News Feeds

A film made of lithium niobate and gratings of silicon dioxide converts infrared light into visible light better than the other leading compound, potentially allowing nighttime vision

Five children have had gene therapy to treat inherited deafness, this time in both ears, following the success of earlier treatments in just one ear

On January 19th, 2024, the Japanese Aerospace Exploration Agency (JAXA) successfully landed its Smart Lander for Investigating Moon (SLIM) on the lunar surface. In so doing, JAXA became the fifth national space agency to achieve a soft landing on the Moon – after NASA, the Soviet space program (Interkosmos), the European Space Agency, and the China National Space Agency (CNSA). SLIM has since experienced some technical difficulties, which included upending shortly after landing, and had to be temporarily shut down after experiencing power problems when its first lunar night began.

On the Moon, the day/night cycle lasts fourteen days at a time, which has a drastic effect on missions that rely on solar panels. Nevertheless, SLIM managed to reorient its panels and recharge itself and has survived three consecutive lunar nights since it landed. However, when another lunar night began on May 27th, JAXA announced that they had failed to establish communications with the lander. As a result, all science operations were terminated while mission controllers attempt to reestablish communications, which could happen later this month.

As JAXA stated via its official X account (formerly Twitter):

“We tried again on the night of the 27th, but there was no response from #SLIM. As the sun went down around SLIM on the night of the 27th, it became impossible to generate electricity, so unfortunately this month’s operation will end. Thank you very much for the overwhelming support you have shown us since our post the day before.”

27??????????????????#SLIM ???????????????27??????SLIM??????????????????????? ?????????????????????????????????????????????????????#JAXA

— ????????SLIM (@SLIM_JAXA) May 28, 2024

JAXA further indicated that the command transmission to restore communication was performed using an “unplanned ground station antenna” and with the cooperation of JAXA’s tracking network.” They also indicated that they plan to try reestablishing communications once the current lunar night ends later this month – at which point, they expect the lander will be recharged. “The power was turned off overnight, so we hope that the whole system will be reset and restarted,” they wrote.

The SLIM mission also carried two rovers, which separated from it in lunar orbit and landed independently on the same day. Known as the Lunar Excursion Vehicle-1 and -2 (LEV-1 and LEV-2), these rovers are the first Japanese robotic missions to traverse and explore the lunar surface. According to JAXA, LEV-1 is the world’s first “hopping exploration rover” while LEV-2 is the world’s smallest and lightest. During the four months since they landed, LEV-1 has measure the local temperatures, topography, and taken images.

The rovers can conduct operations autonomously and transmit data to Earth directly without assistance from the lander. As such, JAXA’s mission controllers are still likely to hear from LEV-1 and LEV-2 while attempting to restore communications with SLIM.

Further Reading: Twitter.com

The post Japan’s Lunar Lander Fails to Check-in appeared first on Universe Today.

Engineers have developed a model that combines machine learning and collaborative robotics to accelerate the design of aerogel materials used in wearable heating applications.

Engineers have developed a model that combines machine learning and collaborative robotics to accelerate the design of aerogel materials used in wearable heating applications.

Adolescents with an internet addiction undergo changes in the brain that could lead to additional addictive behavior and tendencies, finds a new study.

For a small, lumpy chunk of rock that barely reflects any light, Mars’ Moon Phobos draws a lot of attention. Maybe because it’s one of only two moons to orbit the planet, and its origins are unclear. But some of the attention is probably because we have such great images of it.

Phobos is the largest of Mars’ two moons, the other one being Deimos. Scientists are uncertain about their history. They could be a pair of captured main-belt asteroids, two lobes of what once was a binary asteroid until capture separated them, or a second-generation object formed after Mars had already formed. Or they could be surviving fragments from an ancient collision between more massive objects.

Phobos isn’t very large. It’s about 26 km × 23 km × 18 km and not massive enough to be rounded. Studies of its density show that it’s a rubble-pile body loosely held together by its own gravity.

When the ESA launched its Mars Express Orbiter in 2003, its mission was to study Mars. One of its instruments is the High-Resolution Stereo Camera, a German contribution that produces colour images with up to two meters resolution. The instrument also has a black-and-white mode, and the original image of Phobos was black-and-white.

