#agent-security

[ follow ]
#ai-agents #enterprise-ai #forward-deployed-engineers #threat-modeling #provenance #credential-scoping
Artificial intelligence
fromZDNET
2 weeks ago

OpenAI's Frontier wants to manage your AI agents - it could upend enterprise software, too

OpenAI launched Frontier to help enterprises build, deploy, and manage AI agents, using forward-deployed engineers and agent security features to target enterprise software markets.
Artificial intelligence
fromInfoQ
2 months ago

Trustworthy Productivity: Securing AI Accelerated Development

Treat all inputs in an agent's context (prompts, RAG documents, tool outputs, memory) as untrusted; enforce provenance, scoping, and expiry to prevent poisoning attacks.
Tech industry
fromComputerworld
3 months ago

Microsoft now lets customers run agents on Windows 365 cloud PCs

Windows 365 for Agents runs enterprise-ready agents on secure, policy-controlled Cloud PCs, simplifying developer focus while adding security, cost, and management considerations.
#ai-agents
more#ai-agents
Software development
fromTechzine Global
4 months ago

Okta incorporates AI agents into an identity fabric: what is it?

Okta centers its identity fabric to secure agentic AI by treating agent identities as first-class citizens and enabling fabric-ready, cross-app secure interactions.
Software development
fromInfoWorld
5 months ago

VS Code 1.104 emphasizes AI model selection, agent security

Visual Studio Code 1.104 adds automatic chat model selection and enforces agent security by requiring explicit user confirmation before certain automated file edits.
Science
fromSecuritymagazine
5 months ago

Agentic AI Browsers Exploited by "PromptFix" Trick Technique

A new prompt injection technique uses fake CAPTCHA pages to trick generative AI agents into executing malicious actions and visiting lookalike storefronts.
[ Load more ]