Contact
Red Hot Cyber

Cybersecurity, Cybercrime News and Vulnerability Analysis

Home
    Anthropic Breach: AI Startup Cracks Systems in Just 12 Hours
    Luigi Zullo | Vulnerability | 26/10/2026

    Cathedral, a startup founded by former employees of the US DOJ, gained access to internal data of the artificial intelligence developer Anthropic in 12 hours. The startup’s artificial intelligence age...

    China Hits the Frontier: Kimi K3.1 Pro Beats All OpenAI Models – Weights Open Nov 4
    Carolina Vivianti | innovation | 26/10/2026

    The Chinese artificial intelligence lab Moonshot AI unveiled Kimi K3.1 Pro — an updated version of the Kimi K3 model, which underwent further training using a larger volume of compute resources. In th...

    AI Ransomware Takes Over: Autonomous AI Powers the Next Attack Wave
    Luigi Zullo | Cybercrime | 25/10/2026

    It was to be expected, and it has arrived. Cybercriminals are moving to a new attack model in which artificial intelligence becomes not only an assistant to the attacker, but a tool to manage almost t...

    Google patches 353 Chrome vulnerabilities, blocks insecure HTTP
    Luigi Zullo | Vulnerability | 25/10/2026

    Google has released Chrome 156 with one of the largest security fix packages ever released and at the same time has changed the rules for visiting unprotected sites. The new browser version immediatel...

    Impossible-to-Apply Patches: How AI Is Redefining Cybersecurity
    Luigi Zullo | innovation | 24/10/2026

    It seems we have returned to the days of EternalBlue or Log4j (to be more recent) – an emergency security update for PHP, the programming language used by roughly 70% of websites worldwide, has trigge...

    DeepSeek Secures $15 B to Challenge OpenAI and Anthropic
    Carolina Vivianti | innovation | 23/10/2026

    The real race in artificial intelligence no longer concerns only who can develop the most powerful model, but especially who will be able to dictate how this technology is used and distributed. On one...

    PHP2Shell Alert: Critical PHP Flaws Expose Web Servers Worldwide
    Luigi Zullo | Vulnerability | 22/10/2026

    The 20 October 2026, the PHP security team published a preliminary notice regarding an imminent emergency release aimed at fixing two critical vulnerabilities (both with a CVSS score of 9.2). The patc...

    Google Unveils Gemini 4 Argon – Matches OpenAI at One‑Fifth the Cost
    Carolina Vivianti | innovation | 02/10/2026

    Google has launched Gemini 4 Argon, which will be rolled out to a group of cybersecurity experts through our Fairwind program. The model is designed to support deep reasoning in complex, long‑term wor...

    ShinyHunters Hacks Clop’s Tor Site, Demands Ransom in Cyber Revenge
    Luigi Zullo | Cybercrime | 22/09/2026

    The ShinyHunters group has hacked the Onion site of the ransomware group Clop (also known as Cl0p). ShinyHunters members claim to have stolen data from the server, including logs, source code and priv...

    From “27 wallets” to “27 attack surfaces”: an analysis of the Revolut and Kraken cases
    Olivia Terragni | Cybercrime | 21/09/2026

    Banks, exchanges and fintechs under pressure through fake requests appearing to come from authorities: the Revolut case, the Kraken precedent, the actors involved, public amplification, and the politi...

    Precedente Successivo
    Anthropic Breach: AI Startup Cracks Systems in Just 12 Hours
    Luigi Zullo 26/10/2026
    China Hits the Frontier: Kimi K3.1 Pro Beats All OpenAI Models – Weights Open Nov 4
    Carolina Vivianti 26/10/2026
    AI Ransomware Takes Over: Autonomous AI Powers the Next Attack Wave
    Luigi Zullo 25/10/2026
    Google patches 353 Chrome vulnerabilities, blocks insecure HTTP
    Luigi Zullo 25/10/2026
    Impossible-to-Apply Patches: How AI Is Redefining Cybersecurity
    Luigi Zullo 24/10/2026

    Latest news

    Anthropic Restores Claude Mythos 5 for Critical US Organizations Cyberpolitics

    Anthropic Restores Claude Mythos 5 for Critical US Organizations

    Anthropic has confirmed that Claude Mythos, its most powerful artificial intelligence model for cybersecurity, will be redeployed to a select...
    Carolina Vivianti - June 27, 2026
    China responds to the Anthropic Mythos. Qihoo 360 presents “Tulongfeng,” a swarm-based device. innovation

    China responds to the Anthropic Mythos. Qihoo 360 presents “Tulongfeng,” a swarm-based device.

