Tell HN: Pangram is easily-defeatable with Claude
2 by nunez | 1 comments on Hacker News.
I use Pangram all of the time to detect whether what I'm reading is fully AI-generated or not. Today, I wondered how easy it was to defeat it. All I had to do was show them that the text they generated was detected as 100% AI generated to get them to generate a "human-sounding" text snippet. Claude Sonnet: https://ift.tt/E7Z64PV Interestingly, ChatGPT 5.4 won't do it, at least not with the default model it uses: https://ift.tt/OxKstVY. I'm guessing it can be jailbroken to do it though.
Don't forget to subscribe our youtube channel Click here:- http://www.youtube.com/c/techgk Product of the day
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment