Can Anthropic's Claude Trick OpenAI's ChatGPT at Math?

Channel:
Kyle Kabasares
Subscribers:
16,700
Published on ● Video Link: https://www.youtube.com/watch?v=6Jc6qdflOJ0


Duration: 0:00
1,004 views
46

We use Claude Sonnet 3.5 to try and come up with sneakily disguised math questions that do not pertain to the actual problem.

Link to the Apple Research Paper: https://arxiv.org/pdf/2410.05229



Other Videos By Kyle Kabasares


2024-11-15Should I share my random late night thoughts more often? 😅 #AI #artificialintelligence #climate
2024-11-12Life 3.0 by Max Tegmark is a must read for anyone interested in AI.
2024-11-09I Used Perplexity AI Search for the First Time
2024-11-08Reflecting on OpenAI’s o1-Full, preview, and mini Models
2024-11-0711/7 marks the birthday of Marie Curie, the only one to win a Nobel Prize in Physics AND Chemistry.
2024-11-06Post #election2024 thought: Scientists, we have to start speaking truth to power more than ever.
2024-11-02OpenAI o1 FULL Was Accidentally Released Early?! Let's Test It!
2024-11-01ChatGPT Search: Use at Your Own Risk (A PhD's Perspective)
2024-10-31Can OpenAI's o1-mini Decipher my "Secret Messages"?
2024-10-30Which AI, ML, and Data Science Books Do I Recommend?
2024-10-29Can Anthropic's Claude Trick OpenAI's ChatGPT at Math?
2024-10-28Anthropic’s Claude Computer Use CAN Play Chess…Badly
2024-10-26Can We Trick OpenAI's o1-Preview Model at Math? (Putnam Exam Problem)
2024-10-24Saying Good-Bye to My Favorite Quantum Mechanics Textbook...
2024-10-23Can Anthropic's AI Agent Play Online Chess?!
2024-10-23Using Anthropic's New Agentic Computer Use for the First Time!
2024-10-21Reacting to: How I Become a Homeless Physics Professor
2024-10-21Google's Notebook LM Created a Podcast of My Physics PhD Thesis
2024-10-18Live from Worldcoin’s “A New World”: Keynote Address with Alex Blania and Sam Altman (OpenAI CEO)
2024-10-18I Went to Sam Altman’s Worldcoin Event As a Non-Techie
2024-10-17Live from WorldCoin’s: “A New World” in San Francisco, CA (Pre-Keynote, Breakfast)