SoylentNews

upstart writes:

Meta says its new speech-generating AI tool is too dangerous to release:

Meta has unveiled a new AI tool, dubbed 'Voicebox', which it claims represents a breakthrough in AI-powered speech generation. However, the company won't be unleashing it on the public just yet - because doing so could be disastrous.

Voicebox is currently able to produce audio clips of speech in six languages (all of which are European of origin), and - according to a blog post from Meta - is the first AI model of its kind capable of completing tasks beyond what it was 'specifically trained to accomplish'. Meta claims that Voicebox handily outperforms competing speech-generation AIs in virtually every area.

So what exactly is it capable of? Well, for starters, it can spew out reasonably accurate text-to-speech replications of a person's voice using a sample audio file as short as two seconds, a seemingly innocuous ability that holds a huge amount of destructive potential in the wrong hands.

[...] Meta clearly believes its new tool is good enough to fool at least the majority of people [since] it's explicitly not releasing Voicebox to the public, but instead publishing a research paper and detailing a classifier tool that can identify Voicebox-generated speech from real human speech. Meta describes the classifier as "highly effective" - though notably not perfectly effective.

[...] A little caution, patience, and respect for the magnitude of this technology is a welcome sight - although I doubt Meta will sit on Voicebox for too long, since the shareholders will no doubt be wondering how much money it can make them...

Original Submission

hubie writes:

Writing with AI help can shift your opinions:

Artificial intelligence-powered writing assistants that autocomplete sentences or offer "smart replies" not only put words into people's mouths, they also put ideas into their heads, according to new research.

Maurice Jakesch, a doctoral student in the field of information science asked more than 1,500 participants to write a paragraph answering the question, "Is social media good for society?" People who used an AI writing assistant that was biased for or against social media were twice as likely to write a paragraph agreeing with the assistant, and significantly more likely to say they held the same opinion, compared with people who wrote without AI's help.

The study suggests that the biases baked into AI writing tools – whether intentional or unintentional – could have concerning repercussions for culture and politics, researchers said.

"We're rushing to implement these AI models in all walks of life, but we need to better understand the implications," said co-author Mor Naaman, professor at the Jacobs Technion-Cornell Institute at Cornell Tech and of information science in the Cornell Ann S. Bowers College of Computing and Information Science. "Apart from increasing efficiency and creativity, there could be other consequences for individuals and also for our society – shifts in language and opinions."

[...] These technologies deserve more public discussion regarding how they could be misused and how they should be monitored and regulated, the researchers said.

"The more powerful these technologies become and the more deeply we embed them in the social fabric of our societies," Jakesch said, "the more careful we might want to be about how we're governing the values, priorities and opinions built into them."

Journal Reference:
Maurice Jakesch, Advait Bhat, Daniel Buschek, et al., Co-Writing with Opinionated Language Models Affects Users' Views [open], CHI '23: Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, April 2023 DOI: https://doi.org/10.1145/3544548.3581196

Original Submission

takyon writes:

Megaupload Fugitive Arrested By Armed Police 11 Years After The Raid

Since 2012, Kim Dotcom has been fighting extradition to the United States where he faces serious charges related to cloud storage site Megaupload. During that time, very little has been said about Megaupload graphic designer, Julius Bencko. A wanted man in the U.S., Bencko was arrested this month by armed police in the Czech Republic, who filmed the event and published it online.

[...] In the United States government's superseding indictment dated February 16, 2012, Julius Bencko is described as a citizen and resident of Slovakia. A talented graphic designer, Bencko was responsible for Megaupload's logo and ensuring that other sites in the group looked good and accommodated advertising properly.

According to the United States government, this work earned Bencko more than $1 million in 2010 alone. Even for the guy responsible for integrating Megaupload's Flash player, that was still pretty good money.

The important context here is that Bencko was the director and sole shareholder of Basemax International Limited and through that entity, he effectively held 2.5% of Megaupload's shares. For someone who supposedly played a key role in the most significant piracy conspiracy of all time, the allegations against Bencko were underwhelming and, at times, borderline comical.

Original Submission

Freeman writes:

https://arstechnica.com/gaming/2023/06/ftc-xbox-exclusive-starfield-is-powerful-evidence-against-activision-deal/

For months now, Microsoft has sworn up and down that it has no interest in making Call of Duty exclusive to the Xbox if and when its proposed $69 billion Activision acquisition is approved. But as the FTC's request for an injunction stopping that acquisition heads toward opening arguments this week, the federal regulator cites one piece of what it calls "powerful evidence" that it can't trust Microsoft's assurances. In short, as the FTC puts it, "Microsoft's actions following its 2021 acquisition of ZeniMax speak louder than Defendants' words."
[...]
Rather than focusing on what it calls a "strained analogy" to ZeniMax, Microsoft would prefer the court look at Microsoft's purchase of Minecraft-maker Mojang, which has continued to publish the game on a variety of platforms after becoming part of Microsoft. This is a better analogy for Call of Duty, Microsoft writes, because Minecraft was similarly "an existing, multi-player, cross-platform franchise like COD."
[...]
Call of Duty is unlike Minecraft, the FTC argues, in part because Minecraft is available in largely the same form on mobile phones, tablets, and the Switch. "Even if Microsoft took Minecraft off of rival consoles and subscription and cloud gaming services, it would still be available for play on many other devices. The context for Call of Duty is very different."

Regardless, the FTC also argues that this manufactured categorization doesn't matter, because Microsoft's exclusivity decision applied to "all future ZeniMax games." While Microsoft said in 2021 that "some" future Bethesda games would be Xbox exclusives, no Bethesda non-exclusives have been announced since then.

Previously:
US Moves to Block Microsoft's Activision Takeover - 20230613
Microsoft and Activision Will Miss Their Contractual Merger Deadline - 20230115
FTC Moves to Block Microsoft's Activision Acquisition - 20221209
The Biggest Deal in Gaming is Under Fire From U.S. Senators - 20220403

Related:
Microsoft Acquires ZeniMax Media and Bethesda Softworks for $7.5 Billion - 20200921

Original Submission

Freeman writes:

https://arstechnica.com/information-technology/2023/06/marvel-angers-artists-and-fans-with-ai-generated-secret-invasion-tv-intro/

On Wednesday, Marvel's latest comic book TV show, Secret Invasion, premiered on Disney+ to controversy: It features AI-generated motion graphics during its title sequence, and some fans and creators aren't happy about it. "For Marvel, whose whole empire is built on the work of artists to do this is disgusting and I for one shan't be watching," tweeted Marvel comic artist Christian Ward, who worked on Black Bolt.

For those unfamiliar, Secret Invasion is a Marvel Cinematic Universