Posted by In April 1960 Frank Drake spent his nights at Green Bank, West Virginia, pointing a 26-metre dish at two nearby Sun-like stars, Tau Ceti and Epsilon Eridani, and listening near the 1420 MHz line of hydrogen. He called it Project Ozma, after the princess of the land of Oz. One early detection got his pulse racing and turned out to be a military aircraft. After about 150 hours he had found nothing, which is roughly how every serious attempt to find aliens has gone since. The difference is that today’s searches are millions of times more thorough, and they no longer rely on radio alone.
Why radio came first
The logic goes back to a short paper by Giuseppe Cocconi and Philip Morrison in Nature in 1959. Radio waves cross interstellar distances with little absorption. The galaxy is relatively quiet at certain frequencies. And hydrogen, the most common element in the universe, emits at 1420 MHz, a natural landmark any radio astronomer anywhere might know. If someone wanted to be found, the authors argued, that neighbourhood of the spectrum would be a sensible place to shout.
What SETI researchers look for is not a voice or a picture. It is a narrowband signal: energy squeezed into a sliver of frequency only a few hertz wide. Stars, quasars and gas clouds broadcast across wide bands. Narrowband emission, as far as anyone knows, comes only from transmitters.
The trouble is that Earth is full of transmitters. Phones, satellites, radar, aircraft altimeters, even faulty electronics inside an observatory. Every modern search spends much of its effort throwing those out.
It is also worth asking what we ourselves would look like from a few light-years away. Television and FM radio leak into space, but those broadcasts are weak and spread across wide bands, so a copy of our own telescopes at the distance of even the nearest stars would struggle to notice them. The strongest beams humans have ever sent outward were planetary radar pulses, such as those from Arecibo before it collapsed in 2020. They were intense but brief and pointed at particular asteroids or planets. A civilisation like ours, in other words, would be easy to miss.
Bigger dishes, bigger data
Breakthrough Listen, launched in 2015 with a pledged $100 million over ten years, changed the scale. It bought time on the 100-metre Green Bank Telescope and on Parkes in Australia, later adding the MeerKAT array in South Africa, and recorded enormous amounts of data across billions of channels at once. The Allen Telescope Array in northern California, a field of small dishes built with money from Microsoft co-founder Paul Allen, was designed partly with SETI in mind. China’s FAST, a single 500-metre dish set in a natural limestone basin in Guizhou, has run its own searches.
All of this produces too much data for people to inspect by eye. In 2023 a University of Toronto student, Peter Ma, and colleagues trained a machine-learning model on Breakthrough Listen observations of 820 stars. It flagged eight signals of interest that earlier methods had missed. Follow-up observations of those stars found nothing, so the eight stay as candidates that didn’t repeat. The method mattered more than the result: algorithms can sift, but they cannot confirm.
Looking for flashes instead
In 1961, only a year after the first working laser, Robert Schwartz and Charles Townes, one of the laser’s inventors, suggested that another civilisation might signal with light. A powerful laser pulse lasting a few billionths of a second could briefly outshine its own star in that one colour, at least from the right direction. No natural source is known to produce such short, bright flashes.
Optical SETI has been running in fits and starts since the late 1990s. A newer project called PANOSETI, being developed for Lick Observatory in California, aims to watch a large part of the night sky at once for nanosecond flashes in visible and near-infrared light. It is a strange kind of astronomy. You are not studying anything. You are waiting for something to blink.
Megastructures and a famous false trail
Freeman Dyson pointed out in 1960 that a civilisation using a large fraction of its star’s energy would have to dump waste heat, which should glow in the infrared. A star surrounded by collectors would look dimmer in visible light and oddly bright in infrared. Searches for such “Dyson spheres” have combed infrared sky surveys ever since. A 2024 study using data from Gaia and the WISE mission flagged seven possible candidates among millions of stars, and other researchers quickly argued that background galaxies could explain at least some of them.
Then there was Tabby’s Star. KIC 8462852, spotted by volunteers in the Planet Hunters project and described in 2015 by Tabetha Boyajian, dimmed irregularly by as much as 22 percent, with no planet-like pattern. Speculation about alien construction projects spread fast. Follow-up observations showed that the dimming was deeper in blue light than in red, which is what fine dust would do and a solid structure would not.
That episode is a decent model for the whole enterprise. Weird is cheap. Explaining weird is the work.
Signs a planet might give off
Another strategy skips signals entirely and looks at whole planets. Industrial pollutants such as chlorofluorocarbons have no known natural source, and a few researchers have estimated that large amounts in an exoplanet’s air might, in favourable cases, be detectable by future telescopes. Nitrogen dioxide from combustion has been proposed too. Artificial lighting on a planet’s night side would be far harder to spot.
These ideas are speculative. They also have the advantage of not depending on anyone choosing to talk to us.
When a candidate shows up
Every one of these methods produces candidates, and almost all of them die quickly. The routine is the same whatever the wavelength. Does the signal stay fixed to one point on the sky as Earth rotates? Does it disappear when the telescope nods away to a nearby patch and return when it nods back? Does a second instrument, with different electronics in a different country, see it as well? Can every satellite, aircraft and local transmitter be excluded?
Only a candidate that passes all of that starts to deserve attention outside a research group. None has so far. That is not embarrassing. It is roughly what you would expect from a search that is still young, and it means the filters are doing their job.
Where to aim
Since nobody knows where to look, strategies to find aliens differ. Some surveys target the nearest stars, others the systems with known temperate planets, others sweep broad strips of the galactic plane where stars are densest. One clever approach flips the geometry: Lisa Kaltenegger and Jackie Faherty identified, in 2021, more than a thousand nearby stars from which Earth would have been seen passing in front of the Sun at some point in the last few thousand years. If anyone there studies planets the way we do, they could know about us.
Will any of this find aliens? Nobody can promise it. Searches so far have covered a tiny fraction of the stars, frequencies, signal types and moments in time that are possible. Silence from a small sample tells us surprisingly little.
What it does tell us is how to search better, and that is not nothing.
SETIworld covers this hunt across every wavelength, from Green Bank to the newest optical projects, and if you want to follow the next candidate as it is tested, the portal is the place to keep reading.
