The Jerusalem PostGermany's left-wing Die Linke struggles to form coalition as anti-Israel controversy swirlsNOSWereldleiders eenstemmig bezorgd, maar conflicten drukken stempel op klimaattopMexico News DailySheinbaum calls NYT report on CIA’s alleged bargain with Chihuahua prosecutor ‘very interesting’: Wednesday’s mañanera recappedConsequenceNeurosis Announce Summer 2027 UK/European Tourynet כלכלהרגע לפני ההנפקה: האקסית של אריק שמידט דורשת 5% מהחלק שלו באנתרופיקynet"חנק עם שרשרת, עד שנקרעה": נאשם כי תקף וגרם לשברים ביד - בגלל צפירהComplete SportsEUROMATCH NPFL: 3SC Seek Fourth Straight Home Win Over Bendel InsuranceZDF heuteAktuelle Pressemitteilungen des ZDFDaily MaverickCONSTITUTION @ 30: From equal rights to the death penalty, seven Constitutional Court judgments that shaped SABuenos Aires TimesPope Leo visits prisoners, urges them to ‘restart their lives’ColliderNetflix’s 16-Part Hidden Gem Officially Makes Its Mark on StreamingBusiness AMCEO van Nexperia sluit hereniging met Chinese dochterondernemingen niet uit
The Daily Newsstand · Free, Always
Wednesday, November 11, 2026

Scientists check their work and confirm it with update on iconic black hole image

Translate

The image of the red ring around the black hole M87, produced by the Event Horizon Telescope (EHT) array, made front-page news around the world in 2019. There was one problem: there was a slim chance the depiction might be wrong.

This week, a small team of four Australian researchers have published their work confirming the result based on a new machine learning algorithm using the same base data.

The 2019 Event Horizon Telescope (EHT) image came from hundreds of scientists working together to capture the first direct view of a black hole's shadow. The measurements were taken at eight telescopes around the world, at locations including Chile, the US, and Antarctica.

REG AD

Reconstructing the image from the data so geographically distant meant instrument and environment errors could influence the accuracy of the resulting image. Potentially, alternative analyses of the data could have reconstructed an image without the iconic ring shape.

REG AD

Samuel Lai – a postdoctoral fellow at the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Western Australia – and his team created an independent imaging algorithm to process the historic dataset to validate the result. The new approach used closed triplets (using three telescopes for error correction) rather than station pairs in an effort to reduce error. 

To the relief of the EHT team, the new image recreated the iconic ring-like image seen in the original, according to a paper published in Nature today. 

In a prepared statement, Lai said taking pictures of black holes was not easy. "It is like standing in your backyard and trying to take a photo of a doughnut on the Moon," he said.

Glowing orange ring centered against a black background

The new, composite image reconstruction of the black hole M87 several datasets from the Event Horizon telescope
Pic credit: Lai et al, Nature

He explained that when his team applied new advanced machine learning and artificial intelligence techniques to the original Event Horizon Telescope dataset, they found a black hole almost identical in size and shape to the historic EHT image.

"Importantly, our model had no idea what a black hole looked like when we applied it to the M87 data. This is the first truly independent validation of the EHT image, an important step in the scientific process," he said.

A new shot of the M87 galaxy’s supermassive black hole. Credit: EHT Collaboration

The historic 2019 image estimating/ reconstructing the polarized light from M87's supermassive black hole ...
Credit: EHT Collaboration

Event Horizon Telescope director Laurent Loinard welcomed the new study. "It is extremely gratifying to see that an elegant, entirely independent method recovers the ring structure of M87 across several years, in excellent agreement with the EHT Collaboration's published results," he said. ®

View the original on The Register →

KioskNews shows a cleaned-up reading view extracted from the publisher’s page — the original always lives on their site, not ours.