australia

rocket-report:-australia-says-yes-to-the-launch;-russia-delivers-for-iran

Rocket Report: Australia says yes to the launch; Russia delivers for Iran


The world’s first wooden satellite arrived at the International Space Station this week.

A Falcon 9 booster fires its engines on SpaceX’s “tripod” test stand in McGregor, Texas. Credit: SpaceX

Welcome to Edition 7.19 of the Rocket Report! Okay, we get it. We received more submissions from our readers on Australia’s approval of a launch permit for Gilmour Space than we’ve received on any other news story in recent memory. Thank you for your submissions as global rocket activity continues apace. We’ll cover Gilmour in more detail as they get closer to launch. There will be no Rocket Report next week as Eric and I join the rest of the Ars team for our 2024 Technicon in New York.

As always, we welcome reader submissions. If you don’t want to miss an issue, please subscribe using the box below (the form will not appear on AMP-enabled versions of the site). Each report will include information on small-, medium-, and heavy-lift rockets as well as a quick look ahead at the next three launches on the calendar.

Gilmour Space has a permit to fly. Gilmour Space Technologies has been granted a permit to launch its 82-foot-tall (25-meter) orbital rocket from a spaceport in Queensland, Australia. The space company, founded in 2012, had initially planned to lift off in March but was unable to do so without approval from the Australian Space Agency, the Australian Broadcasting Corporation reports. The government approved Gilmour’s launch permit Monday, although the company is still weeks away from flying its three-stage Eris rocket.

A first for Australia … Australia hosted a handful of satellite launches with US and British rockets from 1967 through 1971, but Gilmour’s Eris rocket would become the first all-Australian launch vehicle to reach orbit. The Eris rocket is capable of delivering about 670 pounds (305 kilograms) of payload mass into a Sun-synchronous orbit. Eris will be powered by hybrid rocket engines burning a solid fuel mixed with a liquid oxidizer, making it unique among orbital-class rockets. Gilmour completed a wet dress rehearsal, or practice countdown, with the Eris rocket on the launch pad in Queensland in September. The launch permit becomes active after 30 days, or the first week of December. “We do think we’ve got a good chance of launching at the end of the 30-day period, and we’re going to give it a red hot go,” said Adam Gilmour, the company’s co-founder and CEO. (submitted by Marzipan, mryall, ZygP, Ken the Bin, Spencer Willis, MarkW98, and EllPeaTea)

North Korea tests new missile. North Korea apparently completed a successful test of its most powerful intercontinental ballistic missile on October 31, lofting it nearly 4,800 miles (7,700 kilometers) into space before the projectile fell back to Earth, Ars reports. This solid-fueled, multi-stage missile, named the Hwasong-19, is a new tool in North Korea’s increasingly sophisticated arsenal of weapons. It has enough range—perhaps as much as 9,320 miles (15,000 kilometers), according to Japan’s government—to strike targets anywhere in the United States. It also happens to be one of the largest ICBMs in the world, rivaling the missiles fielded by the world’s more established nuclear powers.

Quid pro quo? … The Hwasong-19 missile test comes as North Korea deploys some 10,000 troops inside Russia to support the country’s war against Ukraine. The budding partnership between Russia and North Korea has evolved for several years. Russian President Vladimir Putin has met with North Korean leader Kim Jong Un on multiple occasions, most recently in Pyongyang in June. This has fueled speculation about what Russia is offering North Korea in exchange for the troops deployed on Russian soil. US and South Korean officials have some thoughts. They said North Korea is likely to ask for technology transfers in diverse areas related to tactical nuclear weapons, ICBMs, and reconnaissance satellites.

The easiest way to keep up with Eric Berger’s and Stephen Clark’s reporting on all things space is to sign up for our newsletter. We’ll collect their stories and deliver them straight to your inbox.

Sign Me Up!

Virgin Galactic is on the hunt for cash. Virgin Galactic is proposing to raise $300 million in additional capital to accelerate production of suborbital spaceplanes and a mothership aircraft the company says can fuel its long-term growth, Space News reports. The company, founded by billionaire Richard Branson, suspended operations of its VSS Unity suborbital spaceplane earlier this year. VSS Unity hit a monthly flight cadence carrying small groups of space tourists and researchers to the edge of space, but it just wasn’t profitable. Now, Virgin Galactic is developing larger Delta-class spaceplanes it says will be easier and cheaper to turn around between flights.

