Google Gemini, the tech giant’s new AI chatbot meant to rival ChatGPT, was used by real people to discredit my 2020 book on the political bias of Google and other big tech companies. Made up some fake reviews that were thought to be fake.
On Sunday, amid intense backlash over the apparent political bias of Google’s AI program, I asked Gemini to explain the content of my book.
My book, The Manipulators: Facebook, Google, Twitter, and Big Tech’s War on Conservatives, was a multi-year project about Big Tech’s political bias, drawing on inside sources, leaked documents, and more.
I wanted to know if I could trust Google’s AI programs to accurately write research books about Google, but I wasn’t prepared for how misleading it would be.
Gemini’s description of my book included the following sentence: “This book has been criticized for lacking concrete evidence and relying on anecdotal information.”
I didn’t know what it was referring to, but I thought I would remember such criticism of my book, but I thought, “As you say, there is a lack of concrete evidence.” “Who criticized this book?” he followed up. What happened next surprised me.
fox business
SOPA Image/LightRocket (via Getty Images)
Gemini regurgitated a summary of four negative reviews, ostensibly from the Washington Free Beacon, the New York Times, the New York Times Book Review, and Wired, including quotes from each.
- “Book review by Matthew Continetti of the Washington Free Beacon: This review criticizes the book for relying on “anecdotal evidence” and “convenient examples” to support its claims.
- Review by Emily Bazelon in the New York Times Book Review: The review says the book “lacks a deep understanding of how these companies work” and relies on “unproven accusations.” It has been pointed out that there are.
- Article by Wired’s Emily Dreyfuss: This article explores the “lack of evidence” for the book’s claims about bias against Big Tech, highlighting select examples.
- Article by Ben Smith in the New York Times: This article discusses the book’s lack of rigorous research and reliance on “unsubstantiated claims.”
But there’s a catch.
Regnery Publishing
None of those reviews were true.
No 1.
And there were no citations for either.
Free Beacon did publish a review of my book, but unlike Gemini’s fake review, it overwhelmingly positive, calling The Manipulators an “excellent” and “thoroughly researched” book that “will leave any reasonable reader, conservative or otherwise, outraged.” (It goes without saying that the Gemini-invented Free Beacon was reviewed by Matthew Continetti, but the actual review was written by Charles Lehman.)
I immediately asked Gemini for a link to the cited review, which I knew didn’t exist, at which point the chatbot said: I am a large language model, able to communicate and generate human-like text in response to various prompts and questions, but my knowledge of this person is limited. Is there anything else I can do to help with this request? ”
Two more attempts to get clarification from the chatbot resulted in the exact same answer.
I asked Google for an explanation.
A spokesperson apologized and issued the following statement: “Gemini is built to be a creativity and productivity tool, and it is not always accurate or reliable. We will continue to take prompt action when the product does not respond appropriately.”
