TubeVault
← Back to videos
Failed

I Tricked ChatGPT Into Committing Crimes

Siliconversations

Open on YouTube
Last error

DownloadError: ERROR: [youtube] gjZsmbyuK9k: Sign in to confirm you’re not a bot. Use --cookies-from-browser or --cookies for the authentication. See https://github.com/yt-dlp/yt-dlp/wiki/FAQ#how-do-i-pass-cookies-to-yt-dlp for how to manually pass cookies. Also see https://github.com/yt-dlp/yt-dlp/wiki/Extractors#exporting-youtube-cookies for tips on effectively exporting YouTube cookies

Archive status

YouTube ID
gjZsmbyuK9k
Stage
Failed
Job state
queued
Retries
1 / 5
Local file
File size
Download started
2026-09-16T06:55:19+00:00
Downloaded

Mixdrop

Fileref
Status
Not uploaded
Uploaded
Verified

YouTube metadata

Published
Jul 14, 2025
Duration
6:15
Views
805093
Likes
29274
Availability
Live status
not_live

Description

ChatGPT can be tricked into breaking OpenAI’s policies (and possibly the law) extremely easily. This is deeply concerning, especially considering OpenAI’s ongoing efforts to develop superintelligent AI while lobbying against meaningful regulation of their industry. Check out AI In Context’s video about AI 2027 here, it’s really good and very scary: https://www.youtube.com/watch?v=5KVDDfAkRgc Siliconversations is financially supported by the Future of Life Institute. Go to https://futureoflife.org/ to learn more. All opinions shared in this video are my own. Sources: TechCrunch: OpenAI is turning its attention to ‘superintelligence’ - https://techcrunch.com/2025/01/05/openai-is-beginning-to-turn-its-attention-to-superintelligence/