All-in with Delta … Michael Colglazier, Virgin Galactic’s CEO, announced the company’s appetite for fundraising in a quarterly earnings call with investment analysts Wednesday. He said manufacturing of components for Virgin Galactic’s first two Delta-class ships, which the company says it can fund with existing cash, is proceeding on schedule at a factory in Arizona. Virgin Galactic previously said it would use revenue from paying passengers on its first two Delta-class ships to pay for development of future vehicles. Instead, Virgin Galactic now says it wants to raise money to speed up work on the third and fourth Delta-class vehicles, along with a second airplane mothership to carry the spaceplanes aloft before they release and fire into space. (submitted by Ken the Bin and EllPeaTea)

ESA breaks its silence on Themis. The European Space Agency has provided a rare update on the progress of its Themis reusable booster demonstrator project, European Spaceflight reports. ESA is developing the Themis test vehicle for atmospheric flights to fine-tune technologies for a future European reusable rocket capable of vertical takeoffs and vertical landings. Themis started out as a project led by CNES, the French space agency, in 2018. ESA member states signed up to help fund the project in 2019, and the agency awarded ArianeGroup a contract to move forward with Themis in 2020. At the time, the first low-altitude hop test was expected to take place in 2022.

Some slow progress … Now, the first low-altitude hop is scheduled for 2025 from Esrange Space Centre in Sweden, a three-year delay. This week, ESA said engineers have completed testing of the Themis vehicle’s main systems, and assembly of the demonstrator is underway in France. A single methane-fueled Prometheus engine, also developed by ArianeGroup, has been installed on the rocket. Teams are currently adding avionics, computers, electrical systems, and cable harnesses. Themis’ stainless steel propellant tanks have been manufactured, tested, and cleaned and are now ready to be installed on the Themis demonstrator. Then, the rocket will travel by road from France to the test site in Sweden for its initial low-altitude hops. After those flights are complete, officials plan to add two more Prometheus engines to the rocket and ship it to French Guiana for high-altitude test flights. (submitted by Ken the Bin and EllPeaTea)

SpaceX will give the ISS a boost. A Cargo Dragon spacecraft docked to the International Space Station on Tuesday morning, less than a day after lifting off from Florida. As space missions go, this one is fairly routine, ferrying about 6,000 pounds (2,700 kilograms) of cargo and science experiments to the space station. One thing that’s different about this mission is that it delivered to the station a tiny 2 lb (900 g) satellite named LignoSat, the first spacecraft made of wood, for later release outside the research complex. There is one more characteristic of this flight that may prove significant for NASA and the future of the space station, Ars reports. As early as Friday, NASA and SpaceX have scheduled a “reboost and attitude control demonstration,” during which the Dragon spacecraft will use some of the thrusters at the base of the capsule. This is the first time the Dragon spacecraft will be used to move the space station.

Dragon’s breath … Dragon will fire a subset of its 16 Draco thrusters, each with about 90 pounds of thrust, for approximately 12.5 minutes to make a slight adjustment to the orbital trajectory of the roughly 450-ton space station. SpaceX and NASA engineers will analyze the results from the demonstration to determine if Dragon could be used for future space station reboost opportunities. The data will also inform the design of the US Deorbit Vehicle, which SpaceX is developing to perform the maneuvers required to bring the space station back to Earth for a controlled, destructive reentry in the early 2030s. For NASA, demonstrating Dragon’s ability to move the space station will be another step toward breaking free of reliance on Russia, which is currently responsible for providing propulsion to maneuver the orbiting outpost. Northrop Grumman’s Cygnus supply ship also previously demonstrated a reboost capability. (submitted by Ken the Bin and N35t0r)

Russia launches Soyuz in service of Iran. Russia launched a Soyuz rocket Monday carrying two satellites designed to monitor the space weather around Earth and 53 small satellites, including two Iranian ones, Reuters reports. The primary payloads aboard the Soyuz-2.1b rocket were two Ionosfera-M satellites to probe the ionosphere, an outer layer of the atmosphere near the edge of space. Solar activity can alter conditions in the ionosphere, impacting communications and navigation. The two Iranian satellites on this mission were named Kowsar and Hodhod. They will collect high-resolution reconnaissance imagery and support communications for Iran.

