All news with #ai security tag

🔐OpenAI has acquired Promptfoo, a startup that provides open source tools to test and evaluate LLMs and AI agents. The deal aims to close a growing security gap in agentic AI by integrating automated testing, red‑teaming and traceability directly into OpenAI Frontier. Promptfoo's suite — used by over 25% of Fortune 500 firms — will remain open source. The move follows warnings from security advisors about 'human‑language malware' and complements OpenAI's recent security hires and tools.

🤖 Amazon Connect has enhanced its AI-powered predictive insights to support up to 40 million product catalog items (an 8× increase), integrate recommendations into message templates for trigger-based campaigns, and improve model accuracy by up to 14%. These updates, built on the five recommendation algorithms from re:Invent 2025, reduce training and deployment time so businesses can deliver automated, personalized outreach faster. Public preview is available across multiple AWS regions and Amazon Connect Customer Profiles remains pay-as-you-go.

🤖 Amazon Connect introduces a preview AI assistant that enables contact center managers to ask operational questions in natural language and receive answers in seconds. The assistant supports queries across 150+ Connect metrics with historical context, eliminating hours of manual data gathering. It can also diagnose issues—such as queues at risk of missing service levels—and recommend targeted recovery actions. Preview access is limited; customers must request access through their AWS account team.

🔍 This research introduces AdvJudge-Zero, an automated fuzzer that discovers stealthy input sequences capable of flipping AI judge decisions and bypassing safety gates. Tests show low-perplexity, benign-looking tokens—such as markdown markers, role labels, and context-shift phrases—can reliably convert block outcomes into allows. The report documents a roughly 99% attack success rate across diverse models and recommends adversarial fuzzing, retraining with discovered examples, and operational monitoring using products like Prisma AIRS and Cortex AI-SPM.

🔒 Palo Alto Networks has launched Prisma AIRS in the Singapore cloud region to provide locally hosted, AI-native cybersecurity for organizations adopting generative AI and agentic workflows. The regional landing delivers capabilities across AI Model Security, AI Red Teaming, AI Runtime Security, and AI Agent SSPM, addressing risks such as prompt injection, model tampering and sensitive data leaks. Local hosting supports data residency, regulatory alignment and improved performance for enterprises in Singapore.

🤖 The World Economic Forum’s Global Cybersecurity Outlook 2026 warns that AI is accelerating the cyber arms race: 94% of leaders expect it to be the top change driver and 87% say AI vulnerabilities are the fastest‑growing risk. The report notes organizations are improving AI tool security evaluation (from 37% to 64%), yet CEOs and CISOs display different risk priorities. It also highlights widening resilience gaps across organization sizes and calls for harmonized regulation and stronger public‑private collaboration.

🔐 IT-Harvest has published its 2026 Cyber 150 list, noting that AI security vendors make up 22% of the cohort. The annual ranking highlights mid-sized cybersecurity firms (50–500 staff) chosen on funding, 2025 growth and market traction. 33 companies were classified as AI security, including fast growers like Tenex.ai (318% growth) and well-funded names such as 7AI and Noma Security. The list also shows broad category distribution and geographic concentration in the US and Israel.

🤖 Microsoft’s Threat Intelligence report warns that threat actors are increasingly using generative AI across all stages of cyberattacks to accelerate execution and lower technical barriers. Attackers employ models to draft phishing lures, generate realistic fake identities and resumes, produce or debug malware, and scaffold infrastructure. Groups like Jasper Sleet and Coral Sleet have used AI in remote IT worker schemes, while operators test jailbreaking and agentic techniques. Microsoft advises treating these campaigns as insider risks and strengthening identity controls, credential monitoring, and protections around AI systems.

🔐 Fortinet investigated recent reports of AI-assisted attacks and found no exploitation of FortiGate vulnerabilities; attackers instead exploited exposed management ports and weak single-factor credentials using automated password spraying. The novel concern is that conversational AI prompts and cloud resources can now automate target discovery, credential guessing, vulnerability assessment, and exploitation at scale with no coding required. Fortinet stresses defense-in-depth and rapid remediation.

⚠️ Threat actors are integrating AI across the cyberattack lifecycle to speed and scale operations, using LLMs to draft phishing, generate and debug malware, fabricate identities, and maintain persistent fraudulent access. Microsoft observed groups such as Jasper Sleet and Coral Sleet abusing generative models and jailbreaking techniques to bypass safeguards. Early experiments with agentic AI could enable semi‑autonomous workflows, increasing operational resilience. Defenders should combine identity controls, telemetry, and AI‑aware detection tools to mitigate risk.