Andrea Luck is a skilled image processor from Glasgow, Scotland, with a healthy enthusiasm for space images. He decided the original B&W image, which he describes as epic, needed to be updated to colour. “I was kinda tired of seeing this epic photo online only in black and white, so I decided to jazz it up with some colours!” he wrote on his Flickr page.

It’s interesting to note that it’s a single image, not a composite.

Here’s the original B&W image.

This is the original image from the High-Resolution Stereo Camera (HRSC) on ESA’s Mars Express spacecraft. It caught Phobos over Mars’ limb on March 26, 2010. The waviness of Mars in the background is a by-product of HRSC’s line-scanning operation. Image Credit: ESA / DLR / FU Berlin (G. Neukum)

The HRSC’s mission is to take stereographic images of Mars’ surface, capturing geological and morphological details. The goal is to map as much of the surface as possible. But at the bottom of its list of objectives are images of Phobos and Deimos.

The HRSC captured this image of Phobos in 2017. It shows the Stickney Crater, Phobos’ largest impact crater, and the unusual grooves on the moon’s surface. Mars Express images helped scientists conclude that the grooves are likely from impact ejecta. Image Credit: ESA/DLR/FU Berlin. CC BY-SA 3.0 IGO

Images of Phobos have helped scientists better understand the odd moon, but they’re not enough to reach solid conclusions. Fortunately, a mission to Phobos and its sibling Deimos will be launched in a couple of years.

JAXA, the Japan Aerospace Exploration Agency, is launching the MMX mission in 2026. MMX stands for Martian Moons Exploration. Its goal is to understand the origins of Phobos and Deimos. MMX will also return a sample from Phobos in 2031. Once in Earthly labs, those samples should reveal a lot.

But for now, we can enjoy this processed image of Phobos, which captures its nature as a fast-moving, rubble-pile moon with uncertain origins.

The post How Mars’ Moon Phobos Captures Our Imaginations appeared first on Universe Today.

There are plenty of problems that spacecraft designers have to consider. Getting smacked in the sensitive parts by a rock is just one of them, but it is a very important one. A micrometeoroid hitting the wrong part of the spacecraft could jeopardize an entire mission, and the years of work it took to get to the point where the mission was actually in space in the first place. But even if the engineers who design spacecraft know about this risk, how is it best to avoid them? A new programming library from research at NASA could help.

Admittedly, engineers already have a tool for this purpose. NASA’s Meteoroid Engineering Model (MEM) allows them to plug in a planned trajectory for their spacecraft and receive an output that defines where and from which direction they are likely to encounter micrometeoroids.

The James Webb Space Telescope is a perfect example of why such a system is necessary. On its way to the L2 Lagrange point, it has already suffered at least 20 micrometeoroid impacts, at least one of which hit the space telescope’s primary mirror, leaving a dent that still affects the quality of its images to this day.

How do micrometeroids affect spacecraft?
Credit – Chris Pattison YouTube Channel

Due to such high-profile occurrences, spacecraft designers are already aware of the risks. However, many don’t know their trajectories when designing their systems. Without a planned trajectory, the MEM is all but useless.

Enter Althea Moorhead from NASA’s Meteoroid Environment Office at Marshall Space Flight Center and her colleagues Katie Milbrandt from Auburn and Aaron Kingery from ERC, Inc., also based at Marshall. They improved the MEM’s functionality by introducing a library of known spacecraft trajectories and the MEM outputs for each.

Instead of knowing their end trajectory, spacecraft designers would be able to simply look at the library and determine whether there are any significant risks from meteoroids on any number of potential trajectories. In particular, the library includes data on orbital paths around every significant planet, some transfer orbits, and at least two “halo” orbits, where the spacecraft would take advantage of the relative stability of a planet’s Lagrange points.

How Webb deals with the micrometeroid impacts its already suffere.
Credit – Launch Pad Astronomy

The output of the library allows for visualizations of the risks the spacecraft would encounter, which is much easier to understand than complex equations and probabilities for designers who don’t necessarily specialize in micrometeoroid hazards. That was the original impetus for developing the library – to provide generalists who don’t necessarily have time to grok the details of micrometeoroid location and risks but still need to consider it as part of their mission design.

The paper authors stress that the library shouldn’t be used for the formal risk assessment that NASA requires of all missions destined for launch. That requirement can still be met by the MEM itself, along with a well-established orbit. But, if that orbit happens to be informed by the library described in the paper, all the better for it.

