Comments about: AI scientist 'team' joins the search for extraterrestrial life

Comments about the article in Nature: AI scientist 'team' joins the search for extraterrestrial life

Following is a discussion about this article in Nature Vol 593 15 May 2025, by Celeste Beaver
To study the full text select this link: https://www.nature.com/articles/d41586-025-01364-w

The text in italics is copied from the article
Immediate followed by some comments

In the last paragraph I explain my own opinion.

Introduction

: Artificial intelligence (AI) researchers have created a system that can perform autonomous research in astrobiology, the study of the origins of life in the Universe.
AstroAgents comprises eight ‘AI agents’ that analyse data and generate scientific hypotheses. It joins a suite of other AI tools that aim to automate the process of science, from reading the literature to coming up with hypotheses and even writing papers.
The tool’s creators say they will use it to study samples that NASA plans to retrieve from Mars. The agents will help to determine whether the samples harbour organic molecules that indicate the presence of past or present life. The researchers presented AstroAgents on 27 April at the International Conference on Learning Representations in Singapore.
“It’s helping us build a better understanding of how molecules form in space, how molecules form from life on Earth and how they’re preserved — and then which specific signs should we be searching for,” says astrobiologist Denise Buckner at the NASA Goddard Space Flight Center in Greenbelt, Maryland, who co-authored a preprint describing AstroAgents1.

1. AI agents

: The tool is an example of ‘agentic AI’ systems. These are typically based on large language models (LLMs) and are designed to be more-active participants than conventional AI tools, deciding what needs to be done and how to do it, evaluating outcomes and adapting in response. Their emergence has prompted lively debate about whether agentic AI can come up with truly original scientific ideas, and how novelty should even be defined.
One of the most prominent examples is Google’s AI ‘co-scientist’, which was released in February and has searched for potential treatments for liver disease and suggested how antimicrobial resistance arises. Applying agentic AI to astrobiology is new, says astrobiologist Michael Wong at Carnegie Science’s Earth and Planets Laboratory in Washington DC.
To specify the behaviours of the agents, the researchers feed different prompts to an LLM. For example, a ‘data analyst’ is told to identify important patterns in data, a ‘planner’ to decide what to delegate to other ‘scientist’ agents for further research and hypothesis generation, and a ‘critic’ to evaluate the hypotheses and suggest improvements to the data analyst, which then kicks off another round of the process.
The way AstroAgents splits the hypothesis generation up between multiple specialist agents is innovative, says co-author Amirali Aghazadeh, a computer scientist at the Georgia Institute of Technology in Atlanta.
“We realized that because of the complexity of the data, it’s better for the agent to assign multiple tasks to multiple ‘scientists’,” he says. It’s up to the planner to decide what each scientist agent will study, and it does this on its own. “It’s kind of the magic of the system,” he says.

. Heaps of hypotheses


: The research team experimented with using two LLMs to power AstroAgents — Claude Sonnet 3.5 and Gemini 2.0 Flash. They fed each system mass-spectrometry data for eight meteorites and ten soil samples taken from locations around Earth, including Antarctica and the Atacama Desert in Chile, and carried out ten rounds of refinement. Astrobiology: Hunting aliens The result was 101 hypotheses from Gemini and 48 from Claude. One hypothesis posits that certain molecules found on Earth would make “reliable biomarkers” indicating the presence of life. Another suggests that a cluster of the organic molecules found in two meteorites might have formed through the same series of chemical reactions.
Buckner scored each hypothesis. She deemed 36 of the Gemini hypotheses to be plausible and 24 novel. By contrast, none of the Claude-generated hypotheses was original — but they were overall less error-prone and clearer than Gemini’s.
Better than humans? Buckner says that the volume of hypotheses generated and the ability to spot patterns in complex mass-spectrometry plots — which can represent the properties of hundreds of thousands of molecules — makes AstroAgents useful for research. “It’s going a step beyond what a person could do,” she says. Will AI improve your life? Here’s what 4,000 researchers think She looks forward to using AstroAgents to guide analysis of future samples, in particular from an ancient lake bed on Mars. These will be returned to Earth by a mission in the 2030s. “When we analyse those samples, it’ll better help us understand whether or not there’s potential evidence for life,” she says.
But Wong says that it’s not clear whether AstroAgents makes useful contributions, because just one person assessed its hypotheses. “The ratings given would be much more powerful if they had been a collection of, say, a hundred different experts’ scores,” says Wong. Even the hypotheses that received high scores did not teach him “anything new about the mystery of life’s emergence”, he says.
Aghazadeh thinks agentic AI tools will make meaningful contributions. “We are just starting and scratching the surface,” he says. “We agree lots of work should be done to understand origin of life, but agentic AI will play a key part.”

1. Chalenges