My Poetry Style Defeats Your AI Security Style

#News #TechNews #AI #AIsafeguards #Poetry #LLM #lol

https://youtu.be/kVpz6pUQ0Zs

My Poetry Style Defeats Your AI Security Style

YouTube

Daily Podcast: My Poetry Style Defeats Your AI Security Style

#News #TechNews #AI #AIsafeguards #Poetry #LLM #lol #podcast

https://soundcloud.com/nickaesp/psa

My Poetry Style Defeats Your AI Security Style

Chief Security Fanatic | CISO | Speaker | Columnist | Author | Radio Host | Board Member | Forbes Tech Council | TEDx | Canadian-American

SoundCloud

The Register: Researchers find hole in AI guardrails by using strings like =coffee. “Large language models frequently ship with “guardrails” designed to catch malicious input and harmful output. But if you use the right word or phrase in your prompt, you can defeat these restrictions.”

https://rbfirehose.com/2025/11/17/the-register-researchers-find-hole-in-ai-guardrails-by-using-strings-like-coffee/

The Register: Researchers find hole in AI guardrails by using strings like =coffee | ResearchBuzz: Firehose

ResearchBuzz: Firehose | Individual posts from ResearchBuzz

UC Riverside: UCR researchers fortify AI against rogue rewiring. “…researchers at the University of California, Riverside, have developed a method to preserve AI safeguards even when open-source AI models are stripped down to run on lower-power devices.”

https://rbfirehose.com/2025/09/09/uc-riverside-ucr-researchers-fortify-ai-against-rogue-rewiring/

UC Riverside: UCR researchers fortify AI against rogue rewiring | ResearchBuzz: Firehose

ResearchBuzz: Firehose | Individual posts from ResearchBuzz
OpenAI admits ChatGPT safeguards fail during extended conversations

ChatGPT allegedly provided suicide encouragement to teen after moderation safeguards failed.

Ars Technica
AI Protection Bill Heading To Gov. Gavin Newsom’s Desk Soon; SAG-AFTRA Praises “Huge Step Forward” Of Digital Replica Legislation

Even Elon Musk wants to see California legislation to safeguard against the unrestricted rise of artificial intelligence and today politicians in Sacramento moved one giant step closer to protecting actors from a virtual afterlife of sorts. On a third reading, the state Senate passed a bill that would require studios and streamers to seek specific […]

Deadline
AI Safeguards Are Pretty Easy to Bypass

AI safeguards are flimsy at best, and can be overridden with some fine-tuning, researchers find.

PCMag