I Made a Prediction in January. Now I Can Show You Exactly How It Works.
Published: February 4, 2026 - 12 min read
Eleven days ago, I wrote something that felt almost too bold to publish.
In that January post, I predicted that AI would become a human-to-human bridge. I said that experts would package their knowledge into MCPs and sell access to their wisdom 24/7. I said that influence would be measured not by followers, but by how many people create agents inspired by you.
I meant every word. But I also knew something was missing.
I could see the vision. I could feel that it was coming. But I had not yet connected the dots between what I knew about MCPs and how that vision would actually work.
Now, after writing Part 1 and Part 2 of this series, I can finally show you the mechanism.
The three primitives are the answer.
Tools. Resources. Prompts. These are not just technical concepts. They are the building blocks of the human-to-human bridge I predicted.
Let me show you exactly what I mean.
The Prediction, Revisited
In January, I wrote:
"The people who already have status, the thought leaders, domain experts, and successful entrepreneurs, will learn to package their knowledge and insights into MCPs. They will sell these as a way for people to have access to them 24/7."
I gave examples. Lots of them. A fitness coach whose methodology you admire. A financial expert whose podcast resonates with you. A parenting expert whose calm approach you wish you could access at 8 PM during a meltdown. An enterprise consultant whose strategic thinking costs $50,000 for two hours.
I asked: What if you could access their full body of knowledge through AI?
At the time, I was describing a vision. Now I can describe a mechanism.
The Mechanism: Three Primitives, Infinite Possibilities
Here is what I now understand with complete clarity:
Every expert's knowledge can be decomposed into three parts:
- What they know (their methodology, frameworks, accumulated wisdom)
- What they do (the actions they take, the deliverables they produce)
- How they guide you (the workflows they use to solve your specific problem)
And here is the revelation:
| Expert's Knowledge | MCP Primitive | What It Becomes |
|---|---|---|
| What they know | Resources | Curated knowledge bases, research, frameworks |
| What they do | Tools | Functions that produce real outputs |
| How they guide you | Prompts | Workflow templates for specific situations |
That is the bridge.
When an expert packages their knowledge into an MCP, they are not just giving you information. They are giving you their methodology, their capabilities, and their guidance. All three. Working together.
And suddenly, the AI is not a wall between you and the expert. It is a bridge.
Example 1: The Fitness Coach MCP
Let me walk through one of my January examples in detail.
The scenario: It is 6 AM. You woke up motivated, but now you are staring at your home gym and you have no idea what to do. You have a bad knee, 30 minutes before work, and limited equipment.
The expert: A fitness coach whose philosophy has always resonated with you.
How the MCP would work:
| Primitive | What the Coach Packages | What You Experience |
|---|---|---|
| Resources | Their training philosophy, exercise progressions, injury modification database, periodization principles | The AI knows their methodology |
| Tools | generate_workout, log_session, track_progress, adjust_for_injury | The AI creates a workout for you |
| Prompts | "Quick morning workout", "Injury-safe session", "Progressive overload check" | You just select what you need |
The interaction:
You select the "Quick morning workout" prompt. You tell the AI about your bad knee, your 30 minutes, and your equipment.
The AI accesses the coach's injury modification database (Resource). It accesses their periodization principles (Resource). It uses the generate_workout tool to create a session tailored to your constraints. It uses the adjust_for_injury tool to swap out anything that would aggravate your knee.
Result: A workout designed exactly for today's reality. Not generic advice. Not a blog post written for millions. A session designed for one: you.
That is the human-to-human bridge in action.
Example 2: The Financial Mentor MCP
The scenario: You just got a promotion with a significant salary increase. Index funds or individual stocks? Max out your 401k or pay down your mortgage faster?
The expert: A financial advisor whose podcast you listen to every week and whose philosophy deeply resonates with you.
How the MCP would work:
| Primitive | What the Advisor Packages | What You Experience |
|---|---|---|
| Resources | Their investment philosophy, tax optimization strategies, risk assessment frameworks, wealth-building principles | The AI thinks like them |
| Tools | analyze_portfolio, calculate_scenarios, generate_plan, compare_options | The AI runs the numbers for you |
| Prompts | "New income allocation", "Debt vs investment", "Retirement projection" | You describe your situation |
The interaction:
You select "New income allocation." You provide your salary, your debt, your existing investments, your risk tolerance.
The AI accesses their tax optimization strategies (Resource). It uses calculate_scenarios to model different approaches (Tool). It applies their wealth-building principles (Resource) to recommend an allocation.
Result: Financial guidance through their lens. Not generic advice. Their actual investment philosophy and decision-making framework applied to your unique financial situation.
Example 3: The Enterprise Consultant MCP
The scenario: Your company is facing a major strategic decision. Merger or organic growth? Pivot or double down?
The expert: A top-tier strategist who has guided Fortune 500 companies through similar crossroads. Normally $50,000 for a two-hour consultation.
How the MCP would work:
| Primitive | What the Strategist Packages | What You Experience |
|---|---|---|
| Resources | Their decision-making frameworks, industry analysis methodologies, case study patterns, risk assessment criteria | The AI reasons like them |
| Tools | analyze_market, model_scenarios, assess_risks, generate_recommendation | The AI does the analysis |
| Prompts | "Strategic pivot assessment", "M&A evaluation", "Market entry analysis" | Your leadership team asks questions |
The interaction:
Your CEO selects "Strategic pivot assessment" at 2 AM when they cannot sleep. They describe the situation, the options, the constraints.
