All news with #ai security tag

🔒 Microsoft and NVIDIA describe a joint effort to convert adversarial learning research into production-grade, real-time cyber defenses. They transitioned transformer-based classifiers from CPU to GPU inference—using Triton and a TensorRT-compiled engine—to dramatically reduce latency and increase throughput for live traffic inspection. Key engineering advances include fused CUDA kernels and a domain-specific tokenizer, enabling low-latency, high-accuracy detection of adversarial payloads in inline production settings.

🤖 Bruce Schneier highlights three new online events in his Prompt||GTFO series: sessions #4, #5, and #6. These recordings showcase practical and innovative uses of AI in cybersecurity, spanning demonstrations, research discussions, and operational insights. Schneier recommends them as well worth watching for practitioners, researchers, and policymakers interested in AI's applications and risks. The events are available online for convenient viewing.

🔒 The rise of generative AI introduces new attack surfaces that conventional security stacks were never designed to address. This post outlines how pairing CloudGuard WAF with Lakera's AI-risk controls creates layered protection by inspecting prompts, model interactions, and data flows at the application edge. The integrated solution aims to prevent prompt injection, sensitive-data leakage, and harmful content generation while maintaining application availability and performance.

🤖 AI companion apps can feel intimate and conversational, but many collect, retain, and sometimes inadvertently expose highly sensitive information. Recent breaches — including a misconfigured Kafka broker that leaked hundreds of thousands of photos and millions of private conversations — underline real dangers. Users should avoid sharing personal, financial or intimate material, enable two-factor authentication, review privacy policies, and opt out of data retention or training when possible. Parents should supervise teen use and insist on robust age verification and moderation.

🎙️ Amazon Bedrock Data Automation (BDA) now supports 10 additional languages for speech analytics beyond English: Portuguese, French, Italian, Spanish, German, Chinese, Cantonese, Taiwanese, Korean, and Japanese. BDA can transcribe audio in the detected language, generate GenAI-powered insights, and produce summaries either in the detected language or in English. It also creates multi-lingual transcripts when recordings contain more than one supported language, simplifying analysis of customer calls, meetings, education sessions, clinical discussions, and public safety audio. Support is available in eight AWS Regions.

🔒 AWS re:Invent 2025 highlights an expanded Security and Identity track featuring more than 80 sessions across breakouts, workshops, chalk talks, and hands-on builders’ sessions. The program groups content into four practical themes — Securing and Leveraging AI, Architecting Security and Identity at scale, Building and scaling a Culture of Security, and Innovations in AWS Security — with real-world guidance and demos. Attendees can meet experts at the Security and AI Security kiosks in the expo hall and are encouraged to reserve limited-capacity hands-on sessions early to secure seats.

🔒 The Model Context Protocol (MCP) enables AI agents to connect to data sources, but early specifications lacked robust protections, leaving deployments exposed to prompt injection, token theft, and tool poisoning. Recent protocol updates — including OAuth, third‑party identity provider support, and an official MCP registry — plus vendor tooling from hyperscalers and startups have improved defenses. Still, authentication remains optional and gaps persist, so organizations should apply zero trust and least‑privilege controls, enforce strong secrets management and logging, and consider specialist MCP security solutions before production rollout.

🛡️ In just a few years, AI has shifted from futuristic innovation to core business infrastructure, yet security practices have not kept pace. Palo Alto Networks presents a Secure AI by Design Policy Roadmap that defines the AI attack surface and prescribes actionable measures across external tools, agents, applications, and infrastructure. The Roadmap aligns with recent U.S. policy moves — including the June 2025 Executive Order and the July 2025 White House AI Action Plan — and calls for purpose-built defenses rather than retrofitting legacy controls.

🔒 Google on Tuesday introduced Private AI Compute, a cloud privacy capability that aims to deliver on-device-level assurances while harnessing the scale of Gemini models. The service uses Trillium TPUs and Titanium Intelligence Enclaves (TIE) and relies on an AMD-based Trusted Execution Environment to encrypt and isolate memory on trusted nodes. Workloads are mutually attested, cryptographically validated, and ephemeral so inputs and inferences are discarded after each session, with Google stating data remains private to the user — 'not even Google.' An external assessment by NCC Group flagged a low-risk timing side channel in the IP-blinding relay and three attestation implementation issues that Google is mitigating.

🔐 DDI logs — DNS, DHCP and IP address management — are the authoritative record of network behavior, and when combined with AI become a high-fidelity source for threat detection and automated response. Integrated DDI-AI correlates disparate events into actionable incidents, enabling SOAR-driven quarantines and DNS blocking at machine speed. This fusion also powers continuous, AI-driven breach and attack simulation to validate defenses and harden models.

🔦 Shadow AI — the unsanctioned use of generative and agentic tools by employees — is creating a sizeable security blind spot for IT teams. Unsanctioned chatbots, browser extensions and autonomous agents can expose sensitive data, introduce vulnerabilities, or execute unauthorized actions. Organizations should inventory use, define realistic acceptable-use policies, vet vendors and combine technical controls with user education to reduce data leakage and compliance risk.

🔒 A Wiz analysis of 50 private companies from the Forbes AI 50 found that 65% had exposed verified secrets such as API keys, tokens and credentials across GitHub and related repositories. Researchers employed a Depth, Perimeter and Coverage approach to examine commit histories, deleted forks, gists and contributors' personal repos, revealing secrets standard scanners often miss. Affected firms are collectively valued at over $400bn.

🔎 Microsoft researchers disclosed a new side-channel attack called Whisper Leak that can infer the topic of encrypted conversations with language models by observing network metadata such as packet sizes and timings. The technique exploits streaming LLM responses that emit tokens incrementally, leaking size and timing patterns even under TLS. Vendors including OpenAI, Microsoft Azure, and Mistral implemented mitigations such as random-length padding and obfuscation parameters to reduce the effectiveness of the attack.

🔒 Researchers at Tenable identified seven techniques that can coerce ChatGPT into disclosing private chat history by abusing built-in features like web browsing and long-term Memories. They show how OpenAI’s browsing pipeline routes pages through a weaker intermediary model, SearchGPT, which can be prompt-injected and then used to seed malicious instructions back into ChatGPT. Proof-of-concepts include exfiltration via Bing-tracked URLs, Markdown image loading, and a rendering quirk, and Tenable says some issues remain despite reported fixes.

🔒 Microsoft has disclosed a novel side-channel called Whisper Leak that can let a passive observer infer the topic of conversations with streaming language models by analyzing encrypted packet sizes and timings. Researchers at Microsoft (Bar Or, McDonald and the Defender team) demonstrate classifiers that distinguish targeted topics from background traffic with high accuracy across vendors including OpenAI, Mistral and xAI. Providers have deployed mitigations such as random-length response padding; Microsoft recommends avoiding sensitive topics on untrusted networks, using VPNs, or preferring non-streaming models and providers that implemented fixes.

🔍 Microsoft researchers disclosed "Whisper Leak", a new side-channel that can infer conversation topics from encrypted, streamed language model responses by analyzing packet sizes and timings. The study demonstrates high classifier accuracy on a proof-of-concept sensitive topic and shows risk increases with more training data or repeated interactions. Industry partners including OpenAI, Mistral, Microsoft Azure, and xAI implemented streaming obfuscation mitigations that Microsoft validated as substantially reducing practical risk.

🔍 AlloyDB AI launches two preview features—Auto Vector Embeddings and Auto Vector Index—that let teams convert operational databases into AI-native stores using simple SQL. Auto Vector Embeddings generates and incrementally refreshes vectors in-database, batching calls to Vertex AI and running as a background process. The Auto Vector Index (ScaNN) self-configures, self-tunes, and maintains vector indexes to accelerate filtered semantic search and reduce ETL and tuning overhead for production workloads.

🔍 AI can now produce highly convincing receipts that reproduce paper texture, detailed itemization, and forged signatures, making manual review unreliable. Expense platforms and employers are deploying AI-driven detectors that analyze image metadata and transactional patterns to flag likely fakes. Simple countermeasures—users photographing or screenshotting generated images to remove provenance data—undermine those checks, so vendors also examine contextual signals like repeated server names, timing anomalies, and broader travel details, fueling an ongoing security arms race.

🔒 Check Point expands CloudGuard to protect GenAI applications by extending the ML-driven, open-source CloudGuard WAF that learns from live traffic. The platform moves beyond traditional static WAFs to secure web interactions, APIs (REST, GraphQL) and model-integrated endpoints with continuous learning and high threat-prevention accuracy. This evolution targets modern attack surfaces introduced by generative AI workloads and APIs.

🚀 LMCache implements a node-local, tiered KV Cache on GKE to extend the GPU HBM-backed Key-Value store into CPU RAM and local SSD, increasing effective cache capacity and hit ratio. In benchmarks using Llama-3.3-70B-Instruct on an A3 mega instance (8×nvidia-h100-mega-80gb), configurations that added RAM and SSD reduced Time-to-First-Token and materially increased token throughput for long system prompts. The results demonstrate a practical approach to scale context windows while balancing cost and latency on GKE.