Briefing: Disinfo spreads in wake of “El Mencho” killing

Our weekly Briefing is free, but you should upgrade to access all of our reporting, resources, and a monthly workshop.

Craig published the Indicator Guide to investigating ecommerce sites. It covers how to identify the platform a store runs on and offers tools and techniques for tracing ownership, analyzing traffic and ads, doing a safe test purchase, and more.

Alexios wrote about Digital News Now, a one-person podcast network pushing out 11,000 AI-generated episodes a day. He analyzed more than 100 episodes and found that they often reused the same facts, structure, and even phrases as articles from a range of media outlets.

📍 The killing of Mexican drug lord “El Mencho” was followed by extensive real-life retaliatory violence by members of his gang, the Cártel Jalisco Nueva Generación (CJNG). It also saw false and misleading information spread on social media. El Sabueso has a round-up.

📍 Prediction market platform Kalshi suspended the account of an editor that works for MrBeast and reported him for insider trading. It said he had placed bets about the YouTube creator's videos with “near-perfect” success. (Now find the bigger fish — Alexios)

📍 A hacker circumvented Claude’s guardrails and used the LLM to steal 150 gigabytes of government data, according to Bloomberg.

📍 X is working on a way for users to self-report if their post contains AI-generated content. (This is both late and futile, but presumably the company thinks it will help meet regulatory requirements — Alexios)

📍 Police on the Channel Island of Jersey are investigating a TikTok account that impersonated a local school and posted deepfake videos of teachers eating dog food and brandishing guns.

📍An officer from the US Internet Crimes Against Children (ICAC) taskforce testified in court that he believes Meta’s use of AI for content moderation is generating an onslught of “junk“ reports about cases of child sexual abuse. (via Henry Ajder).

📍A faceswap video purporting to show imprisoned Jeffrey Epstein associate Ghislaine Maxwell walking free in Canada got more than 16 million views.

📍 Twitter Viewer is a free tool to view public X profiles, tweets, photos, and videos without being logged in. The same developer made a similar tool for Instagram.

📍 YT Havester is a new, free Python tool from Aida Kokanovic that downloads “videos, transcripts, metadata, and comments from YouTube — with forensic integrity verification built in.” I’m a pretty committed Stacher 7 user for downloading videos from digital platforms, but Harvester has a bunch of nice features.

📍 Kevin D Reyes, who does OSINT investigations for the Institute for Strategic Dialogue, released ISD's Federal Courts Guide, a Start.me page of resources and links.

📍 BetaMeta is a free to tool that can “extract and visualize temporal metadata from any URL.” (via Mario Santella)

📍 Videosearch.io is a free tool that searches across major video and social platforms. (via Cyber Detective)

📍 Nicole Hurey launched a new tool to craft Google Dorks. It joins the recently-launched dork tool from ShadowDragon that we highlighted in a recent Briefing.

📍 Speaking of ShadowDragon, it also launched the Open Sources Toolkit, a database of tools.

📍 PlaceSpotter is an AI-assisted photo geolocation tool. (via Cyber Detective)

📍 The New York Times ran more than 1,000 tests on AI detectors to see how well they performed at detecting real, synthetic, and AI-edited images, video, and audio. Stuart A. Thompson writes:

📍 We're having a party! Indicator will host its first in-person event at the International Journalism Festival in Perugia, Italy on April 16. We’ll be sharing more details soon but if you’re a paid member and will be in Perugia, let us know so we can invite you for food and drinks. If you’re attending IJF, why not become a member so you can attend?

📍 OpenAI’s latest threat report revealed that bad actors attempted to use its AI models to conduct romance scams in Indonesia, help run a content farm in Russia, and to assist with a Chinese social engineering effort targeting Americans on LinkedIn. Of particular note was an account tied to a Chinese law enforcement official who used ChatGPT for a covert influence operation targeting Japanese Prime Minister Sanae Takaichi.

📍 Developmental psychologists told The New York Times that AI slop videos targeting may kids may be cognitively overwhelming, though they cautioned that systematic research is needed.

📍 Channel News Asia found that ”hundreds of AI-generated, Chinese-language YouTube videos have been targeting Singapore and Prime Minister Lawrence Wong as part of an ongoing disinformation campaign.” The videos denigrated Wong and cast doubt on the city-state’s economic success. YouTube banned two related channels that it said violated its spam, scams, or other deceptive practices policies.

📍 Rapper Nicki Minaj has become increasingly vocal on X about her support for President Donald Trump and for conservative positions on hot-button social issues. The security company Cyabra claims that during this pivot 33% of the accounts engaging with her content were inauthentic. It concluded “with high confidence that a coordinated fake campaign was actively amplifying political content on Nicki Minaj’s X account during the period reviewed.”

Want more studies on digital deception? Paid subscribers get access to our Academic Library with 55 categorized and summarized studies:

Please upgrade to access to all of our content, including our how-to guides and Academic Library, and to our live monthly workshops.