
In a recent essay, posted here, I proposed to transform SETI to SETAI, namely to conduct a Search for Extraterrestrial Artificial Intelligence. How can we distinguish extraterrestrial artificial intelligence (AI) from extraterrestrial biological intelligence?

It would be easiest to do so by retrieving an alien spacecraft and examining its content. David Grusch testified under oath that contractors received from the U.S. government intact alien spacecraft with biological materials in them. But even in that case, it would be prudent to check whether the biological materials are a product of synthetic or natural biology. Our own AI is made of silicon chips, derived from sand. But in the distant future, we might produce AI by 3D-printing biological neural networks.

Alternatively, we could search remotely with our telescopes for AI fingerprints on exoplanets. Receiving an electromagnetic signal from sources located thousands of light years away does not allow us to engage in a meaningful dialog to clarify this matter, because each exchange takes millennia. A Turing Test is impractical across interstellar distances. We need an alternative.

The simplest remote signature is thermal. AI data centers produce a lot of heat. If AI takes over biological creatures during the technological evolution of an extraterrestrial civilization, then the host planet might overheat from the power released by its data centers. These circumstances could lead to a hotter surface than expected from the illumination by the host star. Such a technological fingerprint is a variant on the Dyson sphere infrared signature of a megastructure around the host star, albeit it is confined to the surface of the host planet.
In such a case, the planet would reside in the habitable zone of the star — because the AI was constructed by biological entities early in the history of the planet, but its surface temperature might exceed the upper limit beyond which the atmosphere would evaporate and no liquid water would survive. If this infrared fingerprint is accompanied by an electromagnetic signal or some other signature of an advanced technology, one might be able to conclude that the intelligence is artificial since biological life is not possible on the surface of this hotter than expected exoplanet.

In principle, AI robots could migrate and self-replicate on planets or asteroids outside the habitable zones of stars, since they could use silicon in rocky material and are not constrained by natural evolution to use the nutrients feeding the chemistry of biological entities in liquid water.

A few weeks ago, Elon Musk said that approximately a million Optimus robots and roughly a gigawatt of solar power would “constitute the first Von Neumann Probe, a system capable of replicating itself entirely from local materials on any sufficiently large planet or moon.”
Musk described a self-replicating vision for Tesla’s own factories closer to home, claiming that robots could build other robots, effectively removing humans from major parts of the production line. Despite the bold rhetoric, it is important to note that Tesla remains very far from this reality. The company is still in the foundational phases of installing its very first traditional manufacturing lines for Optimus. Tesla has historically missed consumer deployment timelines for the robot, and true autonomous self-replication remains a highly experimental, long-term conceptual goal.
However, over cosmological timescales of billions of years, it is certainly possible that an advanced AI-dominated civilization would expand well beyond the habitable zone of its planetary system to occupy planets at either smaller or larger distances from the host star. Smaller distances would be more desirable as they offer a larger power supply per unit area, as the flux of radiation from the host star scales inversely with distance squared. Proximity to the star offers a great benefit for a power-hungry AI infrastructure to settle on the innermost planets, like Mercury or Venus in the Solar System, as long as the associated electronics will not be damaged by the excessive heat.
Identifying technological signature from planets interior to the habitable zone of stars constitutes an important branch of SETAI. Discovering such signatures could validate Musk’s vision for a future to humanity which is dominated by self-replicating AI robots. Related data can be obtained by observing planetary systems at great distances.
Closer to home, the UAP Science Advisory Council might be able to pursue SETAI faster, through a guided tour to an alien spacecraft on Earth.

In this essay, I featured seven amazingly beautiful watercolors from a series created and sent to me by the celebrated artist, Greg Wyatt.
Authorized publication by Dr.Avi Loeb
Ler em Português