Gravitational Waves, Event Horizon Images, and Supermassive Black Hole Growth

Sam Black holes are probably the most famous thing in physics that almost nobody can actually picture correctly. So let's start at the foundation — what do we genuinely know a black hole is, versus what's still theory dressed up as fact?

Maya A black hole is a region of spacetime where gravity is strong enough that nothing, not even light, can escape once it crosses a boundary called the event horizon. That part comes straight out of Einstein's general relativity, published in 1916, and it's no longer speculative — the horizon itself has been directly imaged. What happens to matter and information at and inside that boundary, though, stays theoretical, and in places genuinely unresolved.

Sam That's a strange split to sit with — the edge is confirmed, but what's happening right behind it is still guesswork. Is that boundary between observed and theorized something that shows up everywhere in this subject, or mostly just at the horizon itself?

Maya It runs through almost every part of the subject. Take sizes: there are stellar black holes, formed from collapsing massive stars; intermediate-mass ones, roughly 100 to 100,000 times the Sun's mass; and supermassive ones, hundreds of thousands to billions of solar masses, sitting at galaxy centers. Nearly every massive galaxy we've observed has one. How the supermassive ones got so big so early is still an open problem.

Sam Before we get to imaging anything directly, how did we even know black holes were real for all those decades when nobody had a photo or a signal? What were astronomers actually pointing their instruments at?

Maya For decades, the only evidence was indirect — gravitational effects on visible matter nearby, stars or gas swinging around something unseen. That changed on 14 September 2015, when the Laser Interferometer Gravitational-Wave Observatory, LIGO, picked up its first gravitational wave signal, called GW150914: ripples in spacetime from two black holes, 36 and 29 solar masses, merging roughly six sextillion miles away.

Sam A century between Einstein predicting those ripples and anyone catching one feels like an enormous gap. What took so long, and what did that single detection confirm that indirect evidence couldn't?

Maya It confirmed black holes don't just sit there — they collide, merge into bigger ones, and release energy as spacetime ripples, just as the 1916 equations predicted but no instrument had confirmed until a decade ago. It was both the first direct detection of gravitational waves and the first direct observation of a binary black hole merger, turning a century-old prediction into measured data.

Sam That first detection wasn't just an opening act, then — it sounds like the whole field turned into a kind of production line afterward, with runs numbered and scheduled like seasons of something.

Maya That's close to how it works. There have been four observing runs so far: Run 1 from September 2015 to January 2016, Run 2 from November 2016 to August 2017, Run 3 from April 2019 to March 2020, and Run 4, which ran from May 2023 all the way through 18 November 2025 — the richest campaign yet, with detections happening roughly every two days instead of every five.

Sam Going from one merger every five days to one every two days is a steep jump. What changed in the instruments or the collaboration to make that kind of acceleration possible?

Maya The collaboration — LIGO, Virgo, and KAGRA together, known as LVK — set out deliberately to scale up, aiming for more than 200 events by the end of Run 4. They got there and beyond: roughly 300 black hole mergers have now been observed through gravitational waves in total, with 250 confirmed within Run 4 alone by the time it wrapped up.

Sam Three hundred mergers is a real catalogue to work with, not just a handful of highlights. Did anything in that pile of data surprise the people studying it, rather than just confirming what they expected to find?

Maya Two mergers in particular stood out. In October and November 2024, LVK detected GW241011 and GW241110, both with unusual black hole spins. GW241011 came from a collision between black holes of roughly 17 and 7 solar masses, with the larger one among the fastest-rotating ever observed; GW241110 involved black holes around 16 and 8 solar masses, much farther away.

Sam Fast-spinning black holes sound like they'd be hard to explain if you assumed every black hole just formed quietly from one dying star. What's the alternative story the collaboration is telling for how those spins got that way?

Maya Announced in October 2025, the interpretation is that these spin configurations point to hierarchical mergers — black holes living inside dense star clusters rather than as isolated pairs, merging, then merging again with whatever's nearby. That would make these second-generation remnants, products of earlier collisions rather than first-generation collapsed stars, which changes how crowded we think these environments are.

Sam Instead of two black holes forming separately and eventually finding each other, you're saying some of these are already the leftovers of a previous collision, sitting in a kind of cosmic mosh pit waiting for the next one. That reframes the whole population, not just two data points.

Maya It does, and it connects to a second finding — a gap in the expected mass range that gravitational waves have started to fill in. Stellar evolution predicted that a single collapsing star shouldn't produce a black hole within a certain range, the so-called pair-instability mass gap, because the physics of that collapse works differently there.

Sam A gap like that sounds like a prediction daring nature to prove it wrong. Did anything ever turn up inside that supposedly forbidden range?

Maya GW190521, detected in May 2019, produced a remnant of about 142 solar masses from two merging black holes, the heavier one weighing 85 solar masses — squarely inside that gap, and the first black hole ever found there. It was the first conclusive detection of an intermediate-mass black hole, and it directly challenged the stellar evolution models that said it shouldn't exist.

Sam So the gap wasn't a wall after all, just a place single stars can't reach on their own — but mergers clearly can get there some other way.

Maya A more recent signal, GW231123, observed in November 2023 and announced in July 2025, added another intermediate-mass merger to that picture. And at the opposite end of the scale, GW230529, picked up by a single LIGO detector, gave the best evidence yet for objects sitting in the lower mass gap, between the heaviest neutron stars and the lightest black holes.

Sam Now we've got a direct image of the boundary itself, from both M87 and Sagittarius A-star. But the thing everyone wants to know — what's inside — still isn't something a telescope can touch, no matter how good the resolution gets.

Maya That's exactly the limit. The images confirm the horizon exists and that light bends and gets trapped the way Einstein's equations predict, but they say nothing about whether a true singularity — a point of infinite density — actually forms inside. That stays theory, and it's where Hawking's work from 1974 comes in.

Sam Stephen Hawking arguing that black holes leak radiation sounds like a direct contradiction of the "nothing escapes" rule we started with. How does quantum field theory get around a boundary that general relativity says is absolute?

Maya The usual picture is virtual particle pairs forming at the horizon, one falling in while the other escapes as radiation — but that's a simplification, not the real calculation. The deeper reason is that a black hole's spacetime doesn't have one single agreed-upon vacuum state, so what looks empty from one perspective contains outgoing particles from another, and the resulting spectrum comes out looking thermal.

Sam So the black hole isn't spitting out captured matter, it's generating radiation out of a disagreement about what "empty" even means near the horizon. That's a strange kind of leak — nothing is really escaping, something is being created at the edge instead.

Maya That's the shape of it, and the consequence is that the black hole loses mass over time and eventually evaporates, incredibly slowly. But for any realistic astrophysical black hole, the predicted temperature is so low that the radiation would be undetectable with any instrument we can currently build or plausibly build soon.

Sam If it's untestable for anything we've detected, that sounds less like a prediction and more like a thought experiment that happens to use real physics. Is there any object small enough that this would ever become observable?

Maya Only a primordial black hole — a hypothetical class formed in the extremely early universe rather than from a dying star, small enough to run hot enough to radiate detectably. None have ever been found, and no direct trace of Hawking radiation exists anywhere. It remains pure theory, but one of the only places general relativity and quantum mechanics make contact at all.

Sam And that contact point seems to be exactly where things get messy, because if a black hole eventually evaporates completely into thermal radiation, whatever fell in during its life should leave some trace behind. Otherwise it sounds like information just gets erased from the universe.

Maya That's precisely the crisis. Quantum mechanics requires unitary evolution — information is never truly destroyed, only scrambled. But if the outgoing radiation is purely thermal, with no structure encoding what fell in, that principle breaks. The honest answer, even now, is that nobody has resolved whether information is destroyed or somehow encoded onto the radiation as it leaves.

Sam That's an uncomfortable place for a fifty-year-old theory to still be sitting. Has anyone found a plausible way to make the bookkeeping work, where the radiation is secretly carrying the information out without looking any different to us?

Maya There are several attempts rather than a settled answer. A 2023 study suggested Hawking-type radiation can form outside the horizon even if the horizon itself only blocks light from escaping. Holographic calculations have reproduced a feature expected of unitary evolution — work Netta Engelhardt at MIT has pointed to as progress — and a December 2024 study proposed that quantum correlations might leave a subtle signature in gravitational waves, testable by future detectors.

Sam A signature in gravitational waves would be remarkable. It would mean the paradox stops being philosophy and starts being something you could put through a detector.

Maya It would, but multiple 2024 papers are plain that the paradox hasn't been conclusively solved. The broad expectation among physicists is that information survives in some form, but that remains an expectation rather than a demonstration — one of the clearest open questions sitting right at the edge of what these observations have confirmed.

Sam Let's push on the supermassive ones. Confirming they exist is one thing — but how did anything get that big in the time the universe has had available?

Maya That's the open puzzle. We know supermassive black holes exist and we've measured their masses across hundreds of galaxies, but how they reached those masses so quickly is unresolved. The basic question is whether they started as ordinary stellar-mass remnants from dying stars and slowly grew, or whether some mechanism produces much heavier "seeds" right from the start.

Sam A seed sounds like it's meant to shortcut the problem — start bigger, so you don't need as much time to bulk up. What's pointing toward that idea rather than just slow accumulation from a small remnant?

Maya The clearest case is a black hole called UHZ1, found using the Chandra and JWST telescopes. Its data shows it would need to grow continuously, faster than the theoretical accretion limit called the Eddington limit, for more than 200 million years straight — or it started from a seed that was already massive. Sustained super-Eddington growth that long isn't considered likely, which pushes the evidence toward a heavy seed.

Sam Two hundred million years of non-stop overfeeding does sound like a stretch for something that's supposed to be a physical limit. Has anything more recent added to that case, or is UHZ1 still the sharpest example on its own?

Maya JWST confirmed in 2025 an actively growing supermassive black hole in a galaxy just 570 million years after the Big Bang, part of a population JWST keeps finding called "Little Red Dots" — compact, very red objects turning up in far larger numbers than expected. Some look "overmassive" for how little stellar material their host galaxies have, which connects these early objects to the luminous quasars we see later and strains current galaxy-formation theory.

Sam Overmassive relative to its own galaxy sounds backwards, like the black hole got ahead of the galaxy that's supposed to be growing around it. Is direct collapse the leading explanation for how that gap opens up, or just one guess among several?

Maya It's one candidate among several, not a settled answer. The idea is that massive clouds of cold gas at the center of early galaxies collapse straight into a black hole, skipping the dying-star stage entirely. One study found this could happen as early as 13.64 billion years ago, less than 500 million years after the Big Bang, inside overdense, rapidly merging clumps of dark matter.

Sam So instead of a star collapsing down to something small and then slowly bulking up, you'd get a massive seed straight out of gas, no stellar middleman at all. That's a genuinely different birth story, not just a faster version of the usual one.

Maya It is, and the broader synthesis is blunt about it: these giants are far too massive to have formed from a single dying star, and scientists think most likely grew from smaller seeds by consuming gas and merging, though some may have come from direct collapse of gas clouds instead. Both tracks are live, and nothing has settled which dominates.

Sam That seed language keeps coming up, and it makes me think of the in-between category you mentioned earlier. Has finding more of those helped narrow down which growth story is right?

Maya It's helping, and the numbers have moved fast. Before mid-2024, only about ten intermediate-mass black holes had been found anywhere, despite heavy searching, and the working idea was that they formed early and act as seeds for supermassive ones through repeated merging. A 2024 study led by Florian Peißker found a second candidate in a star cluster called IRS 13, just 0.1 light-years from Sagittarius A-star.

Sam From ten total to a second one found that close to the galactic center sounds like the floodgates are starting to open. Did that trend keep going, or was that mostly a one-off discovery?

Maya It kept going. Astronomers confirmed a roughly 100,000-solar-mass intermediate black hole in a stripped galactic nucleus in Andromeda, and found another, NGC 6099 HLX-1, caught disrupting a star in a satellite cluster — described as key for probing these elusive objects through rare, extremely luminous X-ray sources. Then a 2025 survey using the Dark Energy Spectroscopic Instrument found around 300 candidates, tripling the roughly 150 known before.

Sam Tripling the known population in one survey makes it sound like these things were never rare, just badly hidden from the instruments we had. Does that change the seed-versus-collapse debate, or just confirm there's more raw material to study either way?

Maya Mostly the latter — it suggests we'd been missing a substantial number of low-mass black holes, which matters for the census but doesn't settle which formation pathway dominates. Both the seed-growth and direct-collapse pictures are still actively debated, right through the 2025 and 2026 JWST results, alongside the still-unconfirmed hierarchical merger populations and the still-untested Hawking radiation sitting at the theoretical end of the whole subject.

Sam After everything — the imaged horizons, the gravitational wave catalogue, the growth puzzle we just went through — if you had to leave the listener with one thing to carry out of all this, what would it be?

Maya The one thing worth holding onto is that black holes have moved from pure prediction to directly observed objects — their horizons imaged, their mergers measured in gravitational waves — while almost everything about their interiors, their earliest formation, and how quantum mechanics and gravity ultimately fit together there remains open. The honest state of the field is a lot of hard data at the edges and real uncertainty about what's happening just past them.

Sam That's a good place to leave it — the edge is solid, the inside is still unfinished business. Thanks for walking through all of it.

Maya Thanks for pushing on every piece of it the way you did, and it's worth saying plainly before we go: this show is AI-generated, the voices you've heard are synthetic, and the facts came from the source reporting named in the closing note, whose authors have no connection to this show.

Sam Everything today came from reporting by arxiv.org, space.com, the LIGO Caltech collaboration, phys.org, Wikipedia, and Big Think — thanks for listening.