weird, ridiculous rants (see it works), come back with:

“The Biden-Harris administration created more jobs, more new business start-ups, covered more Americans with health insurance, passed more bi-partisan legislation and funded more infrastructure projects than any other first term President in history. Period.”

  1. Create a “NOT GOING BACK FACTS” (NotGoingBackFacts.com) website. mobile site/ app that lets voters (and supporters/ field workers/ down ballot candidates) easily access the facts, charts and sources comparing the impact of both the previous 4 years under Trump vs Biden-Harris and expert forecasts for the next 4 years under Harris-Walz vs Trump-Vance based on his stated policies/ actions/ goals for the major issues important to voters.

Examples could include:

  • Jobs created.
  • People covered by health insurance.
  • Bridges, roads and other infrastructure projects funded.
  • Seniors whose monthly costs for insulin is now $35 vs $400 under Trump.
  • Student loan debts forgiven (both the number of people affected, and the dollar amount of loans forgiven.)
  • Bi-Partisan Bills passed.
  • Fiscal responsibility/ National Debt management.
  • Crime control/ reduction.

The ease and beauty of this is that (a) the data already exists in numerous formats and (b) the website/ mobile app could easily allow the user to look at the impact of the Biden-Harris record and Harris-Walz policies (or pain of Trump’s) on a local basis.

Read more