Is It OK to Launch an AI-Built App? What the Rules Say and How to Disclose

Does any platform stop you launching an AI-built app?

Not on the pages we read. The rules of this board ask you to launch something you made or helped make, and they say nothing about which tools you used. Hacker News is the place that draws a visible line, and it draws it around effort and presence, not around tooling. Its Show HN guidelines, read on 13 November 2026, say the project should be non-trivial, tell people not to post quickly generated one-offs, ask you to explain how and why you made it, and require that you worked on it personally and are around to discuss it.

Read that as a quality bar. An app you steered through many sessions, tested and understand is not a one-off, whichever tool typed the code. A prompt pasted once and published unseen is the thing the guideline is aimed at.

What does Google say about AI-generated content?

We read Google Search Central’s guidance on generative AI content on 13 November 2026 (its page shows an update date of 1 October 2026), along with its page on helpful content. Its points, in brief:

• Generative AI can help when researching a topic and adding structure to original content.
• Using AI to generate many pages without adding value for users may violate its spam policy on scaled content abuse.
• Focus on accuracy, quality and relevance, and manually fact-check AI output before publishing, including the title, meta description, structured data and image alt text.
• Sharing how content was created can give readers context, and disclosures are useful where someone might think “How was this created?”
• Fabricating creator profiles, such as made-up names or AI-generated headshots, is deception and makes a page untrustworthy.

Notice what this covers. It is guidance about content on web pages that appear in search, so it applies to your landing page copy and any articles you publish, and it does not rule on whether an app’s code may be machine-written. For your launch page the practical reading is simple: say who made it, say how, and check every claim.

What do people who try your app actually care about?

Whether it works, whether it is safe with their data and whether you will answer. Code that runs is not code that is safe. Gaps that turn up in generated code, as in any rushed code, include keys left where the browser can read them, pages any signed-in person can open, form input that is never checked and a payment step that charges without delivering. You do not need to distrust the tool, but you do need to read the parts that touch accounts, money and private data, and test them as an attacker or a clumsy user would. The vibe coding security guide lists what to check, and the is vibe coding bad guide weighs the trade-offs.

If you collect email addresses or anything personal, publish a short privacy page that says what you keep and why. That matters more to a visitor than which tool wrote the form.

What does honest disclosure look like on a launch page?

A short, factual note, placed where a curious visitor will find it, such as the foot of the page or the description. It names the tools in plain words, says what you reviewed and tested yourself, and lists what you know is rough. It reads like a fact, not a confession, and it does not hide the tools or lead with them.

How this was built: I made [product] with [AI coding tool] over [number] weeks, and I wrote or rewrote [the parts you did by hand, for example the pricing logic and the data export].

What I checked myself: [sign-in and access rules], [payment flow], [how data is stored and deleted], and [what I tested, for example on a phone and with two new users].

What is still rough: [honest gap one], [honest gap two].

Questions or problems: [contact address]. I answer within a day.

• Do not claim “hand-coded” or “no AI” if it is not true. It is the one line that will damage trust if someone checks.
• Do not apologise at length. One sentence of fact is enough.
• Do not imply a review you did not do. “Reviewed for security” means you read the relevant code.
• Do not invent a team. Google’s guidance calls fabricated profiles deceptive, and visitors agree.

Who should write the launch post and the replies?

You. Hacker News’s guidelines say not to put generated text in posts, and ask for comments that are not generated or AI-edited, because the site is for conversation between humans. Elsewhere too, a reply that reads as machine-written can feel like a reply from nobody. Use AI to build and to draft; use your own voice for the post, the first comment and every answer. Add the specifics only you know: what surprised you, what you cut and why, what a user did that you did not expect. Those are also what make a page worth reading, which is Google’s point about adding value.

How do you tell whether being AI-built is helping or hurting?

Watch the comments. If several people ask how it was built, move the answer into the page. If they ask about safety, answer with the specifics of what you checked. If nobody mentions it, the disclosure is doing its quiet job. You cannot measure the effect with a number, because it is tangled up with everything else on the page, so treat the questions as your evidence.

When you are ready, launch your product free with the note in place, and see the playbook for vibe coders for the wider plan. If you want to see what builders are making, browse AI agent ideas, and read how to launch an unfinished product if parts of yours are still rough.