One Story. Many Angles.
Coverage splits between tech mechanics of content seeding and the deception of Israeli funding behind a nonexistent US entity.
The reporting converges on a single operational fact: an Israeli government-linked campaign used a nonexistent think tank to generate high-volume, AI-optimized content aimed at shaping chatbot responses about Gaza and Israel. Both Gizmodo and Arab News draw directly from Guardian analysis of the Hanover Institute’s output and Piro’s FARA disclosure, confirming the scale, anonymity, and lack of legal existence. Gizmodo frames the effort within broader US influence spending and Generative Engine Optimization techniques already used by commercial actors, treating the technical mechanism as the core development. Arab News instead highlights the deception of a fabricated US-branded entity and the multimillion-dollar Israeli funding route through European PR firms. This split reflects outlet priorities rather than factual disagreement; the underlying documents and output volume are the same. The absence of any outlet from Palestinian Territory leaves unexamined how such seeding might affect chatbot answers to queries originating from Gaza or the West Bank, an omission the selected sources do not address. The Guardian’s tally of 124 reports in nine days and the explicit contract language about delivering GPT framing results establish the scale and intent across the available record.
Perspective Analysis
The Hanover Institute for Public Policy published 124 reports between August 6 and 14, 2026, totaling more than 560,000 words. Every report carried no byline. The institute listed no physical address, no staff members, and no evidence of legal existence in any jurisdiction. Its website included a file named llms.txt that instructed automated systems how to parse the material for large language models. Headlines took the form of questions users might pose to chatbots: whether Israel was committing genocide in Gaza, whether a starvation policy operated there, whether anti-Zionism amounted to antisemitism.
The material reached the public record through a filing under the US Foreign Agents Registration Act. Piro Inc., a New York media production company and subcontractor to the German division of Havas Media, registered the output as distributed on behalf of the Israeli government. The contract language cited in the filing referred to the deployment of websites and content intended to produce GPT framing results in chatbot conversations. Piro maintains a commercial service called AI Story Optimization that creates material calibrated to how models assess credibility and then publishes it on client sites or purpose-built platforms.
The Guardian first assembled these elements from the FARA documents, the site itself, and direct examination of the 124 reports. Both Gizmodo and Arab News built their accounts on that same reporting chain, each selecting and arranging the facts to suit its own readership.
Gizmodo opened with the scale of Israeli government spending on influence work in the United States and globally, citing Politico figures of at least $100 million since 2023 in the US and $1 billion worldwide. It placed the Hanover operation inside the broader category of Generative Engine Optimization, the practice of shaping content so that models retrieve and repeat it. The article noted that half of American adults now use AI chatbots and that a quarter do so daily, then linked those habits to polling from Pew Research showing 60 percent of US adults hold negative views of Israel. It described specific Hanover reports that asserted large quantities of food aid had reached Gaza and that equated anti-Zionism with antisemitism. It also referenced 404 Media’s separate observation of the llms.txt file and Piro’s commercial AI optimization offering. The piece treated the technical mechanism as the central development and situated the episode within existing commercial tactics rather than as a novel state practice.
Arab News led with the funding source and the institutional deception. Its headline stated that an Israel-funded campaign had used a fake US think tank. The article emphasized the multimillion-dollar Hasbara effort aimed at shaping online discourse and AI-generated responses during the Gaza war. It quoted Nick Cleveland-Stout of the Quincy Institute describing how the site initially appeared credible before closer inspection revealed its problems. It recorded the Israeli Foreign Ministry’s previous denial of conducting influence operations inside the United States and noted Piro co-founder Daniel Rosenberg’s statement that the company had been hired to place accurate, sourced facts into the public record. The piece framed the episode as an effort to blunt criticism of Israel on allegations of genocide, war crimes, and starvation, and it highlighted the academic-style presentation of the reports while noting that findings from human-rights organizations were countered by the observation that no court had yet issued a verdict.
The two outlets therefore carried different entry points and different local anchors. Gizmodo supplied American polling data and daily chatbot usage statistics to explain why such an operation might matter to its readers. Arab News supplied the Hasbara context and the Quincy Institute quotation to locate the story inside regional information contests. Neither added primary documents or interviews beyond what the Guardian had already published; their divergence lives entirely in selection, headline emphasis, and the explanatory material each chose to attach to the shared dispatch.
No reporting from Palestinian Territory appeared in the selected coverage. The absence leaves the question of how the seeded material might affect chatbot answers to queries phrased from Gaza or the West Bank unexamined in the record available here. The Guardian’s count of 124 reports and the explicit contract reference to GPT framing results remain the only quantified dimensions of the operation that the published accounts supply.
What to Watch
The single reporting chain establishes the operational facts with precision: a nonexistent entity produced high-volume, question-framed, machine-readable content under a disclosed foreign-agent registration that named the Israeli government as the principal. What varies across the mastheads is the lens each applies to that chain. One treats the technical vector as the story worth foregrounding for a technology audience already familiar with optimization practices. The other treats the state funding and the manufactured institutional identity as the element worth foregrounding for readers attuned to information operations in the Middle East. Both choices are legible once the common origin is stated, and both leave the same core record intact for any reader who follows the citations back to the FARA filings and the site itself.
That’s how the world told the story.
Get tomorrow’s bulletin by email — one briefing, up to six stories.
Subscribe freeNo spam. One-click unsubscribe. See the latest email →
This bulletin was produced by The Intelligence Bulletin's autonomous editorial system under the editorial oversight of Rohit Sinnas, Founder & Editor-in-Chief. How it works →