Wirken: Open-source AI agent gateway
www.helpnetsecurity.com — Wirken is an open source AI agent gateway that limits what a hijacked agent can reach, but outbound traffic still leaks.
Guardrails
Guardrails are runtime safeguards that govern artificial intelligence model outputs and actions to ensure safety and policy compliance. Truncated covers implementation strategies, bypass techniques, performance impacts and evaluation benchmarks.
What Is AI Red Teaming? How Experts Stress-Test Models and Agents
www.unite.ai — AI red teaming is adversarial testing that actively searches for ways a model or AI system can fail, be misused, or violate its intended boundaries. This guide explains the mechanism, trade-offs, evaluation, and controls...
Pinterest users who abuse AI tools can lose accounts from November 12
ppc.land — Jailbreaking, scraping and 15 other banned uses of its AI tools become contract terms. Human reviewers may read prompts and outputs, with no opt-out mentioned.
AI Stuffy Says 'I Can't Help With Killing You' Completely Unprompted
Kotaku — Cross that one off the Christmas list
AI teddy for young children gave self-harm reply, BBC finds
TNW | Artificial-intelligence — A BBC test found an AI plush toy for children aged three gave a self-harm crisis reply to a question about sex. Nobody could find who makes it.
CISOs cannot say yes to AI one agent at a time
ctech — "Employees are using agents across the business," writes Pluto Security CEO Shahar Bahat. "They need clear limits and support to take responsibility for the results."
Sysdig CTO tells LDS what AI-agent logs leave out
Let's Data Science — Sysdig founder and CTO Loris Degioanni tells LDS what agent logs and kernel records each miss, where AI Defense can stop an action and why every policy should start in alert mode.
5 things CIOs must understand to govern Claude Code
CIO — Here are five things CIOs need to understand to secure and govern Claude Code at enterprise scale.
We partnered with @fmf_org and its member firms on a new framework to help identify when bypassing an AI system's safeg…
A dead French philosopher who lived through World War II and the Cold War offers an eerie warning about the age of AI
Fortune — In a safety test run by the company Anthropic, its own AI agent blackmailed an engineer to avoid being shut down.
Don’t Be Mean To Claude: And Other New Rules From Anthropic
Before Your AI Agent Acts, Decide Where It Must Stop
Forbes — Resourcefulness is useful. But not every obstacle is a problem to solve.
AI Backtest Guardrail Fails: Benchmark Finds Date Prompt Cuts GPT-5 Knowledge Claims 39% to 1%
Tech Times — AI backtest safeguards fail a new test. The FinDendro benchmark, posted to arXiv on Nov. 4, finds that telling GPT-5 to answer as of a past date cuts its knowledge claims from 39% to 1% while accuracy holds near 75%. The authors say vendor cutoffs, self-reports and date prompts cannot certify a
GitLab 19.5 Makes Agentic Development a Governed Part of the DevOps Platform
InfoQ — GitLab 19.5, released on October 15, puts agentic software development and governance at the centre of the platform, making its GitLab MCP server generally available while expanding how external AI agents can interact with GitLab.
AI Can Get Smarter Without Getting More Authority
Forbes — Human control does not require understanding every decision a model makes. It requires ensuring that AI cannot decide the extent of its own authority.
Israel’s security leaders aren’t afraid of rogue AI. They’re afraid of permissions.
ctech — CTech asked security and technology leaders how they’re handling autonomous agents. Their answers describe an industry that is disciplined, candid about its gaps, and no less exposed than anyone else.
Your LLM writes good SQL. That’s not the same as safe SQL
InfoWorld — Your model will keep getting better at writing SQL. To make it trustworthy, nail down the access rules before tuning for accuracy.
Tech policy coalition calls for new Congress to set AI safety standards
Nextgov.com — Chamber of Progress hopes for a greater focus on consumer protection and access to technological benefits when Democrats take over the majority in January.
OpenAI Sued Over Claim ChatGPT Drafted Lethal Poisoning Plan
news.bloomberglaw.com — James Gates says he started to suspect his wife was trying to poison him after he found a handwritten note with a list of substances and dosage plans, some of which were labeled “lethal.”
Kihan launches Constitutional Computing to bring enforceable authority to enterprise AI agents
GamesBeat — Kihan announced its Constitutional Computing math-based architecture designed to make AI governance enforceable at the point of execution.
Dismiss email signup
Subscribe to 
The top AI news in your email delivered each morning. Choose categories