Can an AI solve a classic paradox designed to test human decision-making?

LLMs and Newcomb's Problem - Computerphile · 18:32 · Video: Computerphile · Watch at the source

In short: Newcomb's Problem is a famous thought experiment that reveals deep contradictions in how we make choices. This video explores what happens when you apply this logic to a large language model, specifically Anthropic's Claude, to see if it behaves like a rational agent or something else entirely.

Newcomb's Problem presents a scenario that appears straightforward but quickly becomes complex, especially when considering the behavior of an identical twin. The paradox challenges our understanding of rational decision-making and predictive capability.

Aric Floyd demonstrates the experiment, testing whether an AI can navigate the dilemma. The results show that Claude does not always follow the expected logic, highlighting the unpredictable nature of interacting with advanced language models in the context of philosophical puzzles.

Opens the Intellect+ feed with this card first. Swipe up for the next one.

Sources

To cite this card, link to https://intellectplus.co.uk/app/technology-llms-and-newcomb-s-problem-computerphile.

How we source, check and correct cards: Editorial standards.

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

Finding something worth knowing…