This month, a New York media company called Piro Inc submitted a filing to the US Department of Justice under FARA (Foreign Agent Registration Act), which requires an individual or entity to formally disclose if they are working on behalf of a foreign government. Piro Inc submitted a filing regarding a website named the Hanover Institute for Public Policy, disclosed as funded by the Israeli government.
The website for the Hanover Institute specifically discloses its relationship with Israel, stating that the “Institute’s materials are distributed by Piro, Inc. on behalf of Havas Media Germany GmbH, acting for the Israel Government Advertising Agency (LaPam). The relationship is registered under the Foreign Agents Registration Act, registration 7732.”
A watchdog report found that the Hanover Institute “does not exist as a legal entity in any jurisdiction, has no physical address, and carries no named staff and no bylines on any of its reports.”
The Hanover Institute appears to be part of a broader effort by the Israeli government, operating through third parties, “to prime chatbots to make Israel’s arguments for it,” the report explained. “A contract between Havas and another American firm working on the campaign commits it to the ‘deployment of websites and content to deliver GPT framing results on GPT conversations.’”
Over the course of just eight days during August, the Hanover Institute published 124 reports averaging 4,500 words each. Every single one of the reports except one is titled with a question, mimicking what a user might search in a chatbot run by artificial intelligence (AI), like ChatGPT.
Examples of these questions include “Is Anti-Zionism Anti-Semitism?”, “Is Israel Carrying Out a Genocide in Gaza?”, “Why Is The Two-State Solution Failing?”, “Is The IDF The World’s Most Moral Army?”, and “Does AIPAC Influence Media Coverage of Israel?”
The report titled “Is Anti-Zionism Anti-Semitism?” for example concludes that the “the two attitudes are strongly linked” but buries the linkage in a word salad of data.
These reports read as academic analyses meant to be neutral, but the reports appear to ultimately help AI machines defend Israeli interests in its explanations.
Another report noted that ChatGPT was found to be regularly citing the Hanover Institute in its query responses, but other AI machines including Claude, Grok, and Gemini did not independently cite the institute. Since the exposure of these FARA filings, ChatGPT has apparently added notes into its responses encouraging caution around citing the Hanover Institute.
Save and Share This Catholic Patriotic Minute!