Learn More:
Moorhead, Milbrandt, & Kingery – A library of meteoroid environments encountered by spacecraft in the inner solar system
UT – NASA has a Plan to Minimize Future Micrometeoroid Impacts on JWST
UT – What Does Micrometeoroid Damage do to Gossamer Structures Like Webb’s Sunshield?
UT – Ouch. Canadarm2 Took a Direct Hit From a Micrometeorite

Lead Image:
Visualization of one of the trajectories planned out in the new micrometeroid library.
Credit – Moorhead, Milbrandt, & Kingery

The post NASA has a New Database to Predict Meteoroid Hazards for Spaceflight appeared first on Universe Today.

The Standard Model of particle physics does a good job of explaining the interactions between matter’s basic building blocks. But it’s not perfect. It struggles to explain dark matter. Dark matter makes up most of the matter in the Universe, yet we don’t know what it is.

The Standard Model says that whatever dark matter is, it can’t interact with itself. New research may have turned that on its head.

Physicists propose many different candidates for dark matter, including dark photons, weakly interacting massive particles (WIMPs), primordial black holes, and more. Each one is intriguing in its own way, but there’s no confirmation regarding any of them. And each one is a proposed part of the Standard Model.

New research in the journal Astronomy and Astrophysics suggests we may be barking up the wrong tree. It suggests that another model, called the Self-Interacting Dark Matter model, can explain dark matter while the Standard Model and its Lambda Cold Dark Matter (Lambda CDM) simply can’t.

The paper is “An N-body/hydrodynamical simulation study of the merging cluster El Gordo: A compelling case for self-interacting dark matter?” The lead author is Riccardo Valdarnini of SISSA’s (Scuola Internazionale Superiore di Studi Avanzati) Astrophysics and Cosmology group.

El Gordo is an extremely massive, extremely distant galaxy cluster more than seven billion light-years away from Earth. It’s comprised of two galaxy sub-clusters that are colliding with one another at several million kilometres per hour. It’s at the center of a back-and-forth over dark matter and the Lambda CDM.

A 2021 paper claimed that El Gordo presents a challenge for the Lambda-CDM model because it appeared so early in cosmic history, is extremely massive, and has such a high collisional velocity. “Such a fast collision between individually rare massive clusters is unexpected in Lambda cold dark matter cosmology at such high z,” the authors of that paper wrote.

A later paper from 2021 arrived at a lower mass estimate for El Gordo, one that was consistent with Lambda CDM. “Such an extreme mass of El Gordo has stimulated a number of discussions on whether or not the presence of the cluster is in tension with the Lambda CDM paradigm,” those authors wrote. “The new mass is compatible with the current Lambda CDM cosmology.”

A key part of Lambda CDM is that dark matter is both cold and collisionless. In that model, it’s impossible for dark matter particles to collide with one another; they can only interact through gravity and possibly the weak force. This study challenges that notion.

Proving that dark matter can interact with itself via collisions is difficult and complicated. El Gordo is a good place to study the Self-Interacting Dark Matter (SIDM) idea. “There are, however, unique
laboratories that can prove very useful for this purpose, many light years away from us,” said lead author Valdarnini. “These are the massive galaxy clusters, gigantic cosmic structures that, upon collision, determine the most energetic events since the Big Bang.” El Gordo is one of them.

Galaxy clusters like El Gordo can be divided into three components: the galaxies, the dark matter, and the gas mass. The Standard Model says that the colliding gas loses some of its initial energy during the collision. “This is why, after the collision, the peak of gas mass density will lag behind those of dark matter and galaxies,” Valdarnini explained.

But the SIDM says something different. It says that the points where the dark matter reaches its maximum density, called centroids, should be physically separated from the other mass components. The peculiarities of that separation are a signature of SIDM.

Observations of El Gordo show that it consists of two large sub-clusters, the northwest (NW) and the southeast (SE), which are merging into one.

This Hubble Space Telescope image shows El Gordo’s two main components, the NW and SE sub-clusters. Image Credit: NASA, ESA, and J. Jee (University of California, Davis)

X-ray images show different peak locations for the different mass components. The X-ray image below shows a single X-ray emission peak in the SE subcluster and two faint tails elongated beyond the X-ray peak. The X-ray peak precedes the dark matter peak. The Brightest Cluster Galaxy (BCG) is also offset from the SE mass centroid. BCGs are the brightest galaxies in a given cluster, are extremely massive, and are centers of mass in clusters.

