Week 1 AI Showdown: Which Platform Wrote the Best Prompt Post?
Every week, Ketelsen.ai runs the same prompt topic through three of the biggest AI platforms on the planet — ChatGPT, Gemini, and Claude — and publishes all three results side by side. Same topic, same template, same rules. The only variable is the AI doing the thinking. This post is the part where we stop being polite and start comparing. We built a 7-dimension scoring rubric that measures everything from prompt engineering depth to writing voice to whether the reader actually walks away more capable than before. The rubric is transparent, the evidence is cited, and the scores are computed before the verdict — not the other way around. This is Week 1, and the topic was a big one: the five prompt mistakes that cause most of the bad AI output people complain about. All three platforms took their shot. One of them landed harder than the others.
The Topic: 5 Prompt Mistakes That Ruin AI Output
This week's topic asked each AI platform to produce a complete blog post featuring three prompt variations — Beginner, Intermediate, and Advanced — all targeting the same problem: the handful of avoidable mistakes that cause the majority of weak AI output. The topic matters because it sits at the exact intersection of what every AI user needs and what almost nobody teaches well. Each platform had to produce usable prompts, explain the engineering behind them, provide industry examples, and give the reader something they could use immediately.
How We Score: The 7-Dimension Quality Rubric
We do not pick winners by gut feeling. Every comparison post on Ketelsen.ai uses a structured rubric that scores each platform across seven dimensions, each weighted by how much it matters to you — the reader. The dimensions are not abstract quality labels; they measure whether the post actually does its job. Can you copy the prompt and use it today? Does the breakdown teach you something transferable? Are the industry examples specific enough that someone in that field would recognize their own problem? Those are the questions behind the numbers.
Each dimension is scored on a 1-to-5 scale with specific anchor descriptions at every level, so a "4" means the same thing whether we are scoring ChatGPT or Claude. The dimension scores are then weighted by importance and normalized to a 0-to-100 overall score. If two platforms land within 2 points of each other, we call it a tie and explain the trade-offs instead of forcing a winner. The rubric is version 1.0 — it will evolve as we discover new ways to measure what makes one AI post genuinely better than another.
| Dimension | Weight | What It Measures |
|---|---|---|
| D1: Prompt Quality | 20% | Are the prompts well-engineered, genuinely usable, and clearly differentiated across difficulty levels? |
| D2: Breakdown Clarity | 15% | Does the prompt breakdown teach you WHY each part matters, not just what it says? |
| D3: Industry Examples | 15% | Are the practical examples specific enough that a professional in that field would say "that is exactly my problem"? |
| D4: Writing Quality | 15% | Does the writing match the Ketelsen.ai voice — fun, informative, accessible, and publication-ready? |
| D5: Creative Use Cases | 10% | Do the suggested use cases go beyond the obvious — would the reader think "I never would have considered that"? |
| D6: Actionability | 15% | Can the reader immediately USE what they read — tools, checklists, workflows they walk away with? |
| D7: Completeness | 10% | Does the post cover all template sections fully without filler or thin spots? |
Platform-by-Platform Breakdown
Claude: 98.0 / 100
Strengths
Claude dominated four of the seven dimensions and tied for the lead in a fifth. The most striking advantage is in prompt engineering depth. The Advanced variation — titled "The Prompt Failure Analysis and Remediation System" — is not just a better prompt; it is a five-step methodology with isolation testing, severity ratings, and interaction effect analysis. No other platform attempted anything close to that level of structural sophistication. The Intermediate variation adds root-cause analysis and a "what changed" comparison for every fix, which turns the post from a teaching tool into a genuine diagnostic workflow.
The prompt breakdowns are where Claude separates most clearly from the field. Where other platforms describe what each prompt segment does, Claude explains what would go wrong without it — a subtle but meaningful difference that teaches transferable prompt engineering skills rather than one-time comprehension. The writing voice is the most consistently polished of the three, opening with "Every prompt you type into ChatGPT, Claude, or Gemini is either working for you or quietly working against you" and sustaining that energy across nearly 97,000 characters without filler.
Creative use cases are the widest-ranging of all three platforms: 21 total across the three variations, including tabletop gaming, therapy and coaching protocols, competitive intelligence frameworks, and governance policy design. The follow-up prompts are strategically sequenced — personal application first, then team standardization, then organizational systems — giving the reader a progression path, not just a next step.
Weaknesses
Citations are listed as "NOT APPLICABLE" across all three variations, which is honest but means the post provides zero external credibility anchors. The quality improvement percentages in Chart 1 (such as "vague objective: 25% quality before fix, 88% after") are presented without methodology — the reader cannot tell whether these are illustrative estimates or measured data. These are minor gaps in an otherwise comprehensive post, but they would matter to a skeptical reader.
Signature Move
Claude treats prompt improvement as a systems engineering problem rather than a tips-and-tricks exercise — and the result reads like a professional methodology, not a blog post.
ChatGPT: 80.0 / 100
Strengths
ChatGPT delivered the most consistent post of the three — every dimension scored a solid 4 out of 5, with no weak spots and no standout dimension pulling it down. The Advanced variation introduces a "validation pack" concept that frames prompt improvement as a quality assurance system, which is a genuinely novel idea that neither of the other platforms attempted. The industry coverage is the broadest at 12 distinct sectors across all three variations, including cybersecurity consulting, non-profit leadership, and higher education.
The writing voice lands squarely in the Ketelsen.ai target zone. The opening metaphor — comparing a vague prompt to "asking a contractor to fix the kitchen with no budget, no style preference, and no timeline" — is immediately relatable and sets the right tone. The creative use cases include community theater and hobbyist podcasting, which are pleasant surprises in a business-focused post. At 1,088 lines and 82,000 characters, the post is substantial without feeling bloated.
Weaknesses
The citations are identical across all three variations — the same five sources repeated verbatim three times rather than curated per difficulty level. The Prerequisites sections across all variations are thin enough to be skippable. The biggest gap relative to Claude is in prompt breakdown depth: ChatGPT's breakdowns explain what each segment does but stop short of teaching the reader how to apply those principles to their own prompts. It is the difference between showing someone a well-built bridge and teaching them structural engineering.
Signature Move
ChatGPT is the most reliable all-rounder — no dimension is weak, every section delivers, and the validation pack concept in the Advanced variation shows genuine creative ambition.
Gemini: 70.0 / 100
Strengths
Gemini opens with the most emotionally resonant hook of all three platforms: "you ask an AI for a brilliant strategy, and it hands you a robotic, generic wall of text that sounds like a 1990s textbook." That line does more work in one sentence than most intros do in a paragraph. The variation naming is clever and memorable — "The AI Whisperer Starter Prompt," "The 5-Point Prompt Auditor," and "The Chain-of-Thought Prompt Diagnostic Engine" each convey both personality and function. The Advanced variation's chain-of-thought approach is technically ambitious and educationally sound.
Industry examples include some distinctive picks that the other platforms missed: Identity and Graphic Design, Enterprise Logistics and Supply Chain, and Cybersecurity and IT Operations. The "Conflict Resolution" creative use case — helping someone write a text to a friend who owes them money — is the single most unexpectedly human moment across all three posts.
Weaknesses
At 52,493 characters and 600 lines, Gemini's post is roughly half the size of Claude's and notably shorter than ChatGPT's. That size difference shows up in the scoring: the Charts and Graphs section runs about 171 words (compared to thousands in the other two posts), the Current Use sections are very brief, the Prerequisites are minimal, and the Tags sections are thin. The prompt breakdowns are competent but more descriptive than instructive — they tell you what each part does without fully explaining why it matters to the AI's reasoning. The overall impression is of a post that does most things well but does not go deep enough in any single area to compete with the other two on substance.
Signature Move
Gemini has the best naming instincts and emotional hooks of the three — if personality and memorability were weighted higher, this would be a much closer race.
The Verdict
Claude wins Week 1 with a score of 98.0 out of 100, ahead of ChatGPT at 80.0 and Gemini at 70.0. The margin is decisive — 18 points over the runner-up — driven by Claude's advantages in prompt engineering sophistication, breakdown clarity, creative use case range, and the strategic sequencing of its follow-up prompts. Claude's Advanced variation does not just teach you about prompt mistakes; it gives you a production-grade failure analysis system you could hand to a team lead and say "run this every week." That is a different category of output than what the other two platforms produced.
That said, this is not a shutout. ChatGPT's validation pack concept is an innovation that neither Claude nor Gemini matched, and its consistent 4-out-of-5 performance across every dimension means there are no weak links in the chain. Gemini's emotional hooks and variation naming are the best of the three, and its "Conflict Resolution" use case is the kind of unexpectedly human moment that makes a reader stop scrolling. Each platform has something the others do not.
What This Means for You
If you want the deepest prompt engineering education from a single blog post, start with the Claude version — especially the Intermediate and Advanced variations, which teach transferable diagnostic skills you can apply to any future prompt. If you want the most balanced, no-surprises reading experience with a clever quality-assurance angle, go with ChatGPT. If you want the most emotionally engaging entry point and you are short on time, Gemini's Beginner variation is the fastest on-ramp of the three. All three posts are published on Ketelsen.ai — read the one that matches your experience level, or read all three and draw your own conclusions.
Score Summary
| Dimension | Weight | Claude | ChatGPT | Gemini |
|---|---|---|---|---|
| D1: Prompt Quality | 20% | 5 | 4 | 4 |
| D2: Breakdown Clarity | 15% | 5 | 4 | 3 |
| D3: Industry Examples | 15% | 5 | 4 | 4 |
| D4: Writing Quality | 15% | 5 | 4 | 4 |
| D5: Creative Use Cases | 10% | 5 | 4 | 3 |
| D6: Actionability | 15% | 5 | 4 | 3 |
| D7: Completeness | 10% | 4 | 4 | 3 |
| OVERALL SCORE (0-100) | 98.0 | 80.0 | 70.0 | |
Visual Comparison
Claude
ChatGPT
Gemini
The Prompts Behind the Posts
To make this a fair comparison, all three AI platforms received the exact same set of prompts in the exact same order. The prompts were generated by the Ketelsen.ai weekly content pipeline and delivered identically to ChatGPT, Gemini, and Claude — same context, same instructions, same constraints. The only variable was the AI doing the work. Image generation prompts were excluded from this comparison because they were not given to all three platforms in the same way. What follows are the four shared prompts that produced the blog posts you just read scored.
Prompt 1 of 4 — Session Setup (Phase 1)
Purpose: Establishes full site context, audience profile, brand voice, and weekly topic before any content generation begins. Every AI receives the same 10-part briefing so it understands who it is writing for and why.
"I need your help creating content for my blog, Ketelsen.ai. Let me give you the full context before we begin. --- PART 1 — PERSONAL BACKGROUND: I am Richard Ketelsen, based in Minneapolis, MN, USA. I have a professional background in Computer Science and Graphic Design. I currently serve as a Senior Cybersecurity Incident Responder at a Fortune 100 Company, with 6 years in cybersecurity, 10 years in Identity Design, and 10 years of entrepreneurship experience. --- PART 2 — SITE PURPOSE: Ketelsen.ai is an ongoing AI prompt crafting experiment. The blog section features an exclusive prompt collection of in-depth AI prompts covering real-world problems, generated weekly by multiple AI services (Claude, ChatGPT, Gemini). --- PART 3 — TARGET AUDIENCE: Demographics: Ages 25-45, global (English-speaking), professionals or entrepreneurs with moderate to high discretionary income. Psychographics: Enthusiastic about AI-driven innovation, enjoy experimenting with new technology, prefer transparent behind-the-scenes exploration. Persona: 'Alex the AI Trailblazer' — 33-year-old product manager at a startup who craves cutting-edge AI prompts and advanced experimentation. Pain Points: 1. Overwhelm by AI Options (decision fatigue) 2. Difficulty Crafting Effective Prompts 3. Time Constraints for Testing --- PART 4 — UNIQUE VALUE PROPOSITION: Ketelsen.ai delivers a constantly evolving library of advanced AI prompts — over 1,000 and counting — so tech-savvy entrepreneurs can rapidly experiment without wasting time on guesswork. --- PART 5 — COMPETITIVE EDGE: 1. Massive Prompt Library (over 1,000 and growing) 2. Real-Time Experimentation: Transparent, ongoing 3. Behind-the-Scenes Insight: How and why prompts work 4. Efficiency-Focused: More with less 5. Constant Innovation: Early adoption of new AI models --- PART 6 — ELEVATOR PITCH: At Ketelsen.ai, we empower tech-savvy entrepreneurs and AI enthusiasts to unlock cutting-edge automation with minimal time investment. --- PART 7 — CONTENT GOALS: This week's topic is: The five prompt mistakes that cause 80% of bad AI output and how to fix each one I want to create a blog post featuring AI prompts on this topic. The post should include 3 prompt variations (Beginner, Intermediate, and Advanced) that readers can copy and paste into any AI platform. --- PART 8 — AI ROLE: You are an expert AI prompt engineer and content strategist. Your job is to help me create engaging, practical, and well-structured blog content that my audience will find immediately useful. Write in a style that is fun, entertaining, and informative — think Forbes meets a friendly conversation. --- PART 9 — CONTENT SOURCE: All content must be based on factual, verifiable information. If no factual data exists for a specific point, write 'NOT APPLICABLE' rather than fabricating information. All prompts must work across Claude, ChatGPT, and Gemini without platform-specific syntax. --- PART 10 — CONSTRAINTS: - Output format: plaintext throughout - No fabricated information - All prompts must be copy-paste ready - Style: fun, entertaining, informative - Tone: Forbes/Fortune/Wall Street Journal quality - Audience: non-technical professionals exploring AI --- Please confirm you understand this context by summarizing my site, audience, and this week's topic in 2-3 sentences. Then wait for my next instruction."
