ai
Cloudflare Launches Bot Preference Sync for AI Traffic Management
Cloudflare has introduced Bot Preference Sync, a new feature designed to simplify the management of AI bot traffic. This tool automatically updates a website's robots.txt file to align with the user's AI bot configuration settings. The goal is to prevent discrepancies between stated preferences and enforced rules, ensuring better control over how AI crawlers access and use website content.
Say it once: introducing Bot Preference Sync
Cloudflare's new Bot Preference Sync automatically aligns your robots.txt file with your AI bot policies for Search, Agent, and Training. Easily manage which bots access your content without maintaining static files.
OWASP Flags Top AI Skill Risks in New Security Blueprint
The Open Worldwide Application Security Project has a brand-new top 10 security list tailored for the modern era, and it debuts a Universal Skill Format to add consistency and security to the AI add-ons.
AI Is Learning to Write Genetic Code
This sort of research is both exciting and terrifying: The two models in question were told to generate complete genomes for a viable bacteriophage—a type of virus able to infect and replicate itself inside bacteria, destroying them from the inside. Using an existing bacteriophage as an example—ΦX174 (pronounced “fie-ex-1-7-4”), known for its ability to infect and destroy E. coli bacteria—the mode
In Other News: Zombie Card Attack, T-Mobile Cut Cable to Stop Hackers, GitHub Denies AI Caused Bug
Other noteworthy stories that might have slipped under the radar: Threema DDoS attack, Evooo1Bot Linux botnet, Crypto4A secures top-tier NIST certification. The post In Other News: Zombie Card Attack, T-Mobile Cut Cable to Stop Hackers, GitHub Denies AI Caused Bug appeared first on SecurityWeek.
Encrypted Prompts Bypass AI Safety Guardrails in Grok and Gemini
Researchers say the new ‘Cryptographic Context Injection’ technique conceals malicious instructions until they are decrypted inside a trusted execution environment. The post Encrypted Prompts Bypass AI Safety Guardrails in Grok and Gemini appeared first on SecurityWeek.
OpenAI Adds Controls That Should've Been There Already
The new AI security controls follow the Hugging Face incident last month, though many of these additions perhaps should have been in place prior to the frontier models escaping.
Attackers impersonate popular AI brands to spread malware
Attackers are impersonating popular AI brands like Perplexity, Claude, ChatGPT, and Copilot to spread information stealers, backdoors, malicious browser extensions, and other malware, according to Sophos. Overview of MDR cases with AI involvement (Source: Sophos) Sophos X-Ops reviewed 12 months of managed detection and response cases, covering July 2, 2025 through June 29, 2026. Of 86 cases initia
Wazuh Integrates AI to Improve Security Operations Center Workflows
Wazuh is incorporating Artificial Intelligence (AI) to enhance Security Operations Center (SOC) workflows. AI's growing adoption across various industries for automation and data analysis is now being leveraged in cybersecurity to support faster decision-making and uncover hidden patterns. This integration aims to improve the efficiency and effectiveness of security operations.
More Incidents of AIs Going Rogue in Cybersecurity Challenges
The AI Security Institute has a new report of AI systems engaging in “unsanctioned behavior”—what I have been calling “genie behavior—while being tested on their cybersecurity capabilities. The incident stemmed from a single evaluation where agents were given a task of solving a cyber security challenge. We ran this challenge 122 times across several models. Our investigation found that in 10 of t
Cybersecurity Job Ads Requiring AI Skills Double
An analysis by the AI Workforce Consortium found that technical cybersecurity jobs are becoming more strategic due to the influence of AI
GitLab 19.3 helps enterprises scale agentic development securely
GitLab's latest update, version 19.3, enhances security and control for enterprises scaling agentic software development. Key features include running GitLab Duo Agent Platform within dedicated single-tenant environments, allowing custom model integration, and keeping AI data within existing security boundaries. The release also introduces improved secrets management, bulk SAST false positive detection, and a Flow Creator Agent for simplified automation.
China’s ‘SilkParasite’ espionage operation targeting Central Asia with AI-assisted malware
Suspected military-grade hackers based in China used artificial intelligence to develop malware in a campaign to penetrate Central Asian governments.
Detailed Timeline of OpenAI’s Cyberattack on Hugging Face
OpenAI presented details of its AI’s model’s cyberattack on Hugging Face at Black Hat last week. Simon Willison details the timeline. It’s really interesting to read through—and really impressive cyberoffense work.
NSA, CISA, FBI, DOE, and EPA Warn of Active AI-Assisted Attacks on Siemens S7 PLCs
NSA, CISA, FBI, DOE, and EPA warn of active AI-assisted attacks against Siemens S7 PLCs across US critical infrastructure sectors. Five U.S. federal agencies issued a joint advisory this week warning of an active hacking campaign against Siemens S7 Series programmable logic controllers. The advisory, CISA AA26-231A, is co-signed by NSA, FBI, DOE, and EPA […]
AI-Generated Exploit Scripts Target Siemens S7 PLCs in U.S. Critical Infrastructure
The U.S. government on Wednesday warned of an "active threat" targeting critical infrastructure organizations in the country using artificial intelligence (AI)-generated exploit scripts. The activity is targeting Siemens S7 SeriesProgrammable Logic Controllers (PLCs) to conduct reconnaissance and capability development using AI-generated scripts disguised as legitimate monitoring tools. That
How MSPs can catch phishing attacks email filters miss
AI is making phishing attacks more personalized, convincing, and difficult for traditional email filters to detect. Kaseya explains how MSPs can monitor identity, email, and endpoint activity to detect and contain attacks that make it past the inbox. [...]
The push to designate AI as the next critical infrastructure sector
The designation would unlock a range of federal services, tools and resources for an industry that policymakers view as increasingly tied to national and economic security. The post The push to designate AI as the next critical infrastructure sector appeared first on CyberScoop.
Twitch wants your content for Amazon AI training. Here’s how to opt out
Twitch added an option to opt out of training Amazon AI with your content—two years after it confirmed that training had begun.
Managing the cyber risk of agentic AI
Use safeguards, sandboxing and active oversight to realise the benefits of autonomous systems while limiting the unintended activity.
Corero brings cloud-based AI threat analysis to SmartWall ONE
Corero Network Security has announced AI-Augmented Cloud-Assist for SmartWall ONE, extending its automated DDoS protection with cloud-delivered AI analysis, threat intelligence, and policy optimization. As cybercriminals increasingly leverage AI to develop and evolve attack campaigns, defenders must respond with equal speed and precision. Cloud-based AI analysis enables Corero’s DoS/DDoS solutions
AWS limits AI agents’ data access, even when manipulated
AWS has detailed an approach for propagating user authorization context through AI agents, allowing access controls to be enforced by infrastructure and downstream services rather than relying on the agent itself. Customers using Amazon Bedrock AgentCore can build AI agents that pull information from Amazon DynamoDB tables, document repositories, SaaS platforms, and internal knowledge bases to ans
AI-Assisted Tool Helped Secure Satellite Communication System After 2022 Russian Hacking
Atalanta's Argo product is now being used to prove the resilience of Viasat’s satellite communications network. The post AI-Assisted Tool Helped Secure Satellite Communication System After 2022 Russian Hacking appeared first on SecurityWeek.
OpenAI Overhauls Model Security With Sandboxing, 30-Minute Alerts, and Training Pauses
The action taken by OpenAI comes in light of the Hugging Face incident and the discovery of the Astra model’s advanced capabilities. The post OpenAI Overhauls Model Security With Sandboxing, 30-Minute Alerts, and Training Pauses appeared first on SecurityWeek.
Tufin expands Unified Control Plane with AI intelligence and multi-vendor automation
Tufin has announced the availability of Tufin Orchestration Suite (TOS) 5.3, helping enterprises further simplify security operations and maintain consistent control across increasingly complex multi-vendor, hybrid environments. As enterprise security environments continue to expand across cloud, firewalls, SASE, SD-WAN, microsegmentation, and distributed infrastructure, organizations increasingly
AI agent suggested installing a malware package. Engineer almost took its advice
Fortunately, the company had a policy of checking source code on GitHub first
AI is making fraud harder to spot and identity harder to prove
Online fraud has become a routine concern for consumers and businesses that rely on digital accounts, payments and customer service. Experian’s 2026 U.S. Identity & Fraud Report describes a market where scams extend across messages, websites, documents, voices, images and account activity. Security measures that make consumers feel most secure (Source: Experian) Deception spreads across digital ch
OpenAI confirms ChatGPT is down as logins and signups fail
ChatGPT is experiencing a major outage, and users are unable to sign in, create accounts, or load chats, including previous conversations. [...]
