AI uncovers Earth’s oldest life: 3.3-billion-year-old traces found in Ancient Rock

Oldest Life on Earth Identified by AI in 3.3-Billion-Year-Old Geological Time Capsule (Image: Canva)

A quiet signal from ancient rocks offers new insight. Researchers say faint chemical traces now reveal early life on Earth. Their method studies old molecules that survived deep time.

How were these ancient signs of life detected?

Scientists analysed rocks from South Africa that hold early history. These samples date back almost 3.3 billion years. The team identified chemical fingerprints linked to microbial activity. They used machine learning to spot patterns hidden within fragments. The study appears in Proceedings of the National Academy of Sciences. Carnegie Institution for Science researchers led the investigation.

Robert Hazen, a co-lead author, explained the method. He said the team extracted carbon-rich molecules for testing. He added that machines read patterns the eye cannot see. The approach separates biological and non-biological molecules with high accuracy. Hazen said the signals resemble whispers left by ancient cells.

What do the chemical signatures reveal about early microbes?

The team found traces tied to oxygen-producing photosynthesis. These signs appear in rocks about 2.5 billion years old. The discovery suggests marine bacteria used sunlight much earlier. This activity helped oxygen fill Earth’s ancient atmosphere. It also supported later evolution of complex aerobic life.

The oldest known physical fossils date to 3.5 billion years. These include stromatolites from Australia and South Africa. However, such fossils are extremely rare and incomplete. Chemical evidence offers a different window into early biology. Many original biomolecules like fats and sugars are long gone. Only small fragments remain, yet they still carry chemical patterns.

Why does this new method matter for astrobiology?

Study co-lead author Anirudh Prabhu said the technique changes timelines. He said their work doubles the age of detectable biosignatures. It also distinguishes different microbial types in ancient rocks. Machine learning helps identify life signals within damaged molecules.

NASA has funded further development of this technique. Rovers on Mars already collect samples for future missions. Scientists hope these methods can test Martian rocks for life. They also aim to study material from Enceladus and Titan. Europa remains another key world for possible chemical clues.

Hazen said the team hopes to analyse returned samples soon. He added that the approach may guide future rover missions. The researchers believe this tool can support planetary science. It may help determine where life once lived beyond Earth.