How Far Are AI Scientists from Changing the World?

TL;DR

This survey reviews the current state of AI Scientist systems powered by large language models, analyzing their achievements, limitations, and future potential to revolutionize scientific discovery and address grand challenges.

Contribution

It provides a comprehensive, prospect-driven analysis of AI Scientist systems, highlighting key bottlenecks and essential components for future breakthroughs in scientific AI.

Findings

Current AI Scientist systems have made significant progress in automating research tasks.

Major bottlenecks include data limitations, reasoning capabilities, and integration of scientific knowledge.

Achieving human-level scientific discovery requires overcoming these challenges and developing ground-breaking AI agents.

Abstract

The emergence of large language models (LLMs) is propelling automated scientific discovery to the next level, with LLM-based Artificial Intelligence (AI) Scientist systems now taking the lead in scientific research. Several influential works have already appeared in the field of AI Scientist systems, with AI-generated research papers having been accepted at the ICLR 2025 workshop, suggesting that a human-level AI Scientist capable of uncovering phenomena previously unknown to humans, may soon become a reality. In this survey, we focus on the central question: How far are AI scientists from changing the world and reshaping the scientific research paradigm? To answer this question, we provide a prospect-driven review that comprehensively analyzes the current achievements of AI Scientist systems, identifying key bottlenecks and the critical components required for the emergence of a scientific agent capable of producing ground-breaking discoveries that solve grand challenges. We hope this survey will contribute to a clearer understanding of limitations of current AI Scientist systems, showing where we are, what is missing, and what the ultimate goals for scientific AI should be.