Leading Yale astrophysicist Priyamvada Natarajan proved that black holes can form from unstable gas. Now she is observing the invisible universe using new technologies.
In popular fiction and science fiction, black holes are places of intense gravity, sort of like cosmic vacuum cleaners that suck up everything around us. They form when stars explode, becoming supernovae and leaving deep holes in the structure of the universe. These are hard to understand, even for Priyamvada Natarajan, a professor of astronomy and physics at Yale University, who has studied the phenomenon for decades.
In 2006, Natarajan proposed a completely different idea of ​​how black holes formed in the early universe. Not from the explosion of a star, but from the direct collapse of gas. He theorized that gas in the early universe became unstable and funneled into the center very rapidly, like pulling a plug in a bathtub. This bathtub action instantly created supermassive black holes, each 10,000 times the mass of the Sun. In late 2023, two space telescopes proved all of his predictions and his theory correct. Today, we know about direct collapse black holes and an ancient galaxy UHZ1 thanks to this.
We caught up with him at the IndiaSpora 2026 conference in Bengaluru to understand his fascination with invisible space, the competition for space telescopes and how artificial intelligence is reshaping his work as an astrophysicist. Edited excerpts:
Until now we have assumed that stars explode, become supernovae and then leave behind black holes many times the mass of the Sun. Why were you looking for another way for black holes to form in the universe?
We need to think in a different way about how black holes form than when stars explode because we were seeing black holes millions of times the size of the Sun in the early universe and there was no explanation for it.
My calculations show that in the early universe, massive black holes may have formed through bathtub action where the gas becomes unstable and stars become trapped at the center very rapidly. Thanks to sophisticated computers, we were able to make a solid prediction in 2017, theorizing what observations of the real universe would look like if such a black hole existed.
Your prediction was made based on data that could be observed by the largest telescopes at the time available to you. James Webb, 1.5 million kilometers away in space, observes the infrared spectrum to look at the early universe. and the Chandra X-ray Observatory, which is in low Earth orbit and observes X-ray emissions. Why did you need these two observations to prove your prediction?
When the gas heats up and glows around the black hole’s event horizon, it can be seen in X-rays or in the optical or ultraviolet. Seeing X-ray emission from the center of a galaxy is often a very good indication that you have a black hole there.
However, much of what was emitted in the optical early in the universe will be visible to us today in the mid-infrared range because the wavelengths of light will have spread as our universe expands. James Webb has cameras that detect infrared radiation. Our prediction was that for these black holes that formed in the early universe, you should see them not only in X-rays, but also in infrared radiation. So to prove our theory that black holes can be created by this method, the same black hole needed to be observed simultaneously by Chandra X and James Webb.
We predicted six different unique properties, spectrum, amount of energy released, even a shape. Only if all six were satisfied with an object, you can clearly say that there is strong evidence that this was a new way in which black holes were created.
A total of four space telescopes are available to humanity. How did you convince two of them to look in the same direction of the universe to prove your prediction of black hole formation?
As scientists, you compete with each other for data allocations from these telescopes. You submit your proposal and it is evaluated anonymously by peers. The most interesting idea wins. For our prediction, we needed to observe James Webb in a single direction for several hours, but with Chandra, we needed several days of observations because it is capturing faint and distant objects. When the black hole we predicted was found, Chandra had been recording the same part of the sky for 24 days.
You thought about this possible early universe black hole idea 20 years ago. All six predictions were proven in November 2023 by data from both James Webb and Charles. How did you feel about this recognition?
It was incredible. My lab partner sent me the real universe data but joked that it was just one of the models we developed. I actually got in on the joke because the data was very similar to what one of the models we predicted predicted. This rare moment when everything falls into place is almost magical. I think I cried, because it’s every scientist’s dream to have something they’ve predicted mathematically be proven with real data. And this should happen during his lifetime.
Is the differently formed black hole you discovered named after you?
(laughs) It’s called UHZ1 but it will be on my inscription.
What surprised you the most after this?
The kind of recognition I received for the work was completely unexpected. Awards within astrophysics and fellowships were expected but there was a lot of media attention. I was on the Time 100 list of the most influential people in the world, I thought this email was spam. But I was already busy publishing more papers, getting my work reviewed by peers.
You work on the invisible universe, dark matter, dark energy and black holes. One cannot see or image these entities. What are attractions and how can you guess if they exist?
Light is our cosmic messenger, but dark matter or dark energy does not interact with light at all. Their visible absence means that they are physically present. I find them intellectually very fascinating because you have to see these things indirectly to make inferences about them, to understand their nature. It’s like detective work.
Dark matter, which I am currently studying, deflects light. Light rays travel across the fabric of our universe, bouncing up and down matter and dark matter creating cracks or holes in this fabric. We physicists record these signatures of deflected light to understand dark matter. In the case of a black hole, only light beyond the event horizon is absorbed. Anywhere outside the event horizon, you can receive signals and radiation. For example, when gas is drawn into a black hole, it heats up before it reaches the event horizon. It shines. This is how we observe, measure, and map black holes to study them.
How has astrophysics over the past two decades enhanced our understanding of the black holes at the centers of galaxies?
Because black holes can affect the structure of the universe, most black holes, especially supermassive black holes, which are a million times the mass of the Sun, are found at the centers of most, if not all, galaxies. The past two decades have given us a new understanding of how black holes play a very important role in shaping the galaxies in which they are located. Because black holes heat the gas before they cross the event horizon. This is interesting because the gas has to cool, condense, and become a substance. In this way, a black hole acts as a switch to turn the formation of stars on and off. They control visible matter in the galaxy, from stars to planets.
What are you working on now?
I am exploring how dark matter interacts with black holes as they both appear to exist in the centers of galaxies. Another direction I am exploring is the meaning of dark energy, which is not yet understood but is responsible for the rapid expansion of our universe. Both of these are open frontier questions.
How have evolving technologies like telescopes and now AI, helped astrophysics?
New developments in charge-coupled instruments that use highly sensitive silicon chip detectors to capture light provide us with more detailed images of faint, distant astronomical objects. In space telescopes, weight matters, so it is important to make imaginative components lighter and more compact. Recently, GPUs have accelerated our computing capabilities. Many simulations took a year to run on a supercomputer. Now we can simulate thousands of possible scenarios simultaneously.
You think AI could fundamentally change the nature of search? Why?
The big LLM models have digested everything that is published in our field, right? Therefore they are equipped with the expertise of peer-reviewed published material. Can an AI agent consider a new idea more precisely than a group of scientists? Can it identify our blind spots? Could it be possible to become a peer reviewer in the future? Can this really work in the laboratory, in the mess of science? Can an AI really write a scientific paper from start to finish, generate ideas, develop them, write them in a coherent manner? We are certainly in a time of exciting intellectual upheaval. We do not fully understand how AI will change all dimensions of our reality.
