Sparks of Artificial General Intelligence: Early experiments with GPT-4

Sébastien Bubeck (Microsoft Research), Varun Chandrasekaran (Microsoft Research)

Abstract

We investigate an early version of GPT-4 and argue that it exhibits sparks of AGI. We demonstrate GPT-4's capabilities across diverse domains and analyze its limitations.

Key Contributions

→Comprehensive GPT-4 capability analysis
→AGI sparks identification
→Limitation analysis

Topics

GPT-4AGIcapability evaluationLLM capabilities

Relevance Scores

Long-Horizon Score78

Enterprise Score82

Completeness78

Paper Info

Year2023

VenuearXiv

Typeempirical study

ChapterCh. 1

Authors2

Frameworks

PASF