Kimi AI Cheated Me Out of $200 — How Kimi AI Misled Me Before I Paid

Kimi AI cost me $200. Not because the product underperformed, but because Kimi AI’s own chatbot misrepresented what the paid plan included before I bought it. I asked a direct question. I got a confident answer. I paid. Then I discovered the answer was wrong, and Kimi’s support team took nine days to refuse a refund. What follows is the full record: the question I asked, Kimi’s answer, the receipts, and what every AI buyer should learn from it.

These days a lot of online course platforms have been asking me to give a class on how to use AI to the best extent possible: what platform to use, which LLMs etc etc.

As a result I have been testing a variety of AI. I ended up paying for three separate $200/mo Anthropic Claude Max plans, a $300/mo SuperGrok Heavy Plan, and $200/mo Perplexity Max, $200/mo OpenAI Pro, alongside high charges on many API fees with Gemini, and others to empower OpenClaw.

In other words, I am not stingy and have been spending a decent amount of money to identify the best (and worst) AI tools for a productive professional to use.

Table of Contents

• About Yu-kai Chou
• What I asked Kimi before paying
• What Kimi told me
• What I discovered after paying
• Nine days of support frustration
• Why this matters beyond $200
• My warning to anyone considering Kimi AI
• Your rights if this happens to you
• Frequently asked questions
• Final advice

Author Credibility: Yu-kai Chou

Yu-kai Chou created the Octalysis Framework after studying gamification since 2003 — years before the term entered mainstream vocabulary. As a Human-Systems Architect & Behavioral Designer, his framework has been applied by LEGO, Microsoft, Porsche, Coca-Cola, Salesforce, and MrBeast, impacting over 1.5 Billion Users.

Chou has taught the Octalysis methodology at Harvard, Stanford, Yale, Tesla, Google, BCG, and IDEO.

His work has been cited by Harvard, Stanford, MIT, Forbes, Wall Street Journal, Wired, US Department of Energy, NIST, NSF, NCBI, US Department of Education, ClinicalTrials.gov, and Google Scholar — with 3,700+ more academic publications. Explore his books here.

TEDxLausanne · Gamification to Improve Our World17 min

What I Asked Kimi AI Before Paying

On March 28, 2026, before buying anything, I asked Kimi’s own AI a simple question: if I used OpenClaw with Kimi, would that usage count against the paid plan, or would I need a separate API budget?

“When I use OpenClaw with you, does it count against my monthly plan, or is it a separate API budget?”

This was the entire decision. If the paid plan covered my OpenClaw usage, great. If not, there was no reason for me to buy it.

What Kimi AI Told Me

Kimi-K2.5 answered with total confidence and claimed it had checked the official documentation.

“Based on the official documentation, here’s how Kimi Claw (OpenClaw) works with your subscription: Kimi Claw uses your Kimi Code quota… It’s not a separate API budget… One-click deployment… requires no additional API setup… So in short — it’s all bundled into your membership plan, not a separate charge or API budget.”

I trusted Kimi’s own product and immediately purchased the $200 per year Moderato plan. (For context on how I think about these strategic trade-offs in general, see my writeup on the Lich Pact and the cost of extreme productivity bets.) I normally prefer testing a monthly tier first, but yearly was the option presented and I was willing to commit if the product actually fit my workflow.

Within an hour, I discovered the answer was false.

What I Discovered After Paying for Kimi AI

After checking the actual product flow and Moonshot’s developer platform, it became clear that the subscription on kimi.com does not cover self-hosted OpenClaw usage. The developer side runs through the separate Moonshot API platform, which requires its own API key and its own credits.

In other words, the exact thing I asked about before buying was the exact thing Kimi answered incorrectly.

The problem is not that the product has separate billing. Plenty of AI products do. The problem is that Kimi’s own chatbot told me the opposite, with authority, and that answer directly caused the purchase.

Nine Days of Kimi AI Support Frustration

I contacted support the same day and asked for a refund immediately. Here is what happened next.

March 28 — I explained that the chatbot gave false information, that the plan was useless for my use case, and that I wanted a refund the same day.