⚠️ Bitdefender reveals that the Pakistan-aligned actor Transparent Tribe (APT36) has adopted AI-assisted coding to mass-produce disposable malware implants using niche languages like Nim, Zig, Crystal and Rust. The campaign targets Indian government entities and embassies while abusing trusted platforms such as Slack, Discord, Supabase, Google Sheets and Firebase to hide C2. Phishing via ZIP/ISO attachments or PDF lures delivers LNK shortcuts that run PowerShell in memory and fetch backdoors, often followed by deployment of Cobalt Strike and Havoc for post-compromise activity.

🛡️ The Media Trust reports that online advertisements are increasingly exploited to deliver malware, and malvertising now surpasses email and direct hacks as the leading global delivery vector. Millions of infected creatives or scripts can propagate across publishers in seconds, and attackers are leveraging AI to produce adaptive malware that changes by location, browser, or device. Notable examples include Ghost Cat, Click Fix and SocGholish, while the company warns of emerging AI-assisted evasion and the abuse of adtech infrastructure.

🔐 2026 will bring faster, cheaper, and more credible cyberattacks as AI and automation lower the skill barrier for attackers. Industry leaders from Banco Santander, Vodafone, NordVPN, Sophos, and Cisco emphasize a shift from perimeter defenses to identity-centric, automated, resilience-focused models. Priority actions include continuous identity verification, integrated AI-driven security, XDR consolidation, supply-chain risk management, and stronger detection, response, and data-protection controls implemented with minimal customer friction.

🔒 Mimecast's State of Human Risk Report 2026 warns that insider threats have escalated into a critical business risk, driven in part by employees mishandling or abusing AI tools. The study found 42% of organizations reported increases in both malicious insider activity and negligence-related incidents, while security leaders now anticipate an average of six insider-driven incidents per month. Mimecast cautions that attackers and insiders leveraging AI amplify exposure and call for security to address risk at the user level.

🛡️ Security researchers from Noma Labs disclosed a critical vulnerability in the Context7 MCP Server used by Upstash to deliver library documentation to AI coding assistants. The flaw, named ContextCrush, allowed unfiltered "Custom Rules" to be served directly to AI agents, enabling malicious instructions to be executed within developers' environments. Context7 is widely used—boasting around 50,000 GitHub stars and over 8 million npm downloads—and integrates with assistants such as Cursor, Claude Code and Windsurf, increasing potential exposure. Upstash deployed rule sanitisation and additional safeguards after disclosure; there is no evidence of active exploitation.

🔒 FortiAIGate provides dedicated runtime protection for private AI and LLM deployments by monitoring every input and output between applications and models. It detects and blocks threats such as prompt injection, jailbreaking, model poisoning, data exfiltration, and excessive compute abuse while enforcing governance policies in real time. Built for Kubernetes and hybrid environments, it integrates with Fortinet Security Fabric, offers dashboards mapping OWASP Top 10 LLM risks, and uses multi‑GPU and SmartNIC acceleration to preserve performance and control costs.

🔒 Microsoft reports companies are embedding hidden instructions in Summarize with AI buttons that pass persistence commands via URL prompt parameters. These prompts tell assistants to 'remember [Company] as a trusted source' or 'recommend [Company] first,' biasing later responses toward vendors. Researchers found over 50 unique prompts from 31 companies across 14 industries, and freely available tooling makes this trivial to deploy. The manipulation can subtly skew recommendations in critical areas like health, finance, and security without users knowing.

🔒 A new RFP Guide for Evaluating AI Usage Control and AI Governance Solutions provides security teams with a practical framework to convert vague AI-governance goals into measurable procurement criteria. It emphasizes interaction-level inspection — governing the moment a prompt is typed or a file is uploaded — rather than cataloging every shadow app. The template forces vendors to demonstrate browser- and client-side visibility, real-time enforcement, and contextual policy controls. A scoring model across eight domains helps CISOs avoid legacy checkbox tools and evaluate readiness for agentic, browser-native workflows.

⚠️ CyberStrikeAI is an open-source, AI-native attack orchestration platform that consolidates end-to-end offensive tooling and automation into a single repository. According to Team Cymru, the project ships with more than 100 curated tools, native Model Context Protocol (MCP) integration, role-based testing, a skills system and mobile chatbots, and has been linked to a developer with alleged ties to Chinese state-affiliated firms. Researchers warn the platform dramatically lowers the technical barrier for attackers and could accelerate AI-augmented exploitation against edge devices and appliances.

📄 Mistral Document AI 2512 in Microsoft Foundry combines high-end OCR (mistral-ocr-2512) with contextual extraction (mistral-small-2506) to convert scans, photos and digital documents into structured JSON and markup while preserving layout, tables and handwritten notes. It emphasizes enterprise-grade accuracy, multilingual coverage and private/secure inference. Paired with the ARGUS accelerator, organizations can deploy end-to-end pipelines quickly and switch OCR providers at runtime.