Are we alone in the universe? Consider mysterious ”extraterrestrial” radio signals. Unexplained gases on other planets. Things mistaken for UFOs.
In the search for alien life, humans have been fooled more than once by signals that turned out to turned out to be false alarms.
Now, researchers at Michigan State University show that artificial intelligence can fall for alien hoaxes, too.
The study, co-authored by Ankit Gupta and Christoph Adami, finds that current AI models can be duped into seeing signatures of life where none exist, spitting out wrong answers with alarming confidence.
That could spell trouble for the next big multibillion-dollar space missions to look for life beyond Earth, the researchers say.
The team will present their findings in August at the 2026 Conference on Artificial Life in Waterloo, Canada.
While definitive evidence for extraterrestrial life has yet to be detected, scientists continue to look. And some are betting on AI — with its ability to process vast amounts of data and identify patterns — to expedite the search.
A number of current and future NASA missions plan to search for signatures of past or present alien life, from rovers drilling into Martian soil and spacecraft exploring the moons of Saturn and Jupiter to telescopes probing the atmospheres of planets beyond our solar system.
Once there, AI-powered sensors and other equipment will analyze various samples to look for features that suggest they could have come from living things.
There’s no one smoking gun biosignature that could be used to say there’s life out there. But there are certain universal features that are pretty good indicators, said Adami, who is a core faculty member in MSU’s ecology, evolution and behavior program.
“One of them is that life needs to encode information,” Adami said.
Typically this takes the form of chain-like molecules like DNA that can replicate themselves.
So Adami and Gupta did an experiment where they created artificial forms of life using a computer program called Avida, and then trained an AI to detect them.
In the Avida universe, digital organisms written in computer code — essentially strings of commands — copy themselves over and over again in a virtual Petri dish inside the computer. Each time they replicate, the copying process is imperfect and their computer code changes, just like the genetic code of real organisms mutates when they reproduce.
Such forms of “digital life” have been used for several decades to study evolution, Adami explained.
For the study, the researchers used Avida to generate tens of thousands of digital organisms, some of which contained the instructions needed to copy themselves and others not. They then used them to train a neural network to distinguish between the two with 99.97% accuracy.
However, when the researchers put the neural network to the test on examples it had never seen before, the results looked far less impressive.
In their experiments, the researchers started by presenting the neural network with a digital organism that the AI correctly deemed incapable of copying itself. Then by gradually swapping out one operation for another in the organism’s computer code, the researchers were able to trick the AI into misclassifying the organism as self-replicating, in as few as 150 tries.
In other words, with just a few tweaks, the team showed that it was possible to convince the AI that it was seeing signs of life where they didn’t exist.
No matter what sequence of commands we started with, we were able to fool the AI 100% of the time, said Gupta, a Ph.D. student in computer science and engineering at MSU.
The number of sequences that could potentially trip up the network is vast. “So the likelihood of encountering such a sequence is substantial,” Adami added.
This risk of false positives raises the likelihood that space missions could be fooled into thinking they’d made a discovery by the data they gather, only to be proven wrong later.
Having trained an AI on computer-generated data, next the researchers plan to re-train the model with real-world data and see how easy it is to deceive, Gupta said.
The findings underscore a known weakness in many contemporary AI models.
“AI has an Achilles heel,” said Adami, professor in MSU’s departments of microbiology and molecular genetics and physics and astronomy. “It can see a pattern and completely misclassify it.”
Beyond the search for alien life, this weakness could also be problematic as AI continues to move into medical scanners, security cameras, self-driving cars and other devices in everyday use here on Earth.
This doesn’t mean it’s futile to use such methods in astrobiology, medical diagnosis, on the battlefield or other situations. It just means “you need an independent way of checking their work,” Adami said. “There needs to be a human in the loop.”
But in many situations that’s not the case, he added. Take, for example, an AI-powered sensor onboard a Martian rover that must analyze hints of alien life on the spot, before the samples return to Earth.
“It’s a very serious vulnerability,” Adami said.
This story originally appeared on the Ecology, Evolution and Behavior website.