    Qihoo 360 has presented an artificial intelligence-based vulnerability scanning system, which the company considers China's response to Anthropic's Mythos. Qihoo...
    Carolina Vivianti - June 26, 2026
    FortiBleed: 320,000 Fortinet Firewalls Exposed Cybercrime

    FortiBleed: 320,000 Fortinet Firewalls Exposed

    A massive campaign involving Fortinet devices exposed on the Internet is attracting much attention from the international cybersecurity community. The...
    Carolina Vivianti - June 17, 2026
    Anthropic Negotiates to Unblock Mythos 5 and Fable 5 Models Cyberpolitics

    Anthropic Negotiates to Unblock Mythos 5 and Fable 5 Models

    As expected, Anthropic's response to the blockade of the Mythos 5 and Fable 5 AI models did not hesitate to...
    Luigi Zullo - June 16, 2026
    Anthropic Launches Claude Fable 5: Is the Mythos Class Model Jailbreak-Proof? innovation

    Anthropic Launches Claude Fable 5: Is the Mythos Class Model Jailbreak-Proof?

    Claude Fable 5 and Claude Mythos 5 are the new artificial intelligence models launched today by Anthropic, with unprecedented capabilities...
    Luigi Zullo - June 10, 2026
    C0XMO: Malware Infects Servers and Eliminates Other Malware Cybercrime

    C0XMO: Malware Infects Servers and Eliminates Other Malware

    Researchers at Fortinet have warned of a new variant of the Gafgyt botnet, dubbed C0XMO. The malware targets routers with...
    Carolina Vivianti - June 9, 2026
    Linux Under Fire: Just 4 Bytes Are Enough for Root Access Vulnerability

    Linux Under Fire: Just 4 Bytes Are Enough for Root Access

    A newly discovered vulnerability, tracked as CVE-2026-31431 and nicknamed Copy Fail, stands out as one of the most insidious Linux...
    Bajram Zeqiri - April 30, 2026
    Forget Alan Parsons! Apple Music and Deezer have become dumping grounds for AI-generated music innovation

    Forget Alan Parsons! Apple Music and Deezer have become dumping grounds for AI-generated music

    For those who love the music of Pink Floyd, they surely know Alan Parsons, the legendary sound technician who mixed...
    Massimiliano Brolli - April 28, 2026
    Diary of a Threat Actor: from Ghost of the Web to Unmasked Man Culture

    Diary of a Threat Actor: from Ghost of the Web to Unmasked Man

    Twenty-five years old, a piece of malware to remotely control dozens of machines, and a botnet as a final act...
    Olivia Terragni - April 27, 2026
    FABRICKED: AMD Cloud Security Flaw Exposed, Is Your Data Safe? Hacking

    FABRICKED: AMD Cloud Security Flaw Exposed, Is Your Data Safe?

    Cloud servers that promise complete data protection are not as impenetrable as they seem. Researchers have found a way to...
    Bajram Zeqiri - April 18, 2026
    Frontier AI for Defenders: CrowdStrike and OpenAI TAC innovation

    Frontier AI for Defenders: CrowdStrike and OpenAI TAC

    CrowdStrike has been selected for OpenAI'sTrusted Access for Cyber (TAC) program. Today, OpenAI releasedGPT-5.4-Cyber, a frontier model designed for defensive...
    Carolina Vivianti - April 17, 2026
    AI and Security: Is the Mythos Model a Breakthrough or Just Hype? Marcus Hutchins’s Vision Culture

    AI and Security: Is the Mythos Model a Breakthrough or Just Hype? Marcus Hutchins’s Vision