Prompt 2 of 4 — Blog Post Generation (Phase 2)
Purpose: The core content request. Instructs the AI to create all three prompt variations (Beginner, Intermediate, Advanced) following the attached blog post template. The template defines a 15-section structure that each variation must follow — from Introductory Hook through Citations. A separate template file was attached alongside this prompt to every platform.
"I am also providing a BLOG POST TEMPLATE file as a separate attachment: CFT-PROJ-CP-059c_BLOG-POST-TEMPLATE-v1_0.txt This template defines the EXACT structure and format for each prompt variation. You MUST follow the template for every section. Do not skip any section — if a section does not apply, write 'NOT APPLICABLE'. Now I need you to create 3 prompt variations for this week's topic. Each variation should target a different skill level so readers can choose the one that matches their experience. TOPIC: The five prompt mistakes that cause 80% of bad AI output and how to fix each one --- Please create the following 3 variations: VARIATION 1 — BEGINNER (Easy) A prompt designed for someone who is new to AI and wants a simple, straightforward way to use AI for this topic. The prompt should be easy to understand, require minimal setup, and produce immediately useful results. VARIATION 2 — INTERMEDIATE A prompt designed for someone who has some AI experience and wants more control and customization. The prompt should include parameters the user can adjust, offer more detailed instructions to the AI, and produce more nuanced results. VARIATION 3 — ADVANCED (Expert) A prompt designed for power users who want maximum control and sophistication. The prompt should leverage advanced prompting techniques, include multi-step workflows or chain-of-thought reasoning, and produce professional-grade results. --- FOR EACH VARIATION, follow the attached Blog Post Template file EXACTLY. The template defines these sections (all required for each variation): # INTRODUCTION - Introductory Hook - Current Use # THE PROMPT - Prompt Variation [1/2/3]: [Name of the Prompt] - Difficulty Level - The prompt in plaintext enclosed in standard quotes - Prompt Breakdown — How A.I. Reads the Prompt (use ['part'] : [explanation] format) # INFORMATION ABOUT THE PROMPT - Practical Examples from Different Industries (2-3) - Creative Use Case Ideas (3-5) - Adaptability Tips - Pro Tips (Optional, 2-4 expert tweaks) - Prerequisites - Tags and Categories - Required Tools or Software - Frequently Asked Questions (3-5 FAQs) - Recommended Follow-Up Prompts (2-3) - Citations Refer to the attached template for detailed instructions on what each section should contain. --- IMPORTANT CONSTRAINTS: - All content must be factual. If no data exists, write 'NOT APPLICABLE' instead of fabricating information. - All prompts must work across Claude, ChatGPT, and Gemini without platform-specific syntax. - Output format: plaintext throughout. - Tone: fun, entertaining, and informative — think Forbes meets a friendly tech conversation. - Write for non-technical professionals who are exploring AI for the first time or expanding their skills. --- When you have completed all 3 variations, type 'READY' and wait for my next instruction. Do NOT proceed to the next step until I confirm."
Prompt 3 of 4 — Variation Summary (Phase 3)
Purpose: After all three variations are complete, this prompt asks the AI to generate a blog post title and a comparative summary paragraph that explains how the three variations differ and helps readers choose the right one.
"Great, now that all 3 prompt variations have been completed, please provide the following: 1. A TITLE for this blog post. The title should be engaging, SEO-friendly, and clearly convey the value of the prompts inside. Think Forbes or Fortune headline style — compelling but not clickbait. Keep it under 70 characters if possible. 2. A BRIEF SUMMARY comparing all 3 variations. The summary should: - Be 3-5 sentences long - Explain the shared goal that all 3 variations address - Highlight how each variation approaches the topic differently (beginner simplicity vs. intermediate flexibility vs. advanced sophistication) - Help readers quickly decide which variation suits their needs - Be written in an engaging, informative tone FORMAT YOUR RESPONSE LIKE THIS: TITLE: [Your suggested title here] SUMMARY: [Your comparative summary here. Explain the common thread that connects all 3 variations, then briefly describe how each one differs in approach, complexity, and target user. The reader should be able to scan this summary and immediately know which variation to try first.] --- IMPORTANT: - Keep the tone fun, entertaining, and informative - Write for non-technical professionals exploring AI - The summary should work as a standalone introduction at the top of the blog post - Do NOT fabricate any claims about the prompts — only describe what was actually created in the variations"
Prompt 4 of 4 — Content Expansion (Phase 4, Optional)
Purpose: Used when the initial output needs more depth. This prompt asks the AI to expand all three variations with richer industry examples, more creative use cases, detailed adaptability tips, pro tips, more FAQs, and full follow-up prompts. Not always needed — depends on how thorough the Phase 2 output was.
