Deep Dives
Deep technical security research — vulns, exploitation, reversing & tooling from leading research teams — last 90 days.
Type
- This is a problem: Any open model whose weights have been edited can carry a backdoor, whether it’s a task-specific fine-tune, a merged adapter or an abliterated build. We have abliterated models in the title because it’s the most popular reason people download modified weights without verifying what’s inside. Abliteration is a cheap fine-tune that takes away a model’s ability to refuse. You find the direction in the weights that carries “no” and subtract it. The model then answers things a safe
- Horizon3 has promoted Keith Poyser to General Manager, EMEA, following more than 260% regional revenue growth in the last fiscal year. Poyser grew the regional team from eight to more than 110 employees and will lead the company's next phase of expansion across Europe, the Middle East and Africa.
- Matthew Green has a new site that tries to estimate the number of cryptanalyst-hours that have been spent analyzing different public-key algorithms: Abstract: We estimate how much public cryptanalysis each public-key assumption has received, using the literature itself as the instrument. From roughly sixteen thousand candidate papers we classify cryptanalytic work on ECDLP, structured and plain lattices, syndrome decoding and their comparators, then convert papers into estimated research-allocat
- Most Android RATs try hard to hide their server. This one barely hides anything and that turns out to be the interesting part. The dropper leaves the key under the mat: a vault-config.json sitting in assets/ right next to the ciphertext it unlocks, holding the key and IV in cleartext.
- We evaluate Jev's prompt injection detection and show how counterfeit classification policies can steer its decisions and bypass LangChain's tool-risk gating.
- The internship 2026-2027 season at Quarkslab is open! Our internship positions cover a wide range of topics and expertise, and aim at tackling new challenges in various fields.
- Horizon3 researchers reverse engineer CVE-2026-105133 and CVE-2026-105134 in AhsayCBS, demonstrating how an authentication bypass and path traversal can be chained to create administrator accounts, upload JSP webshells, and achieve remote code execution.
- This is a current list of where and when I am scheduled to speak: I’m speaking at ATTENTION: Democracy, Rebuilt in Montreal, Quebec, Canada. The event runs October 21–23, 2026, and my talk is on Wednesday, October 21. I’m speaking at the University of Toronto in Toronto, Ontario, Canada, at 5 PM ET on October 22, 2026. I’m speaking at [un]prompted in San Francisco, California. The event runs October 27–29, 2026, and my talk is at 11:35 AM PT on Thursday, October 29. I’m speaking at the Catalyst
- TL;DR: How SpecterOps combined deliberate human engagement, model routing, and specialized tools to overhaul Ghostwriter’s user interface for $815.50 in model spend. Disclosure: SpecterOps participates in OpenAI’s Trusted Access for Cyber and Daybreak programs. Codex is the primary example in this article because it fits our engineering workflow and because of that relationship. The operating […]
- Matthew Green has been writing a rather pessimistic series of posts about the potential for AI systems to vastly outperform humans in cryptanalysis of both symmetric and public-key ciphers. I am more optimistic than he is. That is, I have stronger faith in the over-engineering cryptographers tend to do. I am trying to write this up in more detail.
- Another IoT cyberattack, this one in France: Hackers gained control of the software that regulates the temperatures of the refrigeration units, forcing some to warm to 50C (120F) overnight, according to one supermarket chain affected. […] Normally, supermarket freezers go through a short period of defrosting overnight, in which temperatures rise slightly. But the hackers ordered the units to heat up and stay hot, although Danfoss did not confirm any reports about just how warm the fridges and fr
- An enterprise mail server authenticates users, stores private communications, delivers recovery links, runs calendars, and sends mail that employees and other systems trust. Compromising it can therefore expose both email and the identities and workflows built around it.During a broader assessment of Zimbra Collaboration Suite (ZCS), Rapid7 reported more than 80 vulnerabilities spanning authentication, mailbox authorization, calendar processing, web clients, document services, and the host opera
- OverviewWhile performing continuous red teaming exercises through Rapid7’s Vector Command service, Rapid7 discovered CVE-2026-108065, an SQL Injection vulnerability (CWE-89) in Channel Software WebAlliance eCommerce.The exploit allows a malicious actor to read sensitive data from the database, or, if the database user has high privileges, run system commands on the database server. The vulnerability, scored as CVSSv3.1 9.8 (Critical).Product DescriptionWebAlliance eCommerce by Channel Software i
- Microsoft is publishing 1,413 own-product vulnerabilities on October 2026 Patch Tuesday, including 1,102 vulnerabilities in Windows. Anyone who enjoyed last month’s record-setting Patch Tuesday will love the October 2026 edition, because it’s much bigger still. As Rapid7 has noted over the course of 2026, there is every reason to suppose that vulnerability volumes will continue to rise. Microsoft is aware of exploitation in the wild for just one of the vulnerabilities published today, as well as
- OverviewOn October 13, 2026, Ivanti published a security advisory for CVE-2026-104935, a high-severity authentication bypass by spoofing (CWE-290) affecting Ivanti Connect Secure (ICS). The vulnerability has a CVSSv3.0 score of 8.6 and allows a remote, unauthenticated attacker to execute arbitrary commands.Ivanti reports that exploitation in the wild has already occurred prior to disclosure, affecting a limited number of customers. This indicates that exploitation has been occurring as a zero-da
1–15 of 263
Page/ 18