AI Briefing#ai-security
The latest AI and developer news about #ai-security, with the original source and a short summary.
The latest AI and developer news about #ai-security, with the original source and a short summary.
GitHub introduced LLM-based contextual reasoning to reduce false positives in secret scanning and improve detection accuracy.
ChatGPT has raised security concerns by recommending fake websites of bankrupt brands due to training data contamination.
Indirect prompt injection attacks using transfer records have been shown to turn banking AI agents into phishing tools.
KakaoBank developed a sequence-based FDS model that uses Vision Transformer to analyze customers' continuous behavioral patterns.
ElevenLabs has signed an MOU with the UK government and is expanding its business by doubling its workforce in the UK.
University of Toronto researchers demonstrated the danger of an 'AI worm' that modifies its own strategy and spreads using publicly available AI models.
Anthropic is expanding its Mythos model, specialized in detecting security vulnerabilities, to 150 companies across 15 countries.
Anthropic is expanding Project Glasswing partners to 150 organizations to strengthen software security.
A new benchmark called CVE-Bench has been released to evaluate LLM agents' ability to patch security vulnerabilities.
This covers the verification of major security vulnerabilities discovered by Anthropic's unreleased model, Mythos.
220,000 valid credentials were found in Hugging Face's public training data.
A vulnerability enabling data exfiltration and phishing attacks via indirect prompt injection has been discovered in the ChatGPT for Google Sheets extension.
Anthropic has revealed the various sandbox and isolation technologies applied to each product, including Claude.ai and Claude Code.
PolyRange, an AI security evaluation benchmark that addresses the data contamination problem and reflects dynamic defense environments, has been released.
For public sector CISOs, this presents a roadmap and phased execution strategy for building a security program leveraging AI.
To address the security and governance issues arising from AI agent adoption, the introduction of harness engineering is necessary.
As AI inference costs surge, a robust defense system is needed against inference theft attacks that involve unauthorized use or resale of these resources.
AI Gateway now supports an allowlist feature that lets teams restrict which AI providers can be used.
Amazon has announced 68 ARA winners who will receive support for research in various fields including AI security and agentic AI.
It warns of the security vulnerability risks in code generated by AI coding tools and emphasizes the need for technical guardrails.
The FBI and DHS have officially designated anti-tech extremism targeting the AI industry as a domestic threat factor.
The use of AI-assisted tools is driving a sharp surge in security vulnerability reports for the curl project.
A new threat is emerging in which AI tools find security vulnerabilities in Linux and Windows using nothing more than simple prompts.
Cisco forecast that AI demand will spread beyond hyperscalers to enterprises and sovereign clouds, ushering in a networking-centric infrastructure cycle.