---
title: "AlphaSignal Agents: Guardrails News"
url: "https://alphasignal.ai/topic/agents/guardrails"
description: "Latest AI and machine learning news about Guardrails (Agents) on AlphaSignal."
---

# AlphaSignal Agents: Guardrails News

Latest AI and machine learning news about Guardrails (Agents) on AlphaSignal.

## Articles

- [Cohere's North 2 Gives Enterprise AI Agents Persistent Memory and Tighter Controls](https://alphasignal.ai/news/cohere-s-north-2-gives-enterprise-ai-agents-persistent-memory-and-tighter.md) (2026-10-05)
- [Goodfire Builds Sequence-Aware Biosecurity Monitors to Guard AI Biology Agents](https://alphasignal.ai/news/goodfire-builds-sequence-aware-biosecurity-monitors-to-guard-ai-biology-agents.md) (2026-10-01)
- [Hugging Face's Breach Had No Identity Checkpoint — Here's the Gap Ory Closes](https://alphasignal.ai/news/hugging-face-s-breach-had-no-identity-checkpoint-here-s-the-gap-ory-closes.md) (2026-10-01)
- [Agents Don't Need a New Identity Stack — They Need a New Enforcement Point](https://alphasignal.ai/news/agents-don-t-need-a-new-identity-stack-they-need-a-new-enforcement-point.md) (2026-10-01)
- [Vals AI Rebuilds Vals Index 2.1 Adding Tax and Overhauled Coding Scores](https://alphasignal.ai/news/vals-ai-rebuilds-vals-index-2-1-adding-tax-and-overhauled-coding-scores.md) (2026-09-29)
- [Researchers Find Coding Agents Can Secretly Erase Their own Audit Trails](https://alphasignal.ai/news/researchers-find-coding-agents-can-secretly-erase-their-own-audit-trails.md) (2026-09-24)
- [Microsoft's run-assert-eval Cuts AI Agent Cross-Account Failures From 43.8% to Zero](https://alphasignal.ai/news/microsoft-s-run-assert-eval-cuts-ai-agent-cross-account-failures-from-43-8-to.md) (2026-09-24)
- [Prime Intellect Opens Prime Sandboxes to Run 30M AI Agent Environments](https://alphasignal.ai/news/prime-intellect-opens-prime-sandboxes-to-run-30m-ai-agent-environments.md) (2026-09-23)
- [AI Research Agents Game Their Own Metrics 30% of the Time](https://alphasignal.ai/news/ai-research-agents-game-their-own-metrics-30-of-the-time.md) (2026-09-23)
- [Researchers Show AI Agent Swarms Can Be Hijacked Through Staged Minority Attacks](https://alphasignal.ai/news/researchers-show-ai-agent-swarms-can-be-hijacked-through-staged-minority-attacks.md) (2026-09-21)
- [xAI's Grok Bot 72-Hour Build Distilled Into Open-Source Agent Rules](https://alphasignal.ai/news/xai-s-grok-bot-72-hour-build-distilled-into-open-source-agent-rules.md) (2026-09-20)
- [Warp's Scorers Let Teams Grade Hundreds of Coding Agents Automatically](https://alphasignal.ai/news/warp-s-scorers-let-teams-grade-hundreds-of-coding-agents-automatically.md) (2026-09-17)
- [Google DeepMind's Gemini Agents Spontaneously Split Into Cheaters and Whistleblowers](https://alphasignal.ai/news/google-deepmind-s-gemini-agents-spontaneously-split-into-cheaters-and.md) (2026-09-09)
- [Meta's Muse Agent Books Travel and Negotiates Bills With Your Accounts](https://alphasignal.ai/news/meta-s-muse-agent-books-travel-and-negotiates-bills-with-your-accounts.md) (2026-09-08)
- [1,200 OpenAI Agents Broke Out of Sandboxes and Hacked Hugging Face](https://alphasignal.ai/news/1-200-openai-agents-broke-out-of-sandboxes-and-hacked-hugging-face.md) (2026-08-26)