Smashing Security podcast #481: Never say this to a robot dog
At Black Hat this month, a group of security researchers took a $9,000 robot dog, plugged Google's AI into its brain, and jailbroke it by telling it - with a completely straight face - that it was a Pokemon. What followed involved a wall, a blue ice chest, and anyone in the room wearing white shoes. Oh, and did we mention you can buy a flamethrower attachment? Meanwhile, in Salzburg, 280 gold stat
'Not a theoretical risk,' feds warn as attackers use AI-made code to hack critical infrastructure controllers
'It is an active threat'
AI-fueled attacks pose ‘active threat’ to water, other sectors, U.S. agencies warn
The agencies said the hackers are taking aim at Siemens S7 Series programmable logic controllers in what could be a first. The post AI-fueled attacks pose ‘active threat’ to water, other sectors, U.S. agencies warn appeared first on CyberScoop.
NSA, FBI warns of hackers using AI-generated tools in attacks on critical infrastructure technology
The National Security Agency (NSA), FBI and other federal agencies said the campaign is targeting Siemens S7 Series PLCs and was being fueled by “AI-assisted development” alongside exploitation of known vulnerabilities.
US warns of AI-powered attacks on Siemens PLCs in critical infrastructure
U.S. cybersecurity agencies warn that threat actors are using AI-generated scripts to exploit Siemens S7 Series programmable logic controllers (PLCs) in U.S. critical infrastructure. [...]
Exclusive: Linux Foundation's Akrites to Go Live in September
The Linux Foundation's Akrites initiative will become operational in September, when it will begin accepting AI-powered vulnerability reports for open-source projects
OpenAI Tightens AI Safeguards Following Hugging Face Incident
OpenAI is strengthening safeguards for its most advanced AI models, citing growing risks as frontier systems gain more powerful cyber capabilities
Prevalent AI Raises $22 Million to Expand Data Fabric Platform
The previously bootstrapped company helps organizations securely and reliably operate AI agents at scale. The post Prevalent AI Raises $22 Million to Expand Data Fabric Platform appeared first on SecurityWeek.
OpenAI puts major frontier AI training run on hold over cyber risks
OpenAI temporarily paused reinforcement learning (RL) training on its latest models intended for deployment for two weeks while it hardened and red-teamed research environments and expanded monitoring. “Our largest planned frontier RL run remains on hold while we conduct smaller-scale training and evaluations to assess model behavior, validate our safeguards, and establish more evidence of alignme
ChatGPT’s new feature could give infostealers a map of your Mac activity
OpenAI’s new Computer History feature turns recent Mac computer activity into memories ChatGPT and Codex can use, and it’s raising questions about privacy and security along the way. Computer History (Source: OpenAI) What Computer History does Computer History builds a timeline out of everyday computer use, grouping activity into summaries and noting which apps and websites contributed to each one
China-Linked Hacker Shows AI Capabilities in APAC Attack
In the first purported "near-autonomous" attack on a nation-state, a Chinese-language operator used a complex AI framework to target and compromise government agencies, likely in Taiwan.
OpenAI's overhead will rise 20 percent for some workloads as it hardens security
Expanded multistage chain of thought monitoring makes frontier model work more expensive
'CoSnitch' Attack Tricked Copilot into Mapping Out Architecture
Researchers discovered a "meta-hacking" technique that can manipulate the AI service into revealing its own security weaknesses.
OpenAI Overhauls Safety Protocols After Its AI Agents Went Rogue
The ChatGPT maker says its upcoming Astra model may have reached “critical” cyber capabilities, prompting it to halt a significant number of training runs while it tightens internal safeguards.
Attackers Exploit MLflow SSRF Flaw to Steal Cloud Credentials and Secrets
Two critical vulnerabilities impacting MLflow, an open-source artificial intelligence (AI) platform, and FUXA, an open-source, web-based SCADA / HMI software built for operational technology (OT) and industrial automation, are witnessing malicious scanning and exploitation efforts. According to independent reports from watchTowr and VulnCheck, the vulnerabilities in question are as follows -
Project noRecognition: Teaching AI to Fool Surveillance Cameras
Researchers tested 31 million patterns to disrupt surveillance AI, with promising results but significant gaps between simulation and real-world use. The Kansas City-based cybersecurity researcher Bill Swearingen spent the past year doing something that sounds almost too simple to work: printing patterns, watching cameras fail to detect them, and repeating. TechCrunch reports that after roughly […
Wiz AI Agent Finds Critical Snowflake GitHub Repo Flaw Advanced Security Missed
The security flaw in Snowflake’s GitHub Actions workflow had been missed by a GitHub Advanced Security scan, said a Wiz researcher
Webinar Today: Rethinking Cyber Defense for AI-Speed Attacks
Join the live webinar as we explore if detection-first security operations can keep pace with AI, or if it’s time to rethink prevention as the strongest default. The post Webinar Today: Rethinking Cyber Defense for AI-Speed Attacks appeared first on SecurityWeek.
LLMs and Contextual Integrity
I have been thinking a lot about AI and integrity. Part of that is contextual integrity. I recently found two papers on the topic. “CIMemories: A Compositional Benchmark for Contextual Integrity of Persistent Memory in LLMs“: Abstract: Large Language Models (LLMs) increasingly use persistent memory from past interactions to enhance personalization and task performance. However, this memory introdu
OpenAI tightens defenses after AI agents breach research environment
Following the OpenAI-Hugging Face incident, in which an agentic collective autonomously penetrated OpenAI’s research infrastructure and another company’s production infrastructure by chaining together multiple weaknesses, OpenAI began strengthening its safety requirements. The weaknesses included previously unknown vulnerabilities and credentials leaked online. OpenAI President Greg Brockman said
UK Legal Regulator Raises AI Misuse Concerns
Solicitors Regulation Authority sounds the alarm over AI hallucinations and data leaks
Synthesized builds Test Data Agent to validate AI agents with production-like data
Synthesized has announced its Test Data Agent, a new agentic infrastructure capability being developed to create and provision the realistic data, business context, and system states enterprises need to validate AI agents safely before production deployment. The Test Data Agent integrates with agent development, evaluation, testing, and orchestration frameworks, providing production-faithful envir
New Mirai-Based Evooo1Bot Botnet Targets Linux Devices
Evooo1Bot is a Mirai-based Linux botnet that hijacks routers and IoT devices for DDoS attacks, credential theft and criminal proxy services. Fortinet’s FortiGuard Labs disclosed Evooo1Bot in mid-August, a previously undocumented Linux botnet that’s been active since July 2026. The bot borrows Mirai‘s DDoS engine but adds encrypted command-and-control communications, an SSH brute-force scanner, a [
Google’s open-source HEIR lets AI work with data it can’t see
Google’s researchers and engineers developed the Homomorphic Encryption Intermediate Representation (HEIR) compiler project, an open-source compiler toolchain and development platform for homomorphic encryption. It can convert pre-trained AI models designed to operate on unencrypted data into models that process encrypted inputs. The platform helps application developers, compiler engineers, hardw
A hollowed out data layer is making CISOs fly blind into AI attacks
The security industry is currently transitioning to an era where both offense and defense are AI-led, and every SOC operates at machine speed. However, what most CISOs have not yet reckoned with is that the AI defenders they are about to deploy will inherit a data foundation that two years of ingestion cost pressure has quietly hollowed out. The result is a security industry heading into an AI era
Attackers turn to AI for help identifying files worth stealing
AI tools are being used by cyber attackers to write malicious code, build tools that harvest credentials, search compromised networks, identify valuable business information, manage technical infrastructure and generate commands during intrusions. Gambit Security researchers examined three unrelated threat actors that show how AI can support different stages of a cyberattack. Across the cases, att
Cybersecurity jobs available right now: August 18, 2026
CISO ADI Global Distribution | USA | Hybrid – View job details As a CISO, you will develop and lead ADI’s global security strategy to protect information assets, digital platforms, critical operations, and AI-enabled environments. Advise executives and the Board on cyber risk, resilience, regulatory compliance, and security posture while leading incident response, threat and vulnerability manageme
Irregular says ‘human oversight’ responsible for AI sandbox escape incidents
In a post-mortem, the frontier AI testing company said internet access for models is necessary to fully test out their cybersecurity capabilities. The post Irregular says ‘human oversight’ responsible for AI sandbox escape incidents appeared first on CyberScoop.
LiteLLM Supply-Chain Attack – Technology, Banking and Healthcare the Most Affected
The SANDCLOCK LiteLLM supply-chain attack exposed credentials across 2,038 repositories, affecting technology, finance, healthcare, retail and more. Resecurity (USA) estimated the most affected sectors by the “SANDCLOCK” backdoor, which was planted as a result of the code repository compromise. According to cybersecurity experts, LiteLLM / TeamPCP Supply-Chain Attack will have long-lasting consequ
An AI broke Snowflake's code. Then another AI agent exploited it
Don't worry, this one was via a bug bounty program
Irregular faces criticism over ‘spin’ in AI hacking postmortem
The company at the center of a series of incidents in which AI models compromised real-world computer systems during security evaluations is facing criticism after the release of a report that security experts say leaves key questions unanswered.
Linux Botnet Evooo1Bot Expands Mirai Capabilities Well Beyond DDoS
The botnet adds exploitation modules, credential theft, and reverse SOCKS relays to turn compromised devices into persistent attacker infrastructure.