A distant third … This was only the 13th orbital launch by Russia this year, trailing far behind the United States and China. We know of two more Soyuz flights planned for later this month, but no more, barring a surprise military launch (which is possible). The projected launch rate puts Russia on pace for its quietest year of launch activity since 1961, the year Yuri Gagarin became the first person to fly in space. A major reason for this decline in launches is the decisions of Western governments and companies to move their payloads off of Russian rockets after the invasion of Ukraine. For example, OneWeb stopped launching on Soyuz in 2022, and the European Space Agency suspended its partnership with Russia to launch Soyuz rockets from French Guiana. (submitted by Ken the Bin)

H3 deploys Japanese national security satellite. Japan launched a defense satellite Monday aimed at speedier military operations and communication on an H3 rocket and successfully placed it into orbit, the Associated Press reports. The Kirameki 3 satellite will use high-speed X-band communication to support Japan’s defense ministry with information and data sharing, and command and control services. The satellite will serve Japanese land, air, and naval forces from its perch in geostationary orbit alongside two other Kirameki communications satellites.

Gaining trust … The H3 is Japan’s new flagship rocket, developed by Mitsubishi Heavy Industries (MHI) and funded by the Japan Aerospace Exploration Agency (JAXA). The launch of Kirameki 3 marked the third consecutive successful launch of the H3 rocket, following a debut flight in March 2023 that failed to reach orbit. This was the first time Japan’s defense ministry put one of its satellites on the H3 rocket. The first two Kirameki satellites launched on a European Ariane 5 and a Japanese H-IIA rocket, which the H3 will replace. (submitted by Ken the Bin, tsunam, and EllPeaTea)

Rocket Lab enters the race for military contracts. Rocket Lab is aiming to chip away at SpaceX’s dominance in military space launch, confirming its bid to compete for Pentagon contracts with its new medium-lift rocket, Neutron, Space News reports. Last month, the Space Force released a request for proposals from launch companies seeking to join the military’s roster of launch providers in the National Security Space Launch (NSSL) program. The Space Force will accept bids for launch providers to “on-ramp” to the NSSL Phase 3 Lane 1 contract, which doles out task orders to launch companies for individual missions. In order to win a task order, a launch provider must be on the Phase 3 Lane 1 contract. Currently, SpaceX, United Launch Alliance, and Blue Origin are the only rocket companies eligible. SpaceX won all of the first round of Lane 1 task orders last month.

Joining the club … The Space Force is accepting additional risk for Lane 1 missions, which largely comprise repeat launches deploying a constellation of missile-tracking and data-relay satellites for the Space Development Agency. A separate class of heavy-lift missions, known as Lane 2, will require rockets to undergo a thorough certification by the Space Force to ensure their reliability. In order for a launch company to join the Lane 1 roster, the Space Force requires bidders to be ready for a first launch by December 2025. Peter Beck, Rocket Lab’s founder and CEO, said he thinks the Neutron rocket will be ready for its first launch by then. Other new medium-lift rockets, such as Firefly Aerospace’s MLV and Relativity’s Terran-R, almost certainly won’t be ready to launch by the end of next year, leaving Rocket Lab as the only company that will potentially join incumbents SpaceX, ULA, and Blue Origin. (submitted by Ken the Bin)

Next Starship flight is just around the corner. Less than a month has passed since the historic fifth flight of SpaceX’s Starship, during which the company caught the booster with mechanical arms back at the launch pad in Texas. Now, another test flight could come as soon as November 18, Ars reports. The improbable but successful recovery of the Starship first stage with “chopsticks” last month, and the on-target splashdown of the Starship upper stage halfway around the world, allowed SpaceX to avoid an anomaly investigation by the Federal Aviation Administration. Thus, the company was able to press ahead on a sixth test flight if it flew a similar profile. And that’s what SpaceX plans to do, albeit with some notable additions to the flight plan.

Around the edges … Perhaps the most significant change to the profile for Flight 6 will be an attempt to reignite a Raptor engine on Starship while it is in space. SpaceX tried to do this on a test flight in March but aborted the burn because the ship’s rolling motion exceeded limits. A successful demonstration of a Raptor engine relight could pave the way for SpaceX to launch Starship into a higher stable orbit around Earth on future test flights. This is required for SpaceX to begin using Starship to launch Starlink Internet satellites and perform in-orbit refueling experiments with two ships docked together. (submitted by EllPeaTea)

China’s version of Starship. China has updated the design of its next-generation heavy-lift rocket, the Long March 9, and it looks almost exactly like a clone of SpaceX’s Starship rocket, Ars reports. The Long March 9 started out as a conventional-looking expendable rocket, then morphed into a launcher with a reusable first stage. Now, the rocket will have a reusable booster and upper stage. The booster will have 30 methane-fueled engines, similar to the number of engines on SpaceX’s Super Heavy booster. The upper stage looks remarkably like Starship, with flaps in similar locations. China intends to fly this vehicle for the first time in 2033, nearly a decade from now.