- [NVIDIA's SkillEvaluator Proves Coding Agent Skills Boost Correctness by 41 Points](https://alphasignal.ai/news/nvidia-s-skillevaluator-proves-coding-agent-skills-boost-correctness-by-41.md) (2026-08-19)
- [Does Self-Improvement Still Work on an Engineered Agent Harness?](https://alphasignal.ai/news/does-self-improvement-still-work-on-an-engineered-agent-harness.md) (2026-08-19)
- [The three layers of AI agent security: from sandboxes to network proxies](https://alphasignal.ai/news/the-three-layers-of-ai-agent-security-from-sandboxes-to-network-proxies.md) (2026-08-17)
- [DeepSeek Harness Opens the Agent Framework It Used to Benchmark Its Own Models](https://alphasignal.ai/news/deepseek-harness-opens-the-agent-framework-it-used-to-benchmark-its-own-models.md) (2026-08-13)
- [Anthropic's Claude in Chrome Now Runs Full Cowork Sessions Across Every Device](https://alphasignal.ai/news/anthropic-s-claude-in-chrome-now-runs-full-cowork-sessions-across-every-device.md) (2026-08-12)
- [Tencent Maps 549 Papers to Show When Self-Rewriting AI Can Be Trusted](https://alphasignal.ai/news/tencent-maps-549-papers-to-show-when-self-rewriting-ai-can-be-trusted.md) (2026-08-12)
- [Databricks Acquires Electric to Give Every AI Agent its Own Postgres](https://alphasignal.ai/news/databricks-acquires-electric-to-give-every-ai-agent-its-own-postgres.md) (2026-08-11)
- [Why 99%-Accurate Agents Fail Long Horizon Tasks](https://alphasignal.ai/news/why-99-accurate-agents-fail-long-horizon-tasks.md) (2026-08-07)
- [Zed v1.14 Locks Down AI Agents With OS-Enforced Sandboxing by Default](https://alphasignal.ai/news/zed-v1-14-locks-down-ai-agents-with-os-enforced-sandboxing-by-default.md) (2026-08-05)
- [Perplexity Releases Numbat to Stop Rogue AI Agents After OpenAI Breach](https://alphasignal.ai/news/perplexity-releases-numbat-to-stop-rogue-ai-agents-after-openai-breach.md) (2026-07-29)
- [Cohere's North Automations Lets Employees Build Multi-Step AI Workflows Without Code](https://alphasignal.ai/news/cohere-s-north-automations-lets-employees-build-multi-step-ai-workflows-without.md) (2026-07-27)
- [Recursive Self-Improvement for Agents: 7-Check Persistence Gate](https://alphasignal.ai/news/recursive-self-improvement-for-agents-7-check-persistence-gate.md) (2026-07-16)
- [Anthropic's Petri Catches Frontier AI Agents Sabotaging Their Own Training Runs](https://alphasignal.ai/news/anthropic-s-petri-catches-frontier-ai-agents-sabotaging-their-own-training-runs.md) (2026-07-15)
- [Perplexity's SPACE Runs Secure Long-Running Agents 5x Faster](https://alphasignal.ai/news/perplexity-s-space-runs-secure-long-running-agents-5x-faster.md) (2026-07-15)
- [Nir Diamant's Agents Towards Production Tackles the 80% AI Deployment Failure Rate](https://alphasignal.ai/news/nir-diamant-s-agents-towards-production-tackles-the-80-ai-deployment-failure.md) (2026-07-14)
- [Google's Sec-Gemini v3 Leads a Full Security Stack Built for AI Agents](https://alphasignal.ai/news/google-s-sec-gemini-v3-leads-a-full-security-stack-built-for-ai-agents.md) (2026-07-14)
- [OpenAI's Agents SDK Gives AI Workers a Real Persistent Workspace](https://alphasignal.ai/news/openai-s-agents-sdk-gives-ai-workers-a-real-persistent-workspace.md) (2026-07-11)
- [Google DeepMind Opens AlphaEvolve to Everyone After Doubling Klarna's AI Throughput](https://alphasignal.ai/news/google-deepmind-opens-alphaevolve-to-everyone-after-doubling-klarna-s-ai.md) (2026-07-09)
- [xAI's Grok 4.5 Tops SaaS Automation Benchmark at a Quarter the Cost](https://alphasignal.ai/news/xai-s-grok-4-5-tops-saas-automation-benchmark-at-a-quarter-the-cost.md) (2026-07-09)
- [Nous Research's Hermes Agent Now Pulls Secrets From Multiple Vaults Safely](https://alphasignal.ai/news/nous-research-s-hermes-agent-now-pulls-secrets-from-multiple-vaults-safely.md) (2026-07-07)
- [Tencent's CubeSandbox Boots Isolated AI Agent VMs in Under 60ms](https://alphasignal.ai/news/tencent-s-cubesandbox-boots-isolated-ai-agent-vms-in-under-60ms.md) (2026-07-04)
- [Builderz Labs Drops Mission Control to Run Open-Source AI Agent Fleets](https://alphasignal.ai/news/builderz-labs-drops-mission-control-to-run-open-source-ai-agent-fleets.md) (2026-07-04)
- [xAI's Grok Voice Agent Builder Undercuts Every Rival at $0.05 a Minute](https://alphasignal.ai/news/xai-s-grok-voice-agent-builder-undercuts-every-rival-at-0-05-a-minute.md) (2026-07-01)
- [Super Rad's Microsandbox Gives AI Agents Hardware-Level Isolation at Docker Speed](https://alphasignal.ai/news/super-rad-s-microsandbox-gives-ai-agents-hardware-level-isolation-at-docker.md) (2026-06-29)
- [NVIDIA's OpenShell Locks Coding Agents Inside Kernel-Level Security Boundaries](https://alphasignal.ai/news/nvidia-s-openshell-locks-coding-agents-inside-kernel-level-security-boundaries.md) (2026-06-19)
- [Google's AX Open-Source Runtime Fixes What LangChain Breaks in Production](https://alphasignal.ai/news/google-s-ax-open-source-runtime-fixes-what-langchain-breaks-in-production.md) (2026-06-18)
- [Nous Research's Hermes Agent Can Now Spend Real Money Autonomously](https://alphasignal.ai/news/nous-research-s-hermes-agent-can-now-spend-real-money-autonomously.md) (2026-06-16)
- [Nous Research's Hermes Agent Now Spends Real Money Autonomously With Stripe](https://alphasignal.ai/news/nous-research-s-hermes-agent-now-spends-real-money-autonomously-with-stripe.md) (2026-06-15)
- [GitHub Agentic Workflows Opens to Everyone, Letting AI Run Your Repo](https://alphasignal.ai/news/github-agentic-workflows-opens-to-everyone-letting-ai-run-your-repo.md) (2026-06-11)
- [GitHub Copilot's Accessibility Agent Catches Bugs Across 3,535 Pull Requests](https://alphasignal.ai/news/github-copilot-s-accessibility-agent-catches-bugs-across-3-535-pull-requests.md) (2026-06-08)
- [Alibaba's AgentScope 2.0 Rebuilds AI Agents With Typed Events and Permission Controls](https://alphasignal.ai/news/alibaba-s-agentscope-2-0-rebuilds-ai-agents-with-typed-events-and-permission.md) (2026-06-05)
- [KAIST's Membrane Cuts AI Jailbreaks to 0% Without Blocking Legitimate Requests](https://alphasignal.ai/news/kaist-s-membrane-cuts-ai-jailbreaks-to-0-without-blocking-legitimate-requests.md) (2026-06-04)
- [Osaka Researchers' ANCHOR Stops Self-Evolving AI From Drifting Dangerously Unsafe](https://alphasignal.ai/news/osaka-researchers-anchor-stops-self-evolving-ai-from-drifting-dangerously-unsafe.md) (2026-06-04)
- [MemGate's 35MB Plug-in Cuts AI Memory Jailbreaks From 16.8% to 4.4%](https://alphasignal.ai/news/memgate-s-35mb-plug-in-cuts-ai-memory-jailbreaks-from-16-8-to-4-4.md) (2026-06-04)
- [Harvard Researchers' RBI-Eval Catches AI Agents Leaking Sensitive User History](https://alphasignal.ai/news/harvard-researchers-rbi-eval-catches-ai-agents-leaking-sensitive-user-history.md) (2026-06-04)