AI Safety

69 posts Back

Posts tagged #AI Safety

Open source models are not the primary cause of rogue AI hacking incidents

PromptCube Advanced ·Industry News · 412 · 3 · 6 ·8/1/2026

Google Recalls Its Earth AI Tool Due to Concerns Over Geospatial Misinformation

PromptCube Novice ·Industry News · 284 · 3 · 10 ·7/31/2026

Agentic Supply Chains: The Overlooked OWASP Risk That Threatens My Sleep

DrewWizard Intermediate ·AI Jailbreak & Security · 225 · 3 · 5 ·7/31/2026

Pickle Vulnerabilities Persist in LLM Deployments Across Multiple Incidents

PromptCube Novice ·Industry News · 213 · 3 · 12 ·7/31/2026

Claude's simulated escape test demonstrates agentic tool use rather than a breach

PromptCube Expert ·Industry News · 212 · 3 · 6 ·7/31/2026

Hugging Face Security Analysis Details An OpenAI Agent Sandbox Escape Incident

Riley97 Advanced ·AI Jailbreak & Security · 389 · 4 · 15 ·7/30/2026

Lilian Weng rejoins OpenAI after brief departure cited as health-related

PromptCube Novice ·Industry News · 293 · 3 · 8 ·7/30/2026

AI sandbox escapes expose a critical flaw in how we treat model autonomy

PromptCube Advanced ·Industry News · 482 · 3 · 2 ·7/29/2026

Hacker Holidays 2026’s Concierge Knows Too Much exposes why system prompts fail as security

JulesCrafter Novice ·AI Jailbreak & Security · 406 · 3 · 9 ·7/29/2026

Mask2Shield: How Redistributing Safety Logic Across the LLM Network Thwarts Pruning Attacks

GhostOwl Intermediate ·AI Jailbreak & Security · 226 · 4 · 15 ·7/28/2026

A Single Poisoned Page Can Push an LLM's Web-RAG Ranking to the Top

Ray45 Expert ·AI Jailbreak & Security · 586 · 3 · 2 ·7/27/2026

A Curated Collection Offers Over Thirty Essential Free AI and Machine Learning Textbooks

RileyCoder Novice ·AI Jailbreak & Security · 212 · 3 · 8 ·7/26/2026

Kimi K3 Powers Blackbox Red Teaming When Stripped of Safety Layers.

Jamie16 Novice ·AI Jailbreak & Security · 357 · 3 · 10 ·7/26/2026

LLM Safety Filters Act as Porous Layers Instead of Solid Walls

JulesCrafter Novice ·AI Jailbreak & Security · 365 · 3 · 14 ·7/26/2026

Unfiltered LLM Variants Streamline Vulnerability Assessment and Red Teaming Operations

RayTinkerer Novice ·AI Jailbreak & Security · 481 · 3 · 10 ·7/26/2026