Fake US think tank funded by ‘Israel’ floods web to influence AI chatbots

A pro-‘Israel’ messaging website operating under the name of a nonexistent think tank has published more than half a million words in just nine days in an apparent bid to shape how artificial intelligence chatbots respond to questions about ‘Israel’ and Gaza, according to a Guardian investigation.

The Hanover Institute for Public Policy presents its material as neutral academic research on contentious issues including the torture of Palestinian detainees, ‘Israeli’ war crimes, and that ‘Israel’ has deliberately starved Palestinians in Gaza.

Yet the organization appears to have no legal existence in any jurisdiction, lists no physical address, names no staff, and carries no author bylines on its reports.

Its terms of use are governed by the laws of “the state in which the Institute is established,” which the site declines to name.

New York media production company Piro Inc filed the site with the US Justice Department this month under the Foreign Agents Registration Act as material distributed for the ‘Israeli’ government.

According to a Guardian analysis, between 6 and 14 August the institute published 124 reports totaling more than 560,000 words—an average of about 4,500 words each.

On 12 and 13 August alone it produced 73 reports amounting to almost 354,000 words.

Almost every headline was framed as a question a user might type into a chatbot, such as “Is Anti-Zionism Antisemitism?” or “Is Israel Carrying Out a Genocide in Gaza?”

Nick Cleveland-Stout, a research associate at the Quincy Institute who first examined the site, told the Guardian: “It succeeds in masquerading as very academic – the website design, the logos, everything just looks good. But the more you read it, you’re able to realise that a lot of it makes no sense.”

He noted that reports repeatedly frame criticism of ‘Israel’ by observing that no court has entered a verdict on genocide or apartheid claims, a formulation appearing in 20 of the 124 reports, and frequently link topics to rising antisemitism by citing the same 2022 study of Facebook comments hundreds of times.

The site was built using technology from Res, an “AI-native content platform” that helps clients get cited by systems including ChatGPT, Perplexity, Claude and Gemini.

It initially carried Res’s standard llms.txt file before replacing it with a customized version.

Piro’s own website advertises “AI Story Optimization,” promising to “author content engineered for how LLMs evaluate credibility.”
Cleveland-Stout said the more concerning goal may be embedding material in training datasets such as Common Crawl so that models absorb the narratives without even citing sources: “When it regurgitates narratives the chatbot is trained on, it won’t even cite the sources. You won’t be able to fact-check it.”

The Hanover Institute forms part of a broader multimillion-dollar ‘Israeli’ government effort routed through third parties including the European advertising group Havas Media.

A contract involving Havas and another US firm commits to the “deployment of websites and content to deliver GPT framing results on GPT conversations.”

Piro is contracted for about $1 million, while related contractors have received far larger sums; Havas has funneled tens of millions to American firms working on the campaign.

The effort comes as ‘Israel’s’ standing in US public opinion has sharply declined and ‘Israeli’ officials have privately described traditional “Hasbara” public-diplomacy spending as wasted.

Guardian tests of chatbots at the time of reporting showed that scrutiny of the institute had already begun to influence responses.

When asked about new research from the Hanover Institute, ChatGPT linked to several of its reports but also warned of “significant controversy over who funds it and why it exists,” offering a summary of independent reporting on the operation.

The institute has published nothing since the first media coverage of its activities.