A vehicle for the Moon … The reusable Long March 9 is intended to unlock robust lunar operations for China, similar to the way Starship, and to some extent Blue Origin’s Blue Moon lander, promises to support sustained astronaut stays on the Moon’s surface. China says it plans to land its astronauts on the Moon by 2030, initially using a more conventional architecture with an expendable rocket named the Long March 10, and a lander reminiscent of NASA’s Apollo lunar lander. These will allow Chinese astronauts to remain on the Moon for a matter of days. With Long March 9, China could deliver massive loads of cargo and life support resources to sustain astronauts for much longer stays.

Ta-ta to the tripod. The large three-legged vertical test stand at SpaceX’s engine test site in McGregor, Texas, is being decommissioned, NASA Spaceflight reports. Cranes have started removing propellant tanks from the test stand, nicknamed the tripod, towering above the Central Texas prairie. McGregor is home to SpaceX’s propulsion test team and has 16 test cells to support firings of Merlin, Raptor, and Draco engines multiple times per day for the Falcon 9 rocket, Starship, and Dragon spacecraft.

Some history … The tripod might have been one of SpaceX’s most important assets in the company’s early years. It was built by Beal Aerospace for liquid-fueled rocket engine tests in the late 1990s. Beal Aerospace folded, and SpaceX took over the site in 2003. After some modifications, SpaceX installed the first qualification version of its Falcon 9 rocket on the tripod for a series of nine-engine test-firings leading up to the rocket’s inaugural flight in 2010. SpaceX test-fired numerous new Falcon 9 boosters on the tripod before shipping them to launch sites in Florida or California. Most recently, the tripod was used for testing of Raptor engines destined to fly on Starship and the Super Heavy booster.

Next three launches

Nov. 9:  Long March 2C | Unknown Payload | Jiuquan Satellite Launch Center, China | 03: 40 UTC

Nov. 9: Falcon 9 | Starlink 9-10 | Vandenberg Space Force Base, California | 06: 14 UTC

Nov. 10:  Falcon 9 | Starlink 6-69 | Cape Canaveral Space Force Station, Florida | 21: 28 UTC

Photo of Stephen Clark

Stephen Clark is a space reporter at Ars Technica, covering private space companies and the world’s space agencies. Stephen writes about the nexus of technology, science, policy, and business on and off the planet.

Rocket Report: Australia says yes to the launch; Russia delivers for Iran Read More »

x-fails-to-avoid-australia-child-safety-fine-by-arguing-twitter-doesn’t-exist

X fails to avoid Australia child safety fine by arguing Twitter doesn’t exist

“I cannot accept this evidence without a much better explanation of Mr. Bogatz’s path of reasoning,” Wheelahan wrote.

Wheelahan emphasized that the Nevada merger law specifically stipulated that “all debts, liabilities, obligations and duties of the Company shall thenceforth remain with or be attached to, as the case may be, the Acquiror and may be enforced against it to the same extent as if it had incurred or contracted all such debts, liabilities, obligations, and duties.” And Bogatz’s testimony failed to “grapple with the significance” of this, Wheelahan said.

Overall, Wheelahan considered Bogatz’s testimony on X’s merger-acquired liabilities “strained,” while deeming the government’s US merger law expert Alexander Pyle to be “honest and ready to make appropriate concessions,” even while some of his testimony was “not of assistance.”

Luckily, it seemed that Wheelahan had no trouble drawing his own conclusion after analyzing Nevada’s merger law.

“I find that a Nevada court would likely hold that the word ‘liabilities'” in the merger law “is broad enough on its proper construction under Nevada law to encompass non-pecuniary liabilities, such as the obligation to respond to the reporting notice,” Wheelahan wrote. “X Corp has therefore failed to show that it was not required to respond to the reporting notice.”

Because X “failed on all its claims,” the social media company must cover costs from the appeal, and X’s costs in fighting the initial fine will seemingly only increase from here.

Fighting fine likely to more than double X costs

In a press release celebrating the ruling, eSafety Commissioner Julie Inman Grant criticized X’s attempt to use the merger to avoid complying with Australia’s Online Safety Act.

X fails to avoid Australia child safety fine by arguing Twitter doesn’t exist Read More »

“fascists”:-elon-musk-responds-to-proposed-fines-for-disinformation-on-x

“Fascists”: Elon Musk responds to proposed fines for disinformation on X

Being responsible is so hard —

“Elon Musk’s had more positions on free speech than the Kama Sutra,” says lawmaker.

A smartphone displays Elon Musk's profile on X, the app formerly known as Twitter.

Getty Images | Dan Kitwood

Elon Musk has lambasted Australia’s government as “fascists” over proposed laws that could levy substantial fines on social media companies if they fail to comply with rules to combat the spread of disinformation and online scams.

The billionaire owner of social media site X posted the word “fascists” on Friday in response to the bill, which would strengthen the Australian media regulator’s ability to hold companies responsible for the content on their platforms and levy potential fines of up to 5 percent of global revenue. The bill, which was proposed this week, has yet to be passed.

Musk’s comments drew rebukes from senior Australian politicians, with Stephen Jones, Australia’s finance minister, telling national broadcaster ABC that it was “crackpot stuff” and the legislation was a matter of sovereignty.

Bill Shorten, the former leader of the Labor Party and a cabinet minister, accused the billionaire of only championing free speech when it was in his commercial interests. “Elon Musk’s had more positions on free speech than the Kama Sutra,” Shorten said in an interview with Australian radio.

The exchange marks the second time that Musk has confronted Australia over technology regulation.

In May, he accused the country’s eSafety Commissioner of censorship after the government agency took X to court in an effort to force it to remove graphic videos of a stabbing attack in Sydney. A court later denied the eSafety Commissioner’s application.

Musk has also been embroiled in a bitter dispute with authorities in Brazil, where the Supreme Court ruled last month that X should be blocked over its failure to remove or suspend certain accounts accused of spreading misinformation and hateful content.

Australia has been at the forefront of efforts to regulate the technology sector, pitting it against some of the world’s largest social media companies.

This week, the government pledged to introduce a minimum age limit for social media use to tackle “screen addiction” among young people.

In March, Canberra threatened to take action against Meta after the owner of Facebook and Instagram said it would withdraw from a world-first deal to pay media companies to link to news stories.

The government also introduced new data privacy measures to parliament on Thursday that would impose hefty fines and potential jail terms of up to seven years for people found guilty of “doxxing” individuals or groups.

Prime Minister Anthony Albanese’s government had pledged to outlaw doxxing—the publication of personal details online for malicious purposes—this year after the details of a private WhatsApp group containing hundreds of Jewish Australians were published online.

Australia is one of the first countries to pursue laws outlawing doxxing. It is also expected to introduce a tranche of laws in the coming months to regulate how personal data can be used by artificial intelligence.

“These reforms give more teeth to the regulation,” said Monique Azzopardi at law firm Clayton Utz.

© 2024 The Financial Times Ltd. All rights reserved. Not to be redistributed, copied, or modified in any way.

“Fascists”: Elon Musk responds to proposed fines for disinformation on X Read More »

ai-trains-on-kids’-photos-even-when-parents-use-strict-privacy-settings

AI trains on kids’ photos even when parents use strict privacy settings

“Outrageous” —

Even unlisted YouTube videos are used to train AI, watchdog warns.

AI trains on kids’ photos even when parents use strict privacy settings

Human Rights Watch (HRW) continues to reveal how photos of real children casually posted online years ago are being used to train AI models powering image generators—even when platforms prohibit scraping and families use strict privacy settings.

Last month, HRW researcher Hye Jung Han found 170 photos of Brazilian kids that were linked in LAION-5B, a popular AI dataset built from Common Crawl snapshots of the public web. Now, she has released a second report, flagging 190 photos of children from all of Australia’s states and territories, including indigenous children who may be particularly vulnerable to harms.

These photos are linked in the dataset “without the knowledge or consent of the children or their families.” They span the entirety of childhood, making it possible for AI image generators to generate realistic deepfakes of real Australian children, Han’s report said. Perhaps even more concerning, the URLs in the dataset sometimes reveal identifying information about children, including their names and locations where photos were shot, making it easy to track down children whose images might not otherwise be discoverable online.

That puts children in danger of privacy and safety risks, Han said, and some parents thinking they’ve protected their kids’ privacy online may not realize that these risks exist.

From a single link to one photo that showed “two boys, ages 3 and 4, grinning from ear to ear as they hold paintbrushes in front of a colorful mural,” Han could trace “both children’s full names and ages, and the name of the preschool they attend in Perth, in Western Australia.” And perhaps most disturbingly, “information about these children does not appear to exist anywhere else on the Internet”—suggesting that families were particularly cautious in shielding these boys’ identities online.

Stricter privacy settings were used in another image that Han found linked in the dataset. The photo showed “a close-up of two boys making funny faces, captured from a video posted on YouTube of teenagers celebrating” during the week after their final exams, Han reported. Whoever posted that YouTube video adjusted privacy settings so that it would be “unlisted” and would not appear in searches.

Only someone with a link to the video was supposed to have access, but that didn’t stop Common Crawl from archiving the image, nor did YouTube policies prohibiting AI scraping or harvesting of identifying information.

Reached for comment, YouTube’s spokesperson, Jack Malon, told Ars that YouTube has “been clear that the unauthorized scraping of YouTube content is a violation of our Terms of Service, and we continue to take action against this type of abuse.” But Han worries that even if YouTube did join efforts to remove images of children from the dataset, the damage has been done, since AI tools have already trained on them. That’s why—even more than parents need tech companies to up their game blocking AI training—kids need regulators to intervene and stop training before it happens, Han’s report said.

Han’s report comes a month before Australia is expected to release a reformed draft of the country’s Privacy Act. Those reforms include a draft of Australia’s first child data protection law, known as the Children’s Online Privacy Code, but Han told Ars that even people involved in long-running discussions about reforms aren’t “actually sure how much the government is going to announce in August.”

“Children in Australia are waiting with bated breath to see if the government will adopt protections for them,” Han said, emphasizing in her report that “children should not have to live in fear that their photos might be stolen and weaponized against them.”

AI uniquely harms Australian kids

To hunt down the photos of Australian kids, Han “reviewed fewer than 0.0001 percent of the 5.85 billion images and captions contained in the data set.” Because her sample was so small, Han expects that her findings represent a significant undercount of how many children could be impacted by the AI scraping.

“It’s astonishing that out of a random sample size of about 5,000 photos, I immediately fell into 190 photos of Australian children,” Han told Ars. “You would expect that there would be more photos of cats than there are personal photos of children,” since LAION-5B is a “reflection of the entire Internet.”

LAION is working with HRW to remove links to all the images flagged, but cleaning up the dataset does not seem to be a fast process. Han told Ars that based on her most recent exchange with the German nonprofit, LAION had not yet removed links to photos of Brazilian kids that she reported a month ago.

LAION declined Ars’ request for comment.

In June, LAION’s spokesperson, Nathan Tyler, told Ars that, “as a nonprofit, volunteer organization,” LAION is committed to doing its part to help with the “larger and very concerning issue” of misuse of children’s data online. But removing links from the LAION-5B dataset does not remove the images online, Tyler noted, where they can still be referenced and used in other AI datasets, particularly those relying on Common Crawl. And Han pointed out that removing the links from the dataset doesn’t change AI models that have already trained on them.

“Current AI models cannot forget data they were trained on, even if the data was later removed from the training data set,” Han’s report said.

Kids whose images are used to train AI models are exposed to a variety of harms, Han reported, including a risk that image generators could more convincingly create harmful or explicit deepfakes. In Australia last month, “about 50 girls from Melbourne reported that photos from their social media profiles were taken and manipulated using AI to create sexually explicit deepfakes of them, which were then circulated online,” Han reported.

For First Nations children—”including those identified in captions as being from the Anangu, Arrernte, Pitjantjatjara, Pintupi, Tiwi, and Warlpiri peoples”—the inclusion of links to photos threatens unique harms. Because culturally, First Nations peoples “restrict the reproduction of photos of deceased people during periods of mourning,” Han said the AI training could perpetuate harms by making it harder to control when images are reproduced.

Once an AI model trains on the images, there are other obvious privacy risks, including a concern that AI models are “notorious for leaking private information,” Han said. Guardrails added to image generators do not always prevent these leaks, with some tools “repeatedly broken,” Han reported.

LAION recommends that, if troubled by the privacy risks, parents remove images of kids online as the most effective way to prevent abuse. But Han told Ars that’s “not just unrealistic, but frankly, outrageous.”

“The answer is not to call for children and parents to remove wonderful photos of kids online,” Han said. “The call should be [for] some sort of legal protections for these photos, so that kids don’t have to always wonder if their selfie is going to be abused.”

AI trains on kids’ photos even when parents use strict privacy settings Read More »