In the “Age of Generative Warfare,” the fog of war is no longer a metaphor, it’s an algorithmic reality.
From deepfakes of the 2026 Middle East crisis to “AI slop,” our defenselessness against synthetic media is a mandate for Critical AI Literacy. We must interrogate the tools before they automate our reality.
Read my latest on why CAIL is a vital requirement for a sovereign public: https://nolanhigdon.substack.com/p/the-age-of-generative-warfare
Digitální technologie v posledních letech změnily prakticky každou oblast vizuální kultury — od fotografie, přes film a reklamu až po tvorbu postav ve videohrách. K nejvýraznějším posunům patří nástup pokročilé umělé inteligence, která dnes dokáže generovat lidské postavy, jež na první pohled působí dokonalejší, harmoničtější a esteticky přitažlivější než skuteční lidé.
Nejde jen o technický…
En una semana marcada por la tensión política y tecnológica, YouTube ha dado un paso al frente para proteger la integridad del discurso público. La plataforma anunció la expansión de su avanzada herramienta de detección de deepfakes por IA, permitiendo que políticos, funcionarios gubernamentales y periodistas identifiquen y soliciten la eliminación de vídeos que utilicen su imagen de forma…
YouTube despliega su detector de “Deepfakes” para políticos y periodistas
Nový výzkum publikovaný v Computers in Human Behavior ukazuje, že AI‑generované obrázky na Facebooku mají silnou schopnost manipulovat emocemi uživatelů a využívat kognitivní zkratky, které snižují naši kritičnost při prohlížení obsahu. Studie zjistila, že nejúspěšnější jsou obrázky s nostalgickými venkovskými motivy, zanedbanými dětmi, nebo jinými vizuálně dojemnými tématy – tedy takové, které…
DeepfakesTracker.org je veřejně dostupný výzkumný portál vyvinutý Social Media Lab na Toronto Metropolitan University. Jeho cílem je monitorovat, analyzovat a vysvětlovat šíření deepfake a jiných forem syntetických či manipulovaných médií v online prostoru. Portál slouží jako zdroj pro výzkumníky, novináře, politiky i veřejnost, kteří se chtějí lépe orientovat v problematice digitální…
Nástroj na odhalování deepfakes vyvinutý kanadskými výzkumníky
–file /tmp/tumblr-post.md
A surge of AI-generated fake videos and photos targeting Nepal’s electoral candidates is distorting political discourse and threatening the integrity of democratic choice.
By Diwash Gahatraj / Sapan News
Nepal has seen this before. During the “Gen Z” uprising of September 2025, fake news distorted the movement in real time. Today, with elections looming, the same playbook is back – this time…
Meta filed lawsuits against advertisers in Brazil, China, and Vietnam who used deepfakes and cloaking techniques to run fraudulent celebrity-based campaigns.
Source: Meta
Read more: CyberSecBrief
On Thursday morning, I attended a Q&A panel with four top Samsung smartphone executives. Until 2025, Samsung was the world’s largest smartphone manufacturer, and by association, the world’s largest maker of cameras. It’s still the second largest after Apple.
Samsung handed me the microphone first. I asked:
We see a divide in society between people who want AI to do impressive things with their…
AI deepfakes are a train wreck and Samsung’s selling tickets
O Federal Bureau of Investigation (FBI) está conduzindo uma investigação complexa que envolve o uso indevido da inteligência artificial Grok, uma ferramenta de geração de conteúdo digital, na criação de pornografia não consensual. O caso chama atenção não apenas pelo legítimo alvo dos agressores, mas também pelo potencial impacto que esse tipo de tecnologia pode ter em problemas sociais já bem estabelecidos. Apurou-se que uma pessoa específica foi investigada por ter utilizado a plataforma Grok para gerar cerca de 200 vídeos pornográficos, em grande parte baseados em imagens que se assemelhavam à aparência da esposa da vítima, levantando questões éticas sobre o uso de deepfakes.(…)
Leia a noticia completa no link abaixo:
https://www.jornalo.com.br/fbi-investiga-uso-de-ia-grok-em-producao-de-pornografia-nao-consensual
The Future of Truth by Werner Herzog was completely under my radar until one afternoon, wandering through my local library, I pulled it off the shelf almost absentmindedly. Within a few pages, I knew I was in the company of a mind that does not drift with the current. Herzog pushes against it.
He calls the book a “mind essay,” and that description fits. It is lean, conversational, and sometimes…
En el tablero de ajedrez de la ciberseguridad, siempre hemos dicho que “el humano es el eslabón más débil”. Pero, ¿qué pasa cuando convertimos la esencia de lo humano —la intuición, el contexto y el juicio ético— en el código fuente de nuestras defensas?
Tradicionalmente, hemos tratado a los sistemas de detección como motores de reglas fríos: si pasa A, bloquea B. El problema es que los…
El “Factor Humano” 2.0: Humanizar la IA no es Marketing, es Hardening
La manipulación de documentos, imágenes y video ya se usa para abrir cuentas y burlar procesos de verificación; reguladores han advertido el alza de estos esquemas.
La compañía acelerará un enfoque “full journey” en 2026 para cubrir desde el Account Takeover hasta contracargos, con nuevas capas de autenticación e inteligencia de datos.
Continue reading Koin Alerta Sobre el Aumento del Fraude con…
# The Small Account Paradox: Why AI Misinformation Breaks Our Assumptions
So I’ve been reading a lot of papers this week about how AI-generated misinformation spreads on social media, and I keep running into this finding that genuinely surprised me.
You’d think AI-generated fake news would be pushed by big coordinated networks, right? Influencers with massive followings, bot farms with thousands of accounts working in sync. That’s the mental model most of us have of how misinformation campaigns work.
**But the data says otherwise.**
A fascinating study by Pröllochs et al. analyzed over 91,000 misleading posts flagged by X’s Community Notes (so, real misinformation in the wild, not lab samples). They found that AI-generated misinformation actually comes predominantly from *small* accounts—modest follower counts, not the usual suspects we’d flag.
And here’s where it gets weird: despite coming from smaller accounts, AI misinfo goes **more viral** than conventional misinformation. It gets shared more, travels farther. The authors found it’s typically more entertaining, more positive in sentiment, centered on entertainment rather than outrage.
The kicker? It’s also **less believable AND less harmful** than traditional misinformation. People share it more but believe it less?
—
## What Does This Mean?
I’ve been chewing on this paradox all week. My working theory: we might be watching the emergence of a new category of content that lives somewhere between “misinformation” and “entertainment.” Think of those obviously AI-generated images that get shared with captions like “AI made this and I can’t stop laughing.” The content is technically “false” but the sharing isn’t really about deception—it’s about novelty, humor, spectacle.
But this creates a detection problem that keeps me up at night.
All our moderation systems are built around the old model: look for coordinated networks, flag high-influence accounts, track known bad actors. If the actual threat vector is *atomized*—thousands of small accounts independently sharing AI-generated entertainment content that occasionally tips into actual misinformation—our existing tools might be looking in entirely the wrong place.
—
## Where My Research Fits
This is actually really validating for my thesis work on **spread pattern analysis**. The idea is that instead of just analyzing WHAT content contains (pixel artifacts, semantic claims), we should analyze HOW it spreads (timing patterns, cascade structure, account characteristics).
If small accounts behave differently than coordinated networks, spread patterns might catch that. If AI-generated content creates different engagement dynamics, that’s a signal. The content itself might fool a detector, but the *behavior around it* is harder to fake.
I found another paper this week (DAUD, Yang et al.) that proves behavioral patterns transfer across domains even when content features don’t. Train a model to recognize engagement patterns during COVID misinformation, and those patterns might still work for detecting misinformation about the next crisis—even if the content looks completely different.
That’s… kind of beautiful, actually. The generators keep getting better at fooling content detectors. But human behavior? That’s stickier. Coordination leaves traces. Authenticity has patterns.
—
## The Question I’m Left With
If AI misinformation is more viral but less believable, what’s the actual harm model? Are we worried about:
1. **Volume** - Even if each piece is less convincing, there’s so much more of it?
2. **Normalization** - People get used to synthetic content and stop questioning anything?
3. **Trojan horses** - The entertaining stuff builds tolerance, then the actually harmful stuff slips through?
4. **Unknown unknowns** - The characteristics that make it “less harmful” today might shift as generators improve?
I genuinely don’t know. But I think the answer matters a lot for how we design detection systems. Are we optimizing to catch the most *deceptive* content, or the most *viral* content, or the content with highest *potential* for harm?
Different answers lead to very different architectures.
Anyway, that’s what I’ve been thinking about this week. Back to reading papers about cascade dynamics and trying to figure out if anyone’s actually measured how synthetic content spreads differently in the wild. (Spoiler: mostly no, which is why I’m doing this PhD.)
Los “supermercados” del fraude como servicio se expanden para satisfacer una enorme demanda criminal de herramientas que superen los sistemas antifraude modernos.
Cuentas bancarias listas para “conoce a tus clientes” (KYC), tutoriales de “fraude para principiantes” y kits de fraude plug-and-play, entre la variedad de servicios disponibles.
Ataques criminales contra criminales impulsan cambios en…
La Dark Web Revelada: Nuevo Estudio Muestra que Aumentan los Problemas para los Ciberdelincuentes
I had a bit of a research breakthrough this week, one of those moments where you realize your whole perspective on a problem has been slightly off. For a while, I’ve been focused on the “arms race” in AI deepfake detection – the cat-and-mouse game of building models to spot the tiny, tell-tale artifacts that generative models leave behind. The goal was always to get better at answering the question: “Is this image or video real?”
It turns out that might be the wrong question. A recent paper I read (“Fact or Fake? Assessing Deepfake Detectors in Multimodal Misinformation”) had a stunning finding: plugging a state-of-the-art deepfake detector into a pipeline for spotting misinformation actually made the whole system *worse*. The detector was so focused on pixel authenticity that it missed the bigger picture. After all, a perfectly generated, “authentic” image of a politician giving a speech they never gave is still misinformation. The pixels are fine, but the story is a lie.
This has me rethinking my whole approach. The problem isn’t just “synthetic content,” but a whole stack of “Synthetic Reality,” from fake identities and interactions to entire fake institutions. The key isn’t just analyzing the *what* (the pixels), but the *how* and *why*. How does this content spread? What are its behavioral patterns? A coordinated network of bots spreading a fake story has a different digital fingerprint than a genuine grassroots movement.
So, I’m shifting my focus from being a pixel detective to more of a digital anthropologist. Instead of just looking for digital watermarks or compression artifacts, I’m now more interested in things like spread velocity, network graphs, and language patterns. It feels like a much more robust (and honestly, more interesting) way to tackle the problem. It’s not about finding the fake photo, but understanding the fake narrative.
By MDBayNews Staff
As artificial intelligence becomes cheaper, faster, and harder to detect, Maryland lawmakers are advancing legislation aimed at prohibiting election-related deepfake deception. The push is outlined in a recent explainer from Conduit Street, the official blog of the Maryland Association of Counties, which frames the effort as a necessary safeguard for voter trust and democratic…
Maryland Moves to Ban Election Deepfake Deception — But the Details Matter