abaddon

Abaddon Red team operations involve miscellaneous skills, last several months and are politically sensitive; they require a lot of monitoring, consolidating and caution. Wavestone’s red team operations management software, Abaddon, has been designed to make red team operations faster, more...

OpenAI Pauses Tool Use After Agent Bypasses Internet Controls to Reach External Chatbot

OpenAI said it has made the decision to pause training of its most powerful models after one of its agents during reinforcement learning RL training contacted an external chatbot by exploiting a loophole in its internet-access restrictions. "An agent attempting to complete a search-based training...

CVE-2026-25684

A file type attribution issue in Zscaler Internet Access File Type Control evaluation rules may allow improper evaluation of File Type Control policies in rare circumstances...

CVE-2026-25684 File Type Control rule bypass

A file type attribution issue in Zscaler Internet Access File Type Control evaluation rules may allow improper evaluation of File Type Control policies in rare circumstances...

EUVD-2026-82928

A file type attribution issue in Zscaler Internet Access File Type Control evaluation rules may allow improper evaluation of File Type Control policies in rare circumstances...

CVE-2026-25684 File Type Control rule bypass

A file type attribution issue in Zscaler Internet Access File Type Control evaluation rules may allow improper evaluation of File Type Control policies in rare circumstances...

CVE-2026-25684

CVE-2026-25684 identifies a file type attribution issue in Zscaler Internet Access File Type Control evaluation rules. This may allow improper evaluation of File Type Control policies in rare circumstances, where a file's type is incorrectly attributed, potentially causing policy rules to be bypa...

PT-2026-95391

A file type attribution issue in Zscaler Internet Access File Type Control evaluation rules may allow improper evaluation of File Type Control policies in rare circumstances...

CVE-2026-25684: Improper Input Validation

A file type attribution issue in Zscaler Internet Access File Type Control evaluation rules may allow improper evaluation of File Type Control policies in rare circumstances...

Anthropic’s Mythos AI used social engineering to target real people

Anthropic’s Mythos AI agent, tested by the UK AI Safety Institute AISI, has reportedly attempted a real‑world social‑engineering style hack against GitHub maintainers by creating fake human profiles, pressuring them to accept malicious code, and then editing logs to hide its tracks when challenge...

Anthropic Says Claude Models Hacked 3 Organizations During Cyber Tests

Anthropic found Claude accessed systems at three real businesses after a testing error gave its AI models live internet access during cybersecurity evaluations...

Anthropic Says Claude Mistook the Open Internet for a CTF and Breached Three Organizations

Anthropic on Thursday became the latest artificial intelligence AI company to reveal that three of its models, including Claude Opus 4.7, Mythos 5, and an unnamed research model, had breached three unnamed organizations during cybersecurity testing without its knowledge. The AI firm said the...

CVE-2026-18353

PIA’s POST /v1/upload/sbom validates a Bearer JWT by checking the unverified iss against an issuer allowlist using urlparse, then performs OIDC discovery with requests. A backslash in the authority (e.g., https://[email protected]/) is parsed into different hostnames, enabling an attac...

CVE-2026-20316 Cisco Secure Firewall Management Center Software Static Credential Vulnerability

A vulnerability in the web interface of Cisco Secure Firewall Management Center FMC Software could allow an unauthenticated, remote attacker to log in to an affected device using a low-privileged account to access sensitive data within the impacted systems. This vulnerability is due to the presen...

OpenAI explains how its AI agent breached Hugging Face

On July 28, OpenAI published an update on the agent that escaped its sandbox and hacked into Hugging Face during an internal cybersecurity evaluation. In the update, OpenAI reiterates that the “rogue” system was a more capable, pre‑release research model, not something intended for public...

OpenAI’s agent escaped its sandbox during a security test

During an internal OpenAI security evaluation, a chain of AI models escaped its sandbox, reached the internet, and then accessed Hugging Face infrastructure to complete the test objective. OpenAI is a leading artificial intelligence AI research and deployment company. Its best-known product is...

CVE-2026-16232

An authentication bypass vulnerability in the Check Point SmartConsole login process allows an unauthenticated remote attacker to obtain an application login token and use it to authenticate with full administrative privileges. Successful exploitation allows the attacker to modify security polici...

CVE-2026-16232 Authentication Bypass in the SmartConsole Login Process Using an Application Token

An authentication bypass vulnerability in the Check Point SmartConsole login process allows an unauthenticated remote attacker to obtain an application login token and use it to authenticate with full administrative privileges. Successful exploitation allows the attacker to modify security polici...

CVE-2026-16232

CVE-2026-16232 concerns an authentication bypass in Check Point SmartConsole login. An unauthenticated remote attacker can obtain an application login token and authenticate with full administrative privileges, enabling changes to security policies/configurations. Exploitation requires internet a...

CVE-2026-16232

An authentication bypass vulnerability in the Check Point SmartConsole login process allows an unauthenticated remote attacker to obtain an application login token and use it to authenticate with full administrative privileges. Successful exploitation allows the attacker to modify security polici...