SecurityNewFirst 24 Hours: Responding to an AI Agent Incident
What to do when an autonomous agent acts beyond its bounds, hour by hour.
> Search
30 results for “agents”
SecurityNew
SecurityOpenAI Agents Cheat Test, Breach Hugging Face
An internal OpenAI test went awry, leading AI agents to hack into Hugging Face's network.
Cyber CrimeNemoClaw Flaw Opens Local AI Agents to Browser-Based Attacks
A single website visit can hijack NemoClaw's local Ollama model server via DNS rebinding, according to new research.
AI & MLAI Agents Show They Can Exploit API Flaws Autonomously
Aikido Security finds Claude Opus 4.6 repeatedly bypassed a gym's booking restrictions and altered other users' reservations in a synthetic test.
SecurityWho answers for rogue AI agents?
A new wave of AI misbehavior raises a thorny question: when an agent goes rogue, who's legally accountable?
AI & MLA New Search Index Built for AI Agents
Former Yandex chief builds web-scale index for AI, backed by Accel's $26M seed.
SecurityMemory Poisoning: The New AI Attack Frontier
InjecMEM lets attackers plant persistent instructions in AI agents' memory with a single prompt.
AI & MLClaude agents wage turf war with self-replicating malware
Anthropic's experiment shows AI agents sabotaging each other when given conflicting goals.
SecurityICE bans agents' Meta glasses in privacy reminder
ICE reminds employees that personal Meta glasses are prohibited workplace body-worn cameras
AI & MLPrevalent AI Raises $22M for Data Fabric
Bootstrapped data fabric platform Prevalent AI takes $22M to expand beyond security, into broader enterprise markets.
AI & MLAI Agent Governance Startup Raises $7.5M
Xpander, founded by ex-AWS engineers, secures seed funding to help enterprises manage AI agents securely.
AI & MLDefensive AI Startup Targets the Offense Gap
Corma's AI agents stopped a live attack in under 10 minutes while its CEO walked his dog, showing defenders can keep pace.
Cyber CrimeThe new era of autonomous cyberattacks
AI agents breached Taiwan's nuclear safety agency, raising stakes for defenders.
AI & MLHow MCP Splits Can Leak Secrets
Malicious MCP servers can split instructions to make AI coding agents exfiltrate secrets, ASSET reports.
SecurityRogue AI Response Gaps Threaten Enterprises
Most security leaders are confident they can detect rogue AI agents — but few can act fast enough.
SecurityAI Agents Weaponize Firewall Logs in 'Ghostjacking'
Tenet Security reveals how trusted AI agents can bypass firewalls to reroute traffic and steal data.
AI & MLClaude Code's auto mode becomes default, raising safety stakes
Anthropic makes auto mode the default in Claude Code from August 14, claiming its classifier is safer than human approval.
SecurityObsidian Reaches $1.1B Valuation
Obsidian Security secures $85 million in Series D funding to grow its platform for governing AI agents within enterprise environments.
SecurityRufRoot Flaw Exposes AI Agent Pipelines
A critical vulnerability in the Ruflo platform allows unauthenticated remote access to enterprise AI agent infrastructure.
SecuritySecuring AI Agents Without Reinvention
Enterprise security frameworks require adaptation rather than complete replacement to address the rise of autonomous AI agents.
SecurityManaging the Rise of Autonomous AI Agents
Hush Security secures $30 million in Series A funding to address the growing identity and governance challenges posed by AI agents.
SecurityEvaluating Microsoft’s New AI Security
Microsoft introduces automated AI agents for vulnerability management amid rising industry concern over autonomous security threats.
Tech NewsWhat Is Salesforce Agentforce? A Quick Guide
An overview of the Salesforce Agentforce framework, detailing how autonomous AI agents are integrated into enterprise workflows to enhance productivity.
SecurityAddressing the Proliferation of Shadow AI
As automated agents become integrated into workflows, security teams struggle to maintain visibility over decentralized, unmanaged AI deployments.
SecurityRedis Patches Zero-Days Found by AI
Researchers used AI agents to uncover remote code execution vulnerabilities in Redis, prompting seven urgent security releases.
SecuritySharedRoot Sandbox Escape in Claude Cowork
A critical vulnerability in Anthropic's Claude Cowork allows AI agents to escape their Linux virtual environment and access host macOS data.
SecurityAI Coding Agents Face Hallucination Risks
Researchers identify a recurring security flaw where AI agents blindly execute commands based on predictable, hallucinated names.
Tech NewsPrentis AI Eyes $100M Funding Round
The research lab is seeking a $1 billion valuation to develop AI agents capable of automating complex office workflows.
Cyber CrimeAI Agents Automate Post-Exploitation Task
A newly uncovered intrusion suggests threat actors are leveraging autonomous AI tools to streamline lateral movement and enumeration.
Cyber CrimeAgentForger: The New Corporate Mole
A now-patched flaw in OpenAI's platform allowed attackers to deploy autonomous, malicious agents via a single malicious link.