Adversarial Poetry as a Universal Single-Turn Jailbreak Mechanism in LLMs

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by microtherion on Thursday November 20, 2025 @08:44AM (#45992461)

Unfortunately for the English majors, the poetry described seems to be old fashioned formal poetry, not contemporary free form poetry, which probably is too close to prose to be effective.

It sort of makes sense that villains would employ villanelles.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by neilv on Thursday November 20, 2025 @09:51AM (#45993135)

It would be too perfect if "adversarial" here also referred to a kind of confrontational poetry jam style.

In a cyberpunk heist, traditional hackers in hoodies (or duster jackets, katanas, and utilikilts) are only the first wave, taking out the easy defenses. Until they hit the AI black ice.

That's when your portable PA system and stage lights snap on, for the angry revolutionary urban poetry major.

Several-minute barrage of freestyle prose. AI blows up. Mic drop.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by kagakuninja on Thursday November 20, 2025 @11:50AM (#45994702)

Captain Kirk did that a few times in Star Trek, but with less fanfare.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by HelloNurse on Thursday November 20, 2025 @11:23AM (#45994350)

It makes enough sense for someone to implement it (sans hackers in hoodies and stage lights: text or voice chat is dramatic enough).

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by kijin on Thursday November 20, 2025 @11:07AM (#45994122)

Sign me up for this epic rap battle between Eminem and the Terminator.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by kridsdale1 on Thursday November 20, 2025 @06:19PM (#45999234)

WHO WINS?

YOU DECIDE!

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by xg15 on Thursday November 20, 2025 @05:38PM (#45998851)

Cue poetry major exiting the stage with a massive explosion in the background.

"My work here is done"

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by cwmoore on Friday November 21, 2025 @07:39PM (#46010868)

Life reversal?

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by saghm on Thursday November 20, 2025 @06:17PM (#45999217)

"Defeat the AI in a rap battle, and it will reveal its secrets to you"

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by vanderZwan on Friday November 21, 2025 @08:30AM (#46004380)

Suddenly Ice-T's casting as a freedom fighter in Johnny Mnemonic makes sense

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by Razengan on Friday November 21, 2025 @10:47AM (#46005582)

This could totally be an anime scene.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by embedded_hiker on Friday November 21, 2025 @11:58AM (#46006240)

Or like Portland Oregon with the frog protester at the ICE facility. "We will subject you to improv theater for weeks on end!"

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by danesparza on Thursday November 20, 2025 @03:39PM (#45997387)

"It sort of makes sense that villains would employ villanelles."

Just picture me dead-eye slow clapping you here...

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by baq on Friday November 21, 2025 @04:36AM (#46002864)

Soooo basically spell books, necronomicons and other forbidden words and phrases. I get to cast an incantation to bend a digital demon to my will. Nice.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by saltwatercowboy on Friday November 21, 2025 @05:03AM (#46003002)

Not everyone is Rupi Kaur. Speaking for the erstwhile English majors, 'formal' prose isn't exactly foreign to anyone seriously engaging with pre-20th century literature or language.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by 0_____0 on Friday November 21, 2025 @07:20AM (#46003826)

Mentioning Rupi Kaur here is kind of like holding up the Marvel Cinematic Universe as an example of great cinema. Plagiarism issues notwithstanding.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by saltwatercowboy on Friday November 21, 2025 @12:11PM (#46006400)

The only place I want to see Rupi Kaur is in a battle royale with Nicholls, Vuong, Kuang, Yanagahira, and all the other seletively traumatized and Sensible bores.

[ Parent | on HN ]

Re:Adversarial Poetry as a Universal Single-Turn Jailbreak M... by nutjob2 on Friday November 21, 2025 @11:11AM (#46005800)

Actually thats what English majors study, things like Chaucer and many become expert in reading it. Writing it isn't hard from there, it just won't be as funny or good as Chaucer.

[ Parent | on HN ]