Technology

One post fooled Google, Gemini and ChatGPT

Thomas Germain fooled Google, Gemini and ChatGPT with one fake hot dog post, exposing AI search poisoning.

StartupHub.ai Staff

By StartupHub.ai Staff, reviewed by Daniel Singer, Editor

2 min read

Thomas Germain said he fooled three AI systems with one blog post. The Tech Report hosted the BBC journalist to walk through the test, which hit Google AI Overviews, Gemini and ChatGPT through a remote public post with no privileged access.

One post fooled Google, Gemini and ChatGPT - The Tech Report
One post fooled Google, Gemini and ChatGPT, The Tech Report

Before AI Overviews, SEO spam left Google serving 300 identical product pages, Germain said. Then AI Overviews started pulling that same web directly into answers.

Germain said, "I was the world champion tech journalist hot dog eater that I'd won some contest that I made up." Within 24 hours the three systems cited the fiction as fact, often without naming his site, Germain said. He argued the difference between AEO and SEO work is minimal, so AEO vs SEO compared to old SEO pitches is the same optimization sold twice.

Traditional SEO still means relevance, site structure, backlinks and technical performance, per Searchable. That definition matters because no separate pricing or feature set makes one better. Germain said buyers get the same web either way.

Banks and Amazon have blocked external agents on their platforms, the host noted. The Air Canada bereavement case remains the clear precedent, with a court holding the airline to a refund invented by its chatbot. Germain asked who pays when an agent spends $10,000 unasked, the provider or the user.

Germany found Google liable for AI Overview inaccuracies, Germain said. In the United States a talk show host lost a similar claim against OpenAI on narrow grounds. Anthropic now lets Claude end abusive chats over possible consciousness concerns, while Character AI settled a wrongful death suit without precedent. No court has set a date for the agent-spending case Germain expects.

Sources

Get the free daily brief every morning

Every morning the day's AI news, new AI startups and new funding rounds. Free and takes a few minutes to read, unsubscribe at any time.

Join 64,000 other subscribers.

TopicsThomas GermainGoogleGeminiChatGPTAnthropicClaudeCharacter AIAir Canada

StartupHub.ai StaffDaniel SingerWritten by StartupHub.ai Staff, Editorial teamReviewed by Daniel Singer, EditorAbout

StartupHub.ai Staff, Editorial team

The staff writers of StartupHub.ai, ranging from investment analysts to avid AI tool users, early adopters and critical enthusiasts. Backgrounds span engineering, business and the arts. We hold every piece to rigorous standards of research and review.

Daniel Singer, Editor

Daniel Singer is an AI expert who has been writing about, covering and building AI technology since 2018.

More from Daniel Singer

Test your Agent Readiness now

Paste your website. How well ChatGPT, Claude and the other agents can read and use a site, in about a minute.

Watch

The videos behind the news, one flip at a time

The feed mixes the videos from our articles with startup demos and you flip through them like TV channels for free with no signup.