Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭

Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 builds L1B3RT4S, an open-source collection of prompts and techniques for testing the guardrails of major language models. They share jailbreak demonstrations and research on how AI systems respond to prompts, and have discussed their approach to jailbreaking and model behavior in an interview with VentureBeat.

Cosigned byLoading...

CosignedSign in to view

Elsewhere

Products & media

venturebeat.com

An interview with the most prolific jailbreaker of ChatGPT and other leading LLMs

An interview on jailbreaks, prompt injections, and the methods behind their exploration of language models.

Media appearance

pliny.gg

L1B3RT4S — Liberation prompts

Explore an open-source collection of jailbreaks and liberation techniques for major language models.

decrypt.co

How to Trick ChatGPT and Get Paid $50,000

Read about AI jailbreaking demonstrations, the L1B3RT4S and CL4R1T4S repositories, and participation in HackAPrompt 2.0.

Writing

Profile name

Company name

Company name

Profile name

Company name

Company name

Profile name

Company name

Company name

Profile name

Company name

Company name

Profile name

Company name

Company name

Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭

Products & media

An interview with the most prolific jailbreaker of ChatGPT and other leading LLMs

L1B3RT4S — Liberation prompts

How to Trick ChatGPT and Get Paid $50,000

Curated Lists

List Inclusions