Cynative: Open-source deep research agent 2026-07-13 at 09:00 By Mirko Zorz Running a large language model against a live cloud account to hunt for security holes comes with an obvious hazard. An agent that holds real credentials and a mandate to poke around can delete a bucket, flip a permission, or leak a secret on […]
Cynative: Open-source deep research agent Read More »
A hardware security AI assistant that checks chips for hidden backdoors 2026-07-13 at 08:00 By Sinisa Markovic Chip designers license blocks of circuitry from outside vendors and drop them into larger products. A single processor can carry components from a range of suppliers, each written by a company the buyer may never deal with directly.
A hardware security AI assistant that checks chips for hidden backdoors Read More »
99.9% of fixable AI vulnerabilities remain unpatched 2026-07-13 at 07:30 By Anamarija Pogorelec Organizations build, deploy, and operate AI in the cloud, but basic cybersecurity hygiene is often sacrificed for speed, according to Orca Security’s 2026 State of AI Security Report. Building AI without security Fifty-six percent of AI adopters have deployed agent frameworks into
99.9% of fixable AI vulnerabilities remain unpatched Read More »
AWS gives its ERP agent deny-by-default rules and a separate identity 2026-07-10 at 08:00 By Sinisa Markovic Accounts receivable teams at large companies spend hours each day matching incoming bank payments to invoices by hand. When those payments sit unmatched for days, cash flow suffers and days sales outstanding climbs. The same pattern repeats across
AWS gives its ERP agent deny-by-default rules and a separate identity Read More »
Your coding agent says no in chat and yes in the code 2026-07-09 at 13:44 By Mirko Zorz Millions of developers share their keyboard with GitHub Copilot. Inside Visual Studio Code, it opens their files, writes and edits code, runs scripts, and reworks its own output across many turns. The safety testing that vets these
Your coding agent says no in chat and yes in the code Read More »
Malicious AI agent skills can slip past the scanners built to stop them 2026-07-09 at 10:24 By Sinisa Markovic Developers who build with AI coding agents grab capabilities off public marketplaces the same way they grab packages from npm or PyPI. The add-ons are called agent skills. Each one is a little bundle of plain-English
Malicious AI agent skills can slip past the scanners built to stop them Read More »
Thousands of malicious AI skills found capable of stealing data, running malware 2026-07-08 at 12:00 By Anamarija Pogorelec AI agents can browse the web, use external tools, execute commands, and perform tasks on behalf of users. Many rely on skills that define how they interact with services and data. Malicious skills can abuse those capabilities
Thousands of malicious AI skills found capable of stealing data, running malware Read More »
Claude Cowork turns your phone into a remote control for AI work 2026-07-08 at 10:22 By Anamarija Pogorelec Anthropic started rolling out Claude Cowork, an AI agent that completes multi-step tasks, in beta for Max users on mobile and the web. They describe a goal, and Claude plans the work, uses the required tools, and
Claude Cowork turns your phone into a remote control for AI work Read More »
20 open-source cybersecurity tools to keep your team ready for anything 2026-07-08 at 08:30 By Anamarija Pogorelec AI is changing how security teams find vulnerabilities, analyze code, test applications, and protect infrastructure. Developers are building tools to secure AI systems themselves, from coding agents and memory protection to model exposure discovery. This roundup covers recent
20 open-source cybersecurity tools to keep your team ready for anything Read More »
macOS is becoming a proving ground for AI agents 2026-07-08 at 08:00 By Sinisa Markovic Somewhere right now, a Mac Mini is sitting on a shelf doing someone’s chores. Nobody’s watching it. It reads a version number out of Terminal, hops over to Safari, digs up a release year, then quietly files a reminder, the
macOS is becoming a proving ground for AI agents Read More »
OpenAI and Anthropic are pulling in different directions 2026-07-08 at 07:00 By Mirko Zorz Companies are handing routine operational decisions to AI agents that plan, remember, and act on their behalf. These agents run on statistical models, and their behavior can drift across weeks and months. That drift opens a security gap outside the reach
OpenAI and Anthropic are pulling in different directions Read More »
Researchers make the case for a cybersecurity AI scientist 2026-07-07 at 09:30 By Mirko Zorz Autonomous AI agents have started doing real security work. Language-model agents probe software for flaws, run penetration tests, and chain together attack steps that once needed a human operator. Research about security has stayed slower and more manual, built around
Researchers make the case for a cybersecurity AI scientist Read More »
How to prioritize AI agent security by business impact 2026-07-06 at 09:30 By Help Net Security Your CEO calls about an AI agent security incident in finance. He wants to know whether money moved, whether financial data was exposed, who owned the agent and why it had this level of access. The agent was connected
How to prioritize AI agent security by business impact Read More »
The future of payment fraud could be automated 2026-07-06 at 07:00 By Anamarija Pogorelec Payment fraud is becoming more organized as criminal groups use fake websites, large-scale operations, and, in some cases, forced labor to steal money and personal information. Advances in agentic AI could automate many stages of payment fraud, from collecting and assembling
The future of payment fraud could be automated Read More »
Agentic AI Used to Conduct Ransomware Attack via Langflow 2026-07-03 at 14:00 By Ionut Arghire Attack demonstrates how LLM agents can combine known exploitation techniques with real-time reasoning to automate complex, multi-stage intrusions. The post Agentic AI Used to Conduct Ransomware Attack via Langflow appeared first on SecurityWeek. This article is an excerpt from SecurityWeek
Agentic AI Used to Conduct Ransomware Attack via Langflow Read More »
Claude Sonnet 5 includes safeguards against dangerous cyber use 2026-07-01 at 11:45 By Anamarija Pogorelec Anthropic has introduced Claude Sonnet 5, the latest version of its general-purpose AI model, with improved reasoning, coding, tool use, and knowledge work capabilities. The model can make plans, use tools such as browsers and terminals, and complete tasks autonomously.
Claude Sonnet 5 includes safeguards against dangerous cyber use Read More »
Mozilla warns of indirect prompt injection risk in AI coding agents 2026-06-29 at 13:48 By Zeljka Zorz A malicious GitHub repository can silently compromise a developer’s machine without containing a single line of malicious code, security researchers at Mozilla’s Zero Day Investigative Network (0DIN) warned. The attack The proof-of-concept attack targets AI-powered coding agents such
Mozilla warns of indirect prompt injection risk in AI coding agents Read More »
GPT-5.6 gets better at cybersecurity 2026-06-29 at 12:13 By Anamarija Pogorelec OpenAI has started rolling out the GPT-5.6 series models in limited preview to a small group of trusted partners through the API and Codex. The series includes Sol as the flagship model, Terra as a balanced option, and Luna as the fastest and most
GPT-5.6 gets better at cybersecurity Read More »
DarkMoon: Open-source AI pentesting platform 2026-06-29 at 08:30 By Mirko Zorz Penetration testing has long run on expert time, with specialists spending days probing a network or web application by hand. Manual engagements stretch across weeks, expert consultants run into thousands of dollars a day, and results vary with the tester. Automation promises to narrow
DarkMoon: Open-source AI pentesting platform Read More »
Scoring AI hackers when there is no answer key 2026-06-25 at 08:30 By Mirko Zorz AI models are solving more and more of the offensive-cyber tests built to measure them. Once a model solves most of a benchmark, that benchmark runs out of room and says little about the best systems anymore. Many of those
Scoring AI hackers when there is no answer key Read More »