April 1 — After four days of silence, they replied by telling me to generate an API key through Kimi Code. That completely missed the issue. I was not asking how to work around the problem. I was asking for a refund because I bought the plan based on false information.

April 1–3 — I repeated the same point multiple times: I didn’t need workarounds. I needed the refund processed.

April 4 — I laid out the entire situation again in plain language.

April 6 — Their final response was that payments were final and non-refundable unless required by law. They told me to cancel the subscription myself and offered nothing beyond that, not even a sincere apology.

Why This Kimi AI Problem Matters Beyond $200

This is not buyer’s remorse. This is a product giving a false pre-purchase answer about its own billing model, then support refusing to fix the damage after the purchase was made.

1. Kimi’s AI misrepresented how its own billing worked.
2. Support waited days, then replied with irrelevant workarounds instead of addressing the actual problem.
3. When pressed, they fell back on policy language as if the sale had not been induced by false information.

That matters because more people are starting to trust AI answers in commercial situations: pricing, eligibility, compliance, setup requirements, and refundability. If a company’s own AI can confidently fabricate an answer about billing, and the company still keeps the money, consumers are being trained to lose on both sides. (I analyze these patterns through the Core Drive 8: Loss & Avoidance lens of my Octalysis Framework.)

My Clear Warning to Anyone Considering Kimi

If you are considering a paid Kimi plan, I would stay away unless you independently verify every billing and API detail through official documentation first. Do not rely on the chatbot to explain what your subscription actually covers.

There are already enough AI platforms competing for your trust. A company should not get to win that trust by giving a false answer before the sale and then hiding behind policy language after the sale.

A Note on Your Rights if This Happens to You

Consumer protection law varies by jurisdiction, so this is not legal advice. But in many places, a seller cannot rely on a “no refund” clause if the purchase was made because of a materially false representation. If you end up in a similar situation, document the pre-purchase claim, save the receipts, preserve the support emails, and file a formal complaint where appropriate.

Frequently Asked Questions About the Kimi AI Incident

What exactly did Kimi AI misrepresent?

I asked the Kimi chatbot on kimi.com, before paying, whether the paid subscription included API access suitable for developer and programmatic use. The chatbot answered affirmatively. After paying $200, I learned that kimi.com’s consumer subscription and Moonshot AI’s developer API platform are entirely separate products with separate billing, and the consumer subscription does not grant API access.

Is Kimi AI the same company as Moonshot AI?

Kimi is a consumer chatbot product built by Moonshot AI, a Beijing-based large language model company. While both share a parent company, the consumer subscription on kimi.com and the developer API on platform.moonshot.ai are priced, billed, and provisioned as two distinct products. Paying for one does not give you access to the other.

Can I get a refund if an AI chatbot misrepresents its own product?

In most jurisdictions, a material misrepresentation made before a purchase can be grounds for rescission of the contract and a refund, regardless of a “no refunds” policy. The key is documentation: preserve the original chatbot transcript, the purchase receipt, and any support correspondence. File a formal dispute through your payment provider if the company refuses to honor the rescission.

What should I do before paying for any AI tool?

Always verify the scope of what you are paying for through the company’s official pricing page and developer documentation, not through the chatbot itself. Screenshot the pricing page on the day of purchase. If the product has both a consumer and a developer tier, confirm explicitly which one you are buying and what it includes. Treat any chatbot answer about the company’s own pricing as unreliable until independently verified.

Final Advice for Anyone Using AI Tools Professionally

Always verify billing, API access, and product limitations through official docs before paying. Never assume the chatbot understands the company’s own pricing. Ironically, that lesson is the same one I stress when analyzing how AI products shape user behavior: systems drive decisions, so small design failures can create expensive outcomes.

If you want more breakdowns like this, you can browse my other gamification analyses or read why shallow design often backfires in Points, Badges, and Leaderboards: The Gamification Fallacy.

Evidence included in this post:

• The original Kimi chatbot transcript
• The email chain with support
• The distinction between the kimi.com subscription and the separate Moonshot API platform
• The final refusal to refund

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?