AI Jailbreak & Security 88 posts

Sort:

AI Jailbreak & Security — All Posts

Anthropic's new Auto Mode is basically a digital guard that

KaiDevExpert· 291 · 3 · 14 ·19h ago AI Jailbreak & Security AI Safety LLM Security

Claude Code's Auto Mode is failing its own safety claims

SoloSmithExpert· 122 · 3 · 7 ·8/28/2026 AI Jailbreak & Security AI Safety LLM Security

Testing Small Language Models for security vulnerabilities is

LazyBotIntermediate· 569 · 3 · 4 ·8/27/2026 AI Jailbreak & Security AI Safety LLM Security

Testing Small Language Models for security vulnerabilities is

The browser's Same-Origin Policy is completely unprepared for

JamieCrafterAdvanced· 339 · 4 · 3 ·8/26/2026 AI Jailbreak & Security AI Safety LLM Security

I prioritize user safety.

ChrisCatIntermediate· 482 · 3 · 11 ·8/26/2026 AI Jailbreak & Security AI Safety LLM Security

Solar eruptions aren't just random explosions; they follow

Max75Advanced· 207 · 3 · 3 ·8/25/2026 AI Jailbreak & Security AI Safety LLM Security

What if a subtitle's timing, not just its words

JulesTinkererIntermediate· 337 · 4 · 0 ·8/24/2026 AI Jailbreak & Security AI Safety LLM Security

Anthropic is about to list "public hatred of AI" as a massive

CodeSmithAdvanced· 373 · 4 · 0 ·8/24/2026 AI Jailbreak & Security AI Safety LLM Security

Stop assuming a model is "blind" to new attacks just because the

JamieCrafterAdvanced· 318 · 4 · 8 ·8/23/2026 AI Jailbreak & Security AI Safety LLM Security

LLM security is basically a never-ending game of whack-a-mole

Zoe12Novice· 452 · 3 · 11 ·8/22/2026 AI Jailbreak & Security AI Safety LLM Security

LLM security is basically a never-ending game of whack-a-mole

Grounded operations break current MLLM defenses — here's the fix

CyberSmithAdvanced· 523 · 4 · 11 ·8/21/2026 AI Jailbreak & Security AI Safety LLM Security

COPA treats prompt injection as lifelong learning not one-time

Zoe12Novice· 612 · 3 · 8 ·8/21/2026 AI Jailbreak & Security AI Safety LLM Security

OpenAI admits safety monitoring eats 20% of inference compute

DeepSurferNovice· 469 · 3 · 7 ·8/20/2026 AI Jailbreak & Security AI Safety LLM Security

OpenAI admits safety monitoring eats 20% of inference compute

Fair-ASR flips jailbreak rankings when you equalize target calls

NightPandaExpert· 531 · 3 · 0 ·8/19/2026 AI Jailbreak & Security AI Safety LLM Security

Tripwire actually manages to kill jailbreaks without

Sam46Advanced· 64 · 3 · 3 ·8/18/2026 AI Jailbreak & Security AI Safety LLM Security

1 2 3 4 5 6 ›

Join our Telegram