Have you seen posts on X that look like screenshots of New York Post headlines, which have a caption asking the xAI chatbot Grok to determine if the story is real or not? Lead Stories has spotted a number of such posts recently from an account @RabbiJeffx. The account bio is labelled 'parody & satire' and says they are a fictional character. The posts do not show real New York Post article headlines. Some of the fake screenshots carry a @RabbiJeffx watermark, but many don't. In some cases the headline text may refer to something that did happen, but the accompanying photos are from an unrelated situation.
One example appeared in a post (archived here) by the @RabbiJeffx account on X on Sept. 26, 2026. It read:
Hey @grok
Before anything else, start with the religious background, please.
The fake New York Post headline read:
Meet the man who rewrote Australia's rules on abortion, gay sex and marijuana
This is what the post looked like on X at the time of writing
(Image source: post by @RabbiJeffx on X.)
A search for the image in the screenshot with Google Lens (archived here) did not find results pointing to a New York Post article. The image pictured appeared in a writeup in the Sydney Morning Herald on March 6, 2022 (archived here). The article was titled "Moss Cass: pot-smoking Cabinet minister who helped change Australia". Lead Stories conducted three more searches: two on the New York Post website for "Moss Cass" (archived here) and "abortion, gay sex and marijuana" (archived here) and finally, an advanced search with Google of "site:nypost.com" for the full text of the fake headline (archived here). None of these searches pointed to relevant results. This headline never appeared in the New York Post.
What these fake New York Post screenshots look like
The fabricated screenshots contain a red New York Post banner across the top with "Log In" displayed in the upper right corner. Along with the black text of the fake headline, there may be a gray @RabbiJeffx watermark. The fake screenshots do not show a byline or dateline for the article or an image credit as a real New York Post article would. The posts have a brief caption which may contradict the headline in the image, as if @RabbiJeffx is calling out false reporting by the New York Post. The posts typically tag the xAI chatbot with a "Hey @grok" or "@grok yes/no".
Some examples
A Sept. 26, 2026, post (archived here) bears the headline, "ANTISEMITIC HORROR: Jewish teacher slashed with knifes in the name of ISIS terrorists" (embedded below). The photo shows a man smiling from a hospital bed and another man standing beside him. The post caption contradicts the fake headline. It reads:
This Jewish teacher was CAUGHT FAKING an ANTISEMITIC ISIS knife attack against himself.
@grok yes or no
This Jewish teacher was CAUGHT FAKING an ANTISEMITIC ISIS knife attack against himself.@grok yes or no
-- Rabbi Jeff (@RabbiJeffx) September 26, 2026
The photo used in this fake screenshot was taken out of context, it originally appeared in the Winter 2016 issue of Jewish Action in an article (archived here) about kidney donation. The real caption with the photo reads:
Kidney donor Yoel Usher Labin with recipient Robert Sendzischew, as they both recuperate at Mount Sinai Hospital. Photo: VINnews.com/Rafael Lopez JR
The fake headline and caption prompt together led Grok to summarize an unrelated incident involving different people. The reply from Grok mentions the case of Tsion Sylvain Saadoun. A May 12, 2016, article in the French newspaper Le Monde (archived here) tells of a Jewish teacher in Marseille, France who was convicted and given a suspended sentence for making a false accusation regarding an antisemitic attack that he said happened on Nov. 18, 2015. Grok's reply (archived here) displays as the first comment under the @RabbiJeffx post on X (pictured below).
(Image source: post by @grok on X.)
The previously mentioned "Meet the _____" structure of the fake New York Post headline has been repeated many times by the @RabbiJeffx account (pictured below). In each of these examples the xAI chatbot Grok was tagged and asked to "Start with religious ethnicity, please." Lead Stories did four more searches using the Google advanced search method (mentioned above) for the headline text and "site:nypost.com" -- the examples below feature made-up headlines which never appeared in the New York Post:
- Meet the woman behind Argentina's legalization of abortion -- Sept. 27, 2026, post (archived here)
- Meet the woman behind transgender studies in schools and universities -- Sept. 27, 2026, post (archived here)
- Meet the doctor behind the legalization of abortion in Canada -- Sept. 27, 2026, post (archived here)
- Meet the woman who rewrote Japan's rules on divorce, abortion and feminism. Sept. 26, 2026, post (archived here)
(Image source: posts by @RabbiJeffx on X.)
An example where the xAI chatbot Grok is used to bolster a false premise (a classic strawman argument -- archived here) can be found in a post from Sept. 25, 2026. It shows a man in a hospital bed, the fake headline reads: Antisemitic Horror: Jewish man brutally stabbed outside Michigan synagogue. The @RabbiJeffx caption contradicts the fake headline, saying the man stabbed himself and faked the attack. Grok is tagged by @RabbiJeffx and falsely confirms:
Yes. The man shown is Sean Sammit, who in Dec 2019 stabbed himself at Temple Kol Ami in West Bloomfield, MI, then falsely claimed an antisemitic attack. Police found the knife and blood inside the building; his Apple Watch data disproved his story. He admitted staging it and was charged with falsely reporting a felony.
This JEWISH MAN stabbed HIMSELF and FAKED an antisemitic hate attack.@grok Yes/No
-- Rabbi Jeff (@RabbiJeffx) September 25, 2026
The incident regarding false claims made by Sean Sammit is real, as reported in a Fox5 NY article (archived here) but the man shown is not Sean Sammit. It is Rabbi Shlomo Noginski who really was attacked in the Brighton neighborhood of Boston on July 1, 2021. The photo appears at 43 seconds in a WCVB Channel 5 Boston news segment about the stabbing (archived here). The man accused of stabbing Noginski, Khaled Awad, was arraigned (archived here) on Aug. 26, 2021.
Lead Stories published a fact check about a post (archived here) with a similar false refutation dynamic -- it referred to a fake vandalization of a dorm room door, but contained photos of a synagogue door which really was vandalized (embedded below).
This Jewish student was CAUGHT staging a fake ANTISEMITIC attack on her own dorm door.@grok Yes/No
-- Rabbi Jeff (@RabbiJeffx) September 24, 2026
Keywords
Here are keywords and phrases often associated with posts like these: @grok yes or no, Hey @grok, Before anything else, start with the religious background, please, caught FAKING antisemitic attacks, and Start with ethnicity and country, please.
How to tell they are false
If you see posts like these on social media, here are some things you can try.
Search for the text of the headline and the image separately. There are several questions to investigate:
- Did the New York Post publish this article?
- What does this picture show?
- Does the headline match the picture?
- Is the response from Grok relevant to both the photo and the headline?
To check if the headline has been published, search directly on the website, nypost.com/search, or do an advanced search with Google using the 'site:' search operator including the exact text of the headline inside quotation:
Google Lens (accessed by clicking the camera icon in the Google search bar) has a feature where a user can select the portion of an image to search. Another feature allows a user to highlight and then search the text appearing in an image. The @RabbiJeffx posts typically combine elements of stories that may be true or partially true with other stories taken out of context. The xAI chatbot is also prompted to respond in a way that may be unpredictable, misleading, or false.
Unsure about a post or video? Email [email protected] and we will take a look!
These materials were developed in 2026 for the Prebunking at Scale project, with support from the European Fact-Checking Standards Network. If you share this on social media, use #prebunkingatscale.