    With the boom in artificial intelligence, bold claims about breakthroughs in cybersecurity are becoming increasingly common, but not all experts...
    Carolina Vivianti - April 16, 2026
    « Previous   Next »
    L'immagine mostra un tratto della Circonvallazione del Foro Italico a Roma, ripreso da una prospettiva stradale tipica dei servizi di mappe virtuali, come confermato dalla grande scritta bianca digitale impressa verticalmente sull'asfalto in primo piano. L'elemento visivo più straordinario dell'inquadratura è l'intricato motivo geometrico di luci e ombre che avvolge l'intera carreggiata, generato dal sole che filtra attraverso una maestosa tettoia frangisole metallica posizionata in alto. Il traffico scorre in un'unica direzione, guidato in primo piano da una Smart scura, seguita a distanza da una fila di altre vetture tra cui spicca un'auto rossa. A delimitare la carreggiata sulla sinistra si trovano i guardrail metallici e un cartello di divieto di transito, mentre il lato destro è dominato da un'alta barriera antirumore interamente rivestita da una fitta vegetazione rampicante. Sullo sfondo, la strada prosegue superando il tratto coperto e aprendosi verso una zona collinare e alberata molto più luminosa.

    Welcome to Italy, Autopilot: where shadows on the tarmac outsmart artificial intelligence

    When Rome’s tarmac sends Autopilot into a tailspin: the case of Nello Coppeto on the Giovanni XXIII Tunnel Today I’d like to share a story that happened to one of...

    - July 12th, 2026

    Share on Facebook Share on LinkedIn Share on X

    Microsoft 365 in the Crosshairs: $2.5 Million Zero-Day Exploit Up for Auction

    A staggering figure published in an announcement on a well-known underground forum is attracting the attention of the cyber security community. A user with the nickname Orcinus orca claims to...

    - July 10th, 2026

    Share on Facebook Share on LinkedIn Share on X
    Primo piano cinematografico di un tablet con lo schermo in vetro infranto, appoggiato su un pavimento di cemento scuro, ruvido e crepato. Al centro del display compare la scritta “ANTHROPIC”, attraversata da un punto d’impatto da cui si diramano numerose crepe sottili e taglienti. I frammenti riflettono una luce intensa proveniente dall’alto, creando forti contrasti e un’atmosfera cupa e drammatica. Intorno al dispositivo sono visibili piccoli detriti e polvere. L’immagine ha uno stile iperrealistico, con texture molto definite, evocando un’estetica cyber-archeologica decadente e tecnologica.

    Fable 5 Hacked Again After 48 Hours: National Security Alert?

    According to a series of online rumors, the “Fable 5” model would have been hacked again. This time, it happened just 2 days after its release (the second one), following...

    - July 4th, 2026

    Share on Facebook Share on LinkedIn Share on X
    Laptop con la scritta Z.AI davanti alla bandiera cinese, circondato da una città futuristica e ologrammi illuminati di colore rosso.

    GLM-5.2 Arrives: China’s Open-Source Answer to Mythos Runs on Your PC

    We knew it would end like this, even if the question was: how long? Some said six months, others a year. But perhaps we were all wrong after Zhipu AI...

    - June 30th, 2026

    Share on Facebook Share on LinkedIn Share on X
    point 16:01 L'immagine mostra un tablet posizionato verticalmente su una superficie scura e riflettente, probabilmente un tavolo lucido. Lo schermo del dispositivo ha uno sfondo bianco crema su cui campeggia, in caratteri neri, maiuscoli e piuttosto spessi, la scritta "ANTHROPIC", il nome dell'azienda di intelligenza artificiale. Il tablet ha cornici sottili di colore scuro e un design moderno, con bordi arrotondati che richiamano i tipici dispositivi di fascia alta. Sullo sfondo, sfocata per via di un evidente effetto di profondità di campo, si vede la bandiera degli Stati Uniti d'America. Si distinguono chiaramente il campo blu con le stelle bianche nell'angolo in alto a sinistra e le caratteristiche strisce orizzontali rosse e bianche che occupano la parte restante. La bandiera appare leggermente ondulata, come mossa dal vento, e i suoi colori sono ben saturi nonostante la sfocatura. L'illuminazione è morbida e diffusa, e l'inquadratura mette in primo piano il tablet nitido contro lo sfondo patriottico fuori fuoco. La composizione complessiva suggerisce un'associazione tra l'azienda Anthropic e gli Stati Uniti, un tipo di immagine spesso usata in contesti editoriali o giornalistici per articoli che trattano di tecnologia, politica industriale o sviluppi del settore dell'intelligenza artificiale americano.

