Devina Jain

Message

Can Persuasion Break AI Safety? Exploring the Interplay Between Fine-Tuning, Attacks, and Guardrails

This blogpost was created as a part of the AI Safety Fundamentals course by BlueDot Impact. Code available upon request due to sensitive nature of some content. Disclaimer: This post represents the findings of independent research carried out solely for academic and educational purposes. The methods, experiments, and conclusions presented...

Feb 4, 2025•9

Message

8 karma

1 post

Member for a year

Devina Jain — LessWrong

Devina Jain

Message

Devina Jain

Can Persuasion Break AI Safety? Exploring the Interplay Between Fine-Tuning, Attacks, and Guardrails

Feb 4, 2025•9

Message

8 karma

1 post

Member for a year

Can Persuasion Break AI Safety? Exploring the Interplay Between Fine-Tuning, Attacks, and Guardrails

Devina Jain

This blogpost was created as a part of the AI Safety Fundamentals course by BlueDot Impact. Code available upon request due to sensitive nature of some content.

Disclaimer:
This post represents the findings of independent research carried out solely for academic and educational purposes. The methods, experiments, and conclusions presented here are the result of my personal inquiry and are shared in the interest of advancing understanding of AI safety and fine-tuning dynamics. I do not endorse or encourage the use of these methods for any harmful or malicious activities. Any unauthorized or unethical use of this information is strongly discouraged.

TL;DR: Amid growing concerns about adversarial use of AI, this post investigates whether fine-tuning... (read 2715 more words →)

LESSWRONG
LW

LESSWRONG
LW

Devina Jain

Devina Jain

Devina Jain

Can Persuasion Break AI Safety? Exploring the Interplay Between Fine-Tuning, Attacks, and Guardrails

Devina Jain

Devina Jain

Devina Jain

Can Persuasion Break AI Safety? Exploring the Interplay Between Fine-Tuning, Attacks, and Guardrails