Black Hat 2024: Researcher exposes Microsoft Copilot AI vulnerabilities

Microsoft Copilot can be exploited for automated spear-phishing attacks, cite phantom sources, expose communications, and data poisoning

POSTMAN: AI-Powered OSINT Platform for Defense Intelligence and Five Eyes

POSTMAN will unify open source intelligence across the defense intelligence enterprise and to members of “Five Eyes”

Using Generative AI LLM agents to exploit zero-day vulnerabilities

Can Generative AI be used for exploiting zero-day vulnerabilities on websites? A team of research scientists investigates the capability of Large Language Model (LLM) agents to find out.

NVIDIA Nemotron-4 340B: Large Language Model Training with Synthetic Data

Nemotron-4 340B, a family of models optimized for NVIDIA NeMo and NVIDIA TensorRT-LLM, includes cutting-edge instruct and reward models, and a dataset for generative AI training.

Microsoft Windows 11 Recall AI records user desktops every two seconds

The new feature raises significant privacy and cybersecurity concerns as anything captured within a screenshot will not be obfuscated–even passwords.

Microsoft Deploys GPT-4 to Azure Government Top Secret Cloud for DoD

OpenAI’s GPT-4 multimodal large language model is coming to Azure Government Cloud Top Secret

RSAC 2024: Crowdstrike Falcon Cloud Security enhanced for cloud asset visualization

Crowdstrike is enhancing its Falcon Cloud Security platform for AI-assisted cyber incident detection, mitigation and response

RSAC 2024: Google Threat Intelligence unveiled

Google’s new cybersecurity threat intelligence offering is powered by Gemini, VirusTotal, and Mandiant

U.S. Government Releases New AI Security Guidelines for Critical Infrastructure

The new guidance creates a department-wide AI roadmap for securing critical infrastructure from AI and cyberattack vectors.

Google releases Gemini Generative AI prompt guides

Google has released two Gemini generative AI prompt guides for Workspace and Google Cloud to help you create the best prompts.