    Anthropic Restores Claude Mythos 5 for Critical US Organizations

    Anthropic has confirmed that Claude Mythos, its most powerful artificial intelligence model for cybersecurity, will be redeployed to a select group of US organizations responsible for operating and defending critical...

    - June 27th, 2026

    Share on Facebook Share on LinkedIn Share on X

    China responds to the Anthropic Mythos. Qihoo 360 presents “Tulongfeng,” a swarm-based device.

    Qihoo 360 has presented an artificial intelligence-based vulnerability scanning system, which the company considers China's response to Anthropic's Mythos. Qihoo 360's CEO, Zhou Hongyi, presented the novelty at the 14th...

    - June 26th, 2026

    Share on Facebook Share on LinkedIn Share on X

    FortiBleed: 320,000 Fortinet Firewalls Exposed

    A massive campaign involving Fortinet devices exposed on the Internet is attracting much attention from the international cybersecurity community. The operation, dubbed FortiBleed, appears to have targeted around 320,000 FortiGate...

    - June 17th, 2026

    Share on Facebook Share on LinkedIn Share on X
    L'immagine rappresenta la dea del diritto in posizione centrale, con una bandiera degli Stati Uniti d'America che si estende a sinistra e a destra. A destra della bandiera si vede un tablet che mostra il testo 'anthropic', mentre a sinistra si trova una bilancia. Il contesto indica che Anthropic sta iniziando le negoziazioni con gli amministratori americani per sbloccare i modelli Mythos 5 e F.5, riferendosi a un processo di liberazione di modelli linguistici avanzati. Il quadro è realizzato in stile simbolico, con elementi che rappresentano la legge, la tecnologia e l'accesso ai modelli di intelligenza artificiale.

    Anthropic Negotiates to Unblock Mythos 5 and Fable 5 Models

    As expected, Anthropic's response to the blockade of the Mythos 5 and Fable 5 AI models did not hesitate to arrive. Already on Monday, the company met with representatives of...

    - June 16th, 2026

    Share on Facebook Share on LinkedIn Share on X
    Claude Fable 5 e Claude Mythos 5, due nuovi modelli di intelligenza artificiale lanciati da Anthropic, offrono capacità senza precedenti nel settore della cybersecurity. Fable 5, appartenente alla classe Mythos, è reso sicuro per l'uso generale e supera tutti gli altri modelli disponibili al pubblico in termini di performance su benchmark di capacità AI. Eccelle in software engineering, knowledge work, visione, ricerca scientifica e altre aree, con un vantaggio crescente man mano che il compito diventa più lungo e complesso. Questi modelli rappresentano un passo avanti significativo per l'industria della cybersecurity e l'elaborazione di compiti avanzati.

    Anthropic Launches Claude Fable 5: Is the Mythos Class Model Jailbreak-Proof?

    Claude Fable 5 and Claude Mythos 5 are the new artificial intelligence models launched today by Anthropic, with unprecedented capabilities in the cybersecurity sector. Fable 5 is a model of...

    - June 10th, 2026

    Share on Facebook Share on LinkedIn Share on X
    Un robottino bianco in stile cyberpunk, con cingoli neri e occhi rossi e arrabbiati, si trova in una sala server buia. Dal suo braccio-cannone spara un raggio laser rosso che distrugge un rack di server, facendolo esplodere in mille pezzi con fumo e scintille. Un display sul server colpito mostra un teschio e la scritta "SYSTEM COMPROMISED", mentre un elenco a destra segnala che i server da 07 a 10 sono offline. In alto a sinistra, un'interfaccia olografica rossa intitolata "BOT.EXE" mostra lo stato "FURIOUS" e la modalità "DESTRUCTION".

    C0XMO: Malware Infects Servers and Eliminates Other Malware

    Researchers at Fortinet have warned of a new variant of the Gafgyt botnet, dubbed C0XMO. The malware targets routers with DD-WRT firmware, but researchers have also identified versions for ARM,...

    - June 9th, 2026

    Share on Facebook Share on LinkedIn Share on X

    Discover the latest critical CVEs issued and stay updated on the most recent vulnerabilities. Or search for a specific CVE