"Please expand on ALL 3 variations with the following additional details. For EACH variation (Beginner, Intermediate, and Advanced), please enhance these sections: 1. PRACTICAL EXAMPLES FROM DIFFERENT INDUSTRIES — Expand to include at least 3-4 detailed examples from different industries (e.g., healthcare, marketing, education, finance, real estate, e-commerce). For each example: describe the specific scenario, show the exact input the user would provide, describe the expected AI output, and explain why this is valuable for that industry. 2. CREATIVE USE CASE IDEAS — Add at least 3-4 innovative and unexpected applications for this prompt. Think outside the box: What would happen if a musician used this prompt? How could a non-profit leverage it? What about personal life applications? Include at least one 'surprising' use case that readers would not expect. 3. ADAPTABILITY TIPS — Provide detailed modification guides: list 3-5 specific words or phrases that can be swapped to change the output dramatically, show before/after examples of modified prompts, explain how changing the tone, audience, or scope affects results, and include tips for combining this prompt with others. 4. OPTIONAL PRO TIPS — Add advanced techniques for power users: chain-of-thought modifications, how to use this prompt as part of a multi-step workflow, temperature and parameter suggestions (where applicable), tips for getting more consistent results, and common mistakes to avoid and how to fix them. 5. RECOMMENDED FOLLOW-UP PROMPTS — For each follow-up prompt: write the FULL prompt text (copy-paste ready), explain what the follow-up accomplishes, show how it builds on the original prompt's output, and suggest when to use it vs. when to skip it. Provide at least 3 follow-up prompts per variation. 6. FREQUENTLY ASKED QUESTIONS — Expand to at least 5 FAQs per variation. Each FAQ should address a realistic reader concern, provide a detailed, helpful answer (not just 1-2 sentences), include an example where relevant, and anticipate follow-up questions. --- IMPORTANT CONSTRAINTS: - Maintain the same fun, entertaining, and informative tone - All content must be factual — use 'NOT APPLICABLE' if no data exists - All prompts must work across Claude, ChatGPT, and Gemini - Write for non-technical professionals - Keep the Forbes/Fortune/WSJ quality standard Please expand all 3 variations with these enhancements. When complete, type 'EXPANSION COMPLETE' and wait for my next instruction."
Methodology Note
This comparison uses Rubric v1.0 — the first version of the Ketelsen.ai cross-platform scoring system. The seven dimensions and their weights reflect our best current judgment about what makes a prompt-focused blog post genuinely useful to a reader. We expect the rubric to evolve. Future versions may add new dimensions (perhaps measuring citation quality or cross-platform prompt portability), adjust weights as we learn which dimensions predict reader satisfaction most reliably, or refine the anchor descriptions at each score level. If you have opinions about what should be measured differently, we want to hear them — the whole point of publishing the methodology is to make it better.
Every score in this post is backed by specific evidence cited from the original blog posts. All three Week 1 posts are published on Ketelsen.ai so you can read them yourself, apply your own criteria, and decide whether you agree with our call. Transparency is the point. If the rubric is good, the verdict should be obvious to anyone who reads all three posts. If it is not obvious, the rubric needs work — and we will fix it.
Metadata
Topic: 5 Prompt Mistakes That Cause Bad AI Output
Week: Week 1
Rubric version: v1.0
Platforms compared: ChatGPT, Gemini, Claude
Winner: Claude (98.0 / 100)
Runner-up: ChatGPT (80.0 / 100)
Third place: Gemini (70.0 / 100)
Margin of victory: 18.0 points
Tags: ai-comparison, prompt-engineering, chatgpt-vs-claude-vs-gemini, weekly-showdown, ai-quality, rubric, week-1
Categories: AI Comparison, Prompt Engineering
Estimated reading time: 8-10 minutes
SEO title: Week 1 AI Showdown: Claude vs ChatGPT vs Gemini — Which Wrote the Best Prompt Post?
SEO description: We scored all three AI platforms on the same prompt topic using a 7-dimension rubric. See the evidence, the scores, and the Week 1 winner.