The AI accesses the strategist's decision-making frameworks (Resource). It uses model_scenarios to project outcomes (Tool). It applies their risk assessment criteria (Resource) to flag concerns. It uses generate_recommendation to produce a structured analysis (Tool).
Result: World-class strategic thinking, available whenever you need it. Not a canned answer. A genuine engagement with your specific situation through the lens of someone who has seen hundreds of similar decisions.
Example 4: The Parenting Expert MCP
The scenario: It is 8 PM and your child is having a meltdown. You are exhausted. You need guidance right now.
The expert: A parenting coach whose calm, empathetic approach has always resonated with you.
How the MCP would work:
| Primitive | What the Coach Packages | What You Experience |
|---|---|---|
| Resources | Emotional regulation frameworks, age-appropriate scripts, de-escalation techniques, developmental context | The AI understands children like they do |
| Tools | generate_script, suggest_response, analyze_pattern, create_plan | The AI gives you words to say |
| Prompts | "Meltdown response", "Difficult conversation", "Bedtime resistance" | You get help in the moment |
The interaction:
You select "Meltdown response." You describe your child's age, the trigger, what you have already tried.
The AI accesses their emotional regulation frameworks (Resource). It uses generate_script to give you actual words (Tool). It applies their developmental context (Resource) to explain what is happening in your child's brain.
Result: Their wisdom, available in the moment you need it most. Not a parenting book you have to search through. Not a therapist appointment you have to wait for. Help right now.
The Complete Picture
Let me give you the full mapping for every example from my January prediction:
| Expert Type | Resources (What They Know) | Tools (What They Do) | Prompts (How They Guide) |
|---|---|---|---|
| Fitness Coach | Training philosophy, injury modifications, periodization | Generate workout, log session, track progress | Quick workout, injury-safe session |
| Financial Mentor | Investment philosophy, tax strategies, risk frameworks | Analyze portfolio, calculate scenarios, generate plan | New income allocation, debt vs investment |
| Enterprise Consultant | Decision frameworks, industry analysis, case patterns | Model scenarios, assess risks, generate recommendation | Pivot assessment, M&A evaluation |
| Parenting Expert | Emotional regulation, developmental context, scripts | Generate script, analyze pattern, create plan | Meltdown response, difficult conversation |
| Dating Coach | Relationship frameworks, communication principles, boundary setting | Analyze message, suggest response, evaluate compatibility | Text response help, first date prep |
| Domain Expert | Industry knowledge, compliance requirements, unwritten rules | Validate approach, check compliance, identify risks | New project review, compliance check |
| 30-Year Veteran | Institutional knowledge, client relationships, workarounds | Document process, find precedent, recommend approach | New hire onboarding, client history lookup |
Every single one follows the same pattern.
Resources provide the expert's accumulated wisdom. Tools let the AI take action on your behalf. Prompts guide you through specific situations without requiring you to understand the technical details.
That is how the human-to-human bridge gets built.
Why This Is Not Disconnection
Remember the fear I addressed in my January post?
"There is a concern I hear constantly: AI will disconnect us from each other. We will all be talking to machines, losing the human element that makes connection meaningful."
Now I can answer that concern with more precision.
When you use an expert's MCP, you are not talking to a generic AI. You are talking to their methodology, their frameworks, their accumulated wisdom. The AI is the medium, but the mind behind it is human.
Think about it this way:
- A book lets you access someone's thinking, but it cannot respond to your specific situation.
- A podcast lets you hear someone's voice, but it cannot answer your questions.
- A course lets you learn their system, but it cannot adapt to your constraints.
An MCP gives you all three. Their thinking. Their voice. Their system. And it adapts to you.
The AI is not replacing human connection. It is scaling human connection. It is letting one expert's wisdom reach thousands of people in personalized ways that were never possible before.
That is the bridge.
The Industry Is Already Moving
In Part 1, I mentioned the momentum behind MCPs:
- December 2025: Anthropic donated MCP to the Agentic AI Foundation under the Linux Foundation
- March 2025: OpenAI officially adopted MCP for ChatGPT
- Supporting members: Google, Microsoft, Amazon, Cloudflare, Bloomberg
This is not a niche technology anymore. MCPs are becoming infrastructure.
And as the infrastructure matures, the opportunity for experts to package their knowledge grows. The fitness coaches, financial advisors, parenting experts, and enterprise consultants I described are not hypothetical. They are inevitable.
The bridge is being built. The question is: who will cross it first?
What This Means for You
If you are an expert in something, anything, start thinking about what an MCP of your knowledge would look like.
- What Resources would you include? What is the accumulated wisdom you wish you could share with everyone?
- What Tools would you build? What actions do you take that could be automated or assisted?
- What Prompts would you offer? What are the common situations where people need your guidance?
You do not need to build it today. But start thinking about it.
Because the future I predicted in January is not just a vision anymore. It has a mechanism. And that mechanism has three parts.
Tools. Resources. Prompts.
The building blocks of the human-to-human bridge.
What Is Coming Next
This is Part 3 of a 4-part series called MCPs for Humans.
In Part 4, I am going to take you behind the scenes of the Alex Bennett MCP I built. A complete case study of how I designed all 16 tools, 6 resources, and 10 prompts. The full architecture. The decisions I made. The lessons I learned.
I have been showing you the theory. Now I will show you the practice.
Series Navigation
- Part 1: What MCPs are and why they matter
- Part 2: The three superpowers (Tools, Resources, Prompts)
- Part 3 (this post): How MCPs make my Human-to-Human Bridge prediction possible
- Part 4: Behind the scenes of Alex Bennett MCP (deep case study)
As always, thanks for reading!