The ONHT Framework: GenAI Prompting Solutions That Actually Work for People
By John Garner on Sunday, June 15, 2025
Summary: GenAI tools are transforming work, but most people get poor results because they don't understand how to communicate with AI built on structured data. This guide is a series of articles that teaches the ONHT framework—a systematic approach to prompting that transforms vague requests into exceptional outputs by focusing on Objectives (what problem), Needs (what information), How (thinking approach), and Trajectory (path to solution). Master this framework and develop an expert mindset grounded in human-in-the-loop thinking, critical analysis, and empathy, and you'll excel with any AI tool, at any company, in any role.
GenAI Prompting: The Essential Skill for Your Future
Why This Matters More Than You Think
A shift is happening. Right now.
Companies are adopting GenAI tools faster than they adopted email. But here's what most people miss: the tool isn't the advantage. Knowing how to use it is.
Think about this. Two people have access to the same AI. One gets generic, useless outputs. The other gets insights that transform their work. Same tool. Different results.
The difference? Understanding how to communicate with AI.
The New Literacy
GenAI tools work differently from traditional software. They don't follow rigid commands. They understand natural language, drawing from massive knowledge bases of structured, labelled data to help you think, create, and solve problems.
But "natural" doesn't mean "random". These tools respond to patterns, context, and structure - because that's how they learned.
Master these patterns, and you become invaluable. Not just at your current job. At any job.
Here's why: The tool might change. Your company might switch from ChatGPT to Claude to something not yet invented. But the principles stay the same. The company might change. You might move jobs. Your new employer uses different tools. No problem. You know how to adapt. The work will definitely change. Tasks that took hours now take minutes. But only if you know how to ask.
The Gap That's Opening
Watch any workplace right now. You'll see three or four groups:
The Resistant: "AI will never replace human creativity"
The Dabbler: "I tried it. The results were mediocre"
The Skilled: Getting 10x more done, with better quality
The Expert: Champions ways of working like ONHT sharing knowledge with others
The gap between groups 2 and 3 is growing daily.
Group 2 asks: "Write me a report on customer satisfaction" Group 3 asks: "I need three actionable insights from our Q3 customer feedback that I can present to non-technical board members in 5 minutes. Focus on what we can fix with our current budget of £50k."
Same tool. Vastly different value.
What Makes the Difference
It may surprise you, but it's not about being "technical". It's about understanding three things:
1. How GenAI Actually Works
GenAI and similar tools aren't search engines. They're reasoning partners built on foundations most people don't understand.
Here's the reality: LLMs learn from massive amounts of structured, labelled, and categorised data. Meta didn't spend $14 billion with ScaleAI just for raw text. They paid for carefully organised information that teaches AI how concepts connect, how ideas relate, and how human knowledge fits together.
This matters because:
Here's the reality: LLMs learn from massive amounts of structured, labelled, and categorised data. Meta didn't spend $14 billion with ScaleAI just for raw text. They paid for carefully organised information that teaches AI how concepts connect, how ideas relate, and how human knowledge fits together.
This matters because:
AI outputs reflect the patterns in that structured data
Better prompts tap into those patterns more effectively
Understanding this structure helps you get better results
Think of it like a vast library where every book is cross-referenced with every other book. Ask vague questions, get random pages. Ask with structure, get precisely what you need.
They excel when you:
Provide context about your specific situation
Clarify who needs the information and why
Define what success looks like
They struggle when you:
Give vague instructions
Assume they know your constraints
Expect perfection without iteration
2. The Human-AI Partnership
You're not being replaced. You're being amplified.
But amplification requires collaboration:
You provide the wisdom and context
AI provides the processing power and knowledge
Together, you create something neither could alone
Bad approach: "Do my work for me" Good approach: "Help me think through this challenge"
3. The Skill That Transfers Everywhere
Learn to prompt well, and you've learned to:
Clarify your own thinking
Communicate complex needs simply
Break down problems systematically
Iterate towards excellence
These aren't just AI skills. They're structured thinking skills.
Your Competitive Advantage
Here's what most people don't realise: prompting skill compounds.
Every interaction teaches you something. What works. What doesn't. How to adjust.
But only if you're learning, not just copying.
Bad pattern: Collect "perfect prompts" from others Good pattern: Understand WHY prompts work, then create your own
Six months from now, while others still get mediocre results, you'll be operating at a different level. Not because you're smarter. Because you learned the patterns.
And when the tools change? When new capabilities emerge? You'll adapt instantly. Because you understand principles, not just procedures.
Starting Your Journey
This guide teaches you the ONHT framework. It's simple, powerful, and works with any GenAI tool.
But more importantly, it teaches you to think WITH AI, not through it.
O - Objective: What problem are we solving? (The human need, success criteria, audience context) N - Needs: What information matters? (Context, structured flow, layered complexity) H - How: How should the AI think? (Mental models, expertise emulation, reasoning style, constraints) T - Trajectory: What path gets us there? (Step-by-step progression with validation)
Master these four elements, and you'll never get a useless AI response again.
More crucially, you'll understand WHY each response works. You'll build real capability, not just a prompt collection.
What You'll Learn
I'll take you through three levels:
Beginner: Get consistently useful outputs instead of generic fluff Intermediate: Design prompts that solve complex, multi-faceted problems Advanced: Engineer AI interactions that amplify your expertise
But more importantly, you'll develop the expert mindset:
Keeping humans in the loop, not letting AI run blind
Thinking critically about every output
Maintaining empathy for end users and output in every interaction
Each level builds on the last. Start where you are. Progress at your pace. But always remember: techniques are just tools. The mindset is what makes you exceptional.
A Critical Warning: The Learning Trap
Here's what the hype won't tell you.
MIT researchers just discovered something troubling. Students using AI scored brilliantly on exams. But when AI was removed? Their performance plummeted below those who never used it at all.
Why? They'd learned to operate the tool. Not understand the subject.
This is the trap. Use AI to bypass thinking, and you bypass learning. Use it to enhance thinking, and you accelerate growth. Learn to apply the ONHT framework along with its underlying principles, and you'll develop something more valuable: an expert's mindset.
True expertise isn't just knowing techniques. It's a state of mind built on three foundations:
Human-in-the-Loop: You stay actively engaged, improving and evolving outputs rather than accepting first results Critical Thinking: You question outputs, check for bias, and guard against manipulation by bad actors or flawed data Empathy: You always consider the real humans who'll use what you create, their actual needs, and what truly helps them
These aren't just nice principles. They're your protection against AI's limitations and your path to exceptional results.
The difference matters: Bad approach: "Write my report on market analysis" Result: You get a report. You learn nothing. Tomorrow, you're no better.
Good approach: "Help me understand market analysis principles. Then guide me through applying them to my specific situation" Result: You get a report. AND the ability to do it yourself. Tomorrow, you're stronger.
Beyond the Hype: What Actually Works
Yes, AI is revolutionary. No, it won't solve everything.
The hype machine wants you to believe AI is magic. It's not. It's a powerful tool that amplifies whatever you bring to it.
Bring laziness? Get polished mediocrity. Bring curiosity? Get accelerated learning. Bring expertise? Get breakthrough insights.
Look at what's working in education. In Africa, AI platforms are transforming learning by personalising education, not replacing it. They guide students through concepts, adapting to how each person learns best.
That's the model. AI as a mentor, not a substitute.
The Time to Start Is Now
Every day you wait, the gap widens.
But not the gap you think. It's not between those who use AI and those who don't.
It's between those who use AI to learn and those who use it to avoid learning.
Your colleagues are learning this. Your competitors are implementing it. Your industry is transforming around it.
But you? You're about to understand it at a level most never will.
Let's begin with the fundamentals. They're simpler than you think. And more powerful than you imagine.
Critical: Skip the learning, and you'll ace the interview but fail the job. Embrace the learning, and you'll excel at both.
Remember: The goal isn't to become "good at AI". It's to become exceptional at achieving your goals, with AI as your amplifier. The tool serves you, not the other way round.
Next in this Series: - The ONHT Framework for Beginners - The ONHT Framework for Intermediate users - The ONHT Framework for Advanced users - Key Advanced Features of the ONHT Framework
GenAI tools are transforming work, but most people get poor results because they don't understand how to communicate with AI built on structured data. This guide is a series of articles that teaches the ONHT framework—a systematic approach to prompting that transforms vague requests into exceptional outputs by focusing on Objectives (what problem), Needs (what information), How (thinking approach), and Trajectory (path to solution). Master this framework and develop an expert mindset grounded in human-in-the-loop thinking, critical analysis, and empathy, and you'll excel with any AI tool, at any company, in any role.
I question the reliability and accuracy of Generative AI (GenAI) in enterprise scenarios, particularly when faced with adversarial questions, highlighting that current Large Language Models (LLMs) may be data-rich but lack in reasoning and causality. I would call for a more balanced approach to AI adoption in cases of assisting users, requiring supervision, and the need for better LLM models that can be trusted, learn, and reason.
I discuss my experience with chatbots, contrasting older rules-based systems with newer GenAI (General Artificial Intelligence) chatbots. We cannot dismiss the creative capabilities of GenAI-based chatbots, but these systems lack reliability, especially in customer-facing applications, and improvements in the way AI is structured could lead to a "software renaissance," potentially reducing society's technical debt.
The article discusses the contrasting debate on how AI safety is and should be managed, its impact on technical debt, and its societal implications.
It notes the Center for AI Safety's call for a worldwide focus on the risks of AI, and Meredith Whittaker's criticism that such warnings preserve the status quo, strengthening tech giants' dominance. The piece also highlights AI's potential to decrease societal and technical debt by making software production cheaper, simpler, and resulting in far more innovation. It provides examples of cost-effective open-source models that perform well and emphasizes the rapid pace of AI innovation. Last, the article emphasises the need for adaptive legislation to match the pace of AI innovation, empowering suitable government entities for oversight, defining appropriate scopes for legislation and regulation, addressing ethical issues and biases in AI, and promoting public engagement in AI regulatory decisions.