“Another notable aspect can be seen in the NW cluster, where the galaxy number density peak is spatially offset from the corresponding mass peak,” Valdarnini explained.

This combined X-ray and infrared image shows X-rays from Chandra in pink, and the blue shows where dark matter is found. Image Credit: X-ray: NASA/CXC/Rutgers/J. Hughes et al.; Infrared: NASA/ESA/CSA, J.M. Diego (IFCA), B.Frye (Univ. of Arizona), P.Kamieneski, T.Carleton & R.Windhorst (ASU)

But those observations alone aren’t enough. In the new paper in Astronomy and Astrophysics, Valdarnini employed a large number of N-body/hydrodynamical simulations to study El Gordo’s physical properties. The systematic simulations aim to match the observations. Each simulation has slightly different parameters, and when a simulation matches observations, those parameters are likely to offer some explanation of the observations.

Valdarnini explains it clearly in the paper. “… the aim of this paper is to determine whether it is possible to construct merger models for the El Gordo cluster that can consistently reproduce the observed X-ray morphology, as well as many of its physical properties.”

The critical part of this work and its simulations concerns the separations between the centers of mass in El Gordo. If simulations can produce that, it’s evidence in favour of SIDM.

“The most significant result of this simulation study is that the relative separations observed between the different mass centroids of the “El Gordo” cluster are naturally explained if the dark matter is self-interacting,” states Valdarnini.

This figure from the research shows some of the simulation results. The red contours show X-ray surface brightness, and the white shows mass density. Green crosses are mass centroids, and red crosses are X-ray surface brightness centroids. Each row is from a separate simulation run with different parameters, and each panel represents a different viewing angle. The middle top panel is of particular interest. It recreates El Gordo’s twin tails particularly well. Image Credit: Valdarnini et al. 2024.

“For this reason, these findings provide an unambiguous signature of a dark matter behaviour that exhibits collisional properties in a very energetic high-redshift cluster collision,” he continued.

It’s a classic “tip of the iceberg scenario.” While these results are in favour of the Self Interacting Dark Matter model, they’re nowhere near conclusive, as Valdarnini makes clear when he talks about inconsistencies in the results.

Valdarnini’s work shows that while the results are an approximation of how dark matter may behave during cluster mergers, there’s a lot more to it. The “underlying physical processes” are extremely complex.

“The study makes a compelling case for the possibility of self-interacting dark matter between colliding clusters as an alternative to the standard collisionless dark matter paradigm,” he concludes.

For most of the eight billion human beings alive today, dark matter is of little consequence in daily life. But if we want to entertain hopes and enjoy daydreams of human civilization lasting for centuries, millennia, or even longer, expanding into space and travelling to other stars, it’s critical that we understand everything we can about nature. The history of human progress parallels our growing understanding of nature.

Understanding dark matter is critical to understanding nature. If we want civilization to persist, a better understanding of everything about nature is the best way forward.

Now, back to our daily lives under the Standard Model.

The post Evidence of Dark Matter Interacting With Itself in El Gordo Merger appeared first on Universe Today.

Our Moon is shrinking and has been doing so since just after its formation ~4.5 billion years ago from a collision with the young Earth. That shrinkage, along with a constant rain of micrometeorites, causes lunar seismic activity. NASA plans to send two instruments to the Moon to measure its moonquakes. Those dual seismometers share technology first used on Mars by the InSight lander to measure more than a thousand marsquakes.

The seismometers make up part of the Farside Seismic Suite (FSS). It will be delivered to the Moon’s Schrödinger Basin at the South Pole, the first such instrument package deployed since the Apollo program seismic payload operated for a brief time in 1971. That program sent back the first moonquake measurements. Subsequent Apollo missions deployed other seismic instruments that transmitted lunar data until late 1977.

JPL engineers and technicians prepare NASA’s Farside Seismic Suite for testing in simulated lunar gravity, which is about one-sixth of Earth’s. The seismometers in the payload will gather the agency’s first seismic data from moonquakes in nearly 50 years. Credit: NASA/JPL-Caltech

The FSS will send back the first such measurements from the Moon’s far side since Apollo days. Its two seismometers will record a “hum” of seismic background vibrations from icrometeorites pelting the surface. In addition, they will record lunar quakes and return data about their intensity and location.

What Do Moonquakes Tell Us?

Quakes give a great deal of information about more than their location and intensity. The way seismic waves travel through the Moon’s structure should give some insight into the density of its various parts. In addition, they help scientists understand the lunar “shrinkage”.

On Earth, seismic waves travel differently through liquid and solid layers. On the Moon, the Apollo 11 seismic experiment gave planetary scientists the first “look” at the lunar interior. For each moonquake, the instrument recorded the strength, duration, and suspected direction of the event.

Apollo 15’s Lunar Surface Experiments Package (ALSEP). It carried a suite of science instruments, including a seismic experiment to detect moonquakes. Courtesy NASA.

Interestingly, that experiment and others did not detect much seismic activity on the lunar far side. Something in the Moon’s interior plays a role in absorbing the waves from far-side quakes. Scientists want to know what that structure is and what properties prevent transmission of quake waves. Of course, not as many quakes occur on the far side. Interestingly, the surface of the far side is much different than the near side. Are these two related? “FSS will offer answers to questions we’ve been asking about the Moon for decades,” said Mark Panning, the FSS principal investigator at JPL and project scientist for InSight. “We cannot wait to start getting this data back.”

From Marsquakes to Moonquakes

In late 2018, the Mars InSight Lander settled onto the surface of the Red Planet. Its mission was to study the interior of Mars. Essentially, it used the Seismic Experiment for Interior Structure (SEIS) to take the planet’s pulse and measure its interior motions. It measured the strength, duration, and direction of marsquakes. It also detected tiny mini-quakes generated by meteorite impacts. Along with a suite of other instruments that measured wind, temperature, and magnetic field variations, SEIS was able to sense vibrations from wind storms and other atmospheric phenomena.

Engineers at NASA Jet Propulsion Laboratory adapted the seismometer technology used on InSight for the FSS suite. There were a few major differences, however. For one thing, lunar gravity is much less than Mars’s, so they had to adapt the seismic suite’s performance to take that into account. Also, temperatures on the Moon are much colder, and of course, there’s no atmosphere to measure.

The FSS suite contains the Very Broadband Seismometer, which is so sensitive it detects ground motions smaller than the size of a hydrogen atom. The other seismometer is called the Short Period sensor and it measures ground motion in three directions using tiny sensors etched onto chips.

FSS’s Science Goals

This payload, its power sources, and thermal controls are expected to operate for a long time, measuring quakes and background “noise” in the lunar structure. Although scientists know a fair amount about the Moon’s interior, the FSS’s sensitive instruments should help them get a more detailed understanding of its structure. The Moon is a differentiated body—meaning that it has layers beneath it crust.

The Apollo mission instruments measured the thickness of the lunar crust, and the GRAIL mission provided more detailed data. The FSS measurements should determine the thickness of the next layer—the deep mantle. That should come from data recordings and measurements of deep moonquakes. The FSS’s landing site in Schrödinger crater is a great location for quake measurements. It’s an impact basin refilled by rock melted during an impact that occurred some 3.8 billion years ago. There is a great deal of evidence for other volcanic activity in the region, including vents and subsequent lava flows.

Seen here during assembly in November 2023, Farside Seismic Suite’s inner cube houses the NASA payload’s large battery (at rear) and its two seismometers. The gold, puck-shaped device holds the Short Period sensor, while the silver enclosure contains the Very Broadband seismometer. These devices will detect moonquakes on the Moon’s far side. Credit: NASA/JPL-Caltech

The FSS seismometer package is slated for launch in 2025 with a projected landing date in 2026. It’s part of a NASA initiative to work with companies to deliver lunar science and technology packages during the Artemis mission timeline. Artemis astronauts will deploy a seismic network using a distributed acoustic sensing capability to do further work in assessing the Moon’s interior.

For More Information

NASA to Measure Moonquakes With Help From InSight Mars Mission
Apollo 11 Seismic Experiment
InSight Lander

The post Two Seismometers are Going to the Moon to Measure Moonquakes appeared first on Universe Today.

When a large language model expresses doubt about the information it supplies, people are less likely to accept it as fact and more likely to find accurate information elsewhere

Radio telescopes have an advantage over optical telescopes, in that radio telescope can be used even in cloudy conditions here on Earth. That’s because the longer wavelengths of radio waves can pass through clouds unhindered. However, some wavelengths are still partially obscured by portions of Earth’s atmosphere, especially by the ionosphere which traps human-made Radio Frequency Interference (RFI).  

Astronomers have developed a new calibration technique that allows them to take sharp images in low radio frequencies — between 16 and 30 MHz — for the first time, bypassing the influence of the ionosphere. The astronomers say this will allow them to study things like plasmas emanating from ancient black holes and perhaps even detect exoplanets that orbit small stars.

The technique was developed by an international team of researchers led by astronomers from Leiden University in the Netherlands.

“It’s like putting on a pair of glasses for the first time and no longer seeing blurred,” said Christian Groeneveld from Leiden University, who led the research.

The LOFAR central stations on a specially engineered field (“superterp”) between Exloo and Buinen in Drenthe, in the north east of the Netherlands. Image: Aerophoto Eelde.

The astronomers used the LOFAR telescope in Drenthe, the Netherlands, which is currently one of the best low-frequency radio telescopes in the world. They modified a calibration technique that has been used to improve observations for observing in radio at higher frequencies, around 150 MHz.

“We hoped that we could also extend this technique to lower frequencies, below 30 MHz,” said, Reinout van Weeren, also from Leiden University, who came up with the idea. “And we succeeded.”

To test their technique, they studied several galaxy clusters that had previously only been studied in detail at higher frequencies.

“Our observing strategy consisted of simultaneously observing a bright primary calibrator and the target fields,” the team wrote in their paper. “By scheduling the observation after midnight, we minimized RFI caused by the internal reflection of terrestrial RFI by the ionosphere, which is significantly worse during the day, as ionizing radiation from the Sun increases the column density of ions in the ionosphere.”

Then, they split up their field of view into several smaller “facets” and self-calibrated each facet individually, against the calibrator object. “This yields an improved image and model of the sky, partly corrected for direction dependent effects,” they wrote. They then repeated the calibrations three more times.

Left shows an image of a piece of sky observed with the hitherto best calibration technique. Right shows the same piece of sky with the new technique. More detail is visible, and what were once large, blurry patches now appear as single points. (c) LOFAR/Groeneveld et al.

This was the first time radio images at frequencies between 16 and 30 MHz have been taken. Because of this data, the astronomers said that the radio emissions from these clusters is not evenly distributed across the entire cluster, but rather there is a spot pattern.

According to the researchers, the new calibration technique makes it possible to study radio phenomena in frequencies that were previously hidden.

“There is, of course, a chance that we will eventually discover something unexpected,” said Groeneveld.

Read the press release from Astronomy Netherlands
Read the team’s paper

The post Astronomers Have a New Way to Bypass Earth's Atmosphere appeared first on Universe Today.

Auxetics defy common sense, widening when stretched and narrowing when compressed. Researchers have now made the process of using them much easier, paving the way for new types of auxetic products -- from better sneaker insoles to blast-resilient buildings.

Researchers have developed a new method for detecting defects in additively manufactured components.

A new study offers a framework for an objective, non-invasive and zero-effort sleep monitoring system utilizing smart thermostats equipped with motion sensors.

Have you ever wished you could understand what your dog is trying to say to you? Researchers are exploring the possibilities of AI, developing tools that can identify whether a dog's bark conveys playfulness or aggression.

Researchers have developed a model that enables computers to interpret and understand human emotions, utilizing principles of mathematical psychology. In the future, the model can help the computer to adapt its own behavior and guide an irritated or anxious user in different ways. The implications of such technology are profound, offering a glimpse into a future where computers are not merely tools, but empathetic partners in user interaction.

Previous research has demonstrated that adding tin to beta-type titanium alloys improves their strength. But scientists have yet to understand the reasons for this. Now, a research team has pinpointed the exact mechanisms behind this phenomenon.

For years, various models have been developed to describe an important class of mixing effects that occur, for example, in the flow in a chemical reactor. Experimental validation, however, has lagged far behind due to the superimposition of gravity effects. Scientists have now closed this gap with experiments conducted under weightlessness.

Researchers have developed a platform for the interactive evaluation of AI-powered chatbots such as ChatGPT. A team of computer scientists, engineers, mathematicians and cognitive scientists developed an open-source evaluation platform called CheckMate, which allows human users to interact with and evaluate the performance of large language models (LLMs).