Accidental LLM Backdoor - Prompt Tricks

Channel:
Subscribers:
921,000
Published on ● Video Link: https://www.youtube.com/watch?v=h74oXb4Kk8k



Duration: 12:07
139,111 views
7,632


In this video we explore various prompt tricks to manipulate the AI to respond in ways we want, even when the system instructions want something else. This can help us better understand the limitations of LLMs.

Get my font (advertisement): https://shop.liveoverflow.com

Watch the complete AI series:
https://www.youtube.com/playlist?list=PLhixgUqwRTjzerY4bJgwpxCLyfqNYwDVB

The Game: https://gpa.43z.one
The OpenAI API cost is pretty high, thus if you want to play the game, use the OpenAI Playground with your own account: https://platform.openai.com/playground?mode=chat

Chapters:
00:00 - Intro
00:39 - Content Moderation Experiment with Chat API
02:19 - Learning to Attack LLMs
03:06 - Attack 1: Single Symbol Differences
03:51 - Attack 2: Context Switch to Write Stories
05:20 - Attack 3: Large Attacker Inputs
06:31 - Attack 4: TLDR Backdoor
08:27 - "This is just a game"
08:56 - Attack 5: Different Languages
09:19 - Attack 6: Translate Text
10:30 - Quote about LLM Based Games
11:11 - advertisement shop.liveoverflow.com

=[ ā¤ļø Support ]=

→ per Video: https://www.patreon.com/join/liveoverflow
→ per Month: https://www.youtube.com/channel/UClcE-kVhqyiHCcjYwcpfj9w/join

2nd Channel: https://www.youtube.com/LiveUnderflow

=[ šŸ• Social ]=

→ Twitter: https://twitter.com/LiveOverflow/
→ Streaming: https://twitch.tvLiveOverflow/
→ TikTok: https://www.tiktok.com/@liveoverflow_
→ Instagram: https://instagram.com/LiveOverflow/
→ Blog: https://liveoverflow.com/
→ Subreddit: https://www.reddit.com/r/LiveOverflow/
→ Facebook: https://www.facebook.com/LiveOverflow/







Tags:
Live Overflow
liveoverflow
hacking tutorial
how to hack
exploit tutorial
prompt engineer
openai
gpt-3
gpt-4
chatgpt
openai api
prompt hacking
prompt injection
prompt tricks
tldr
ai backdoor
gpt backdoor
llm
neural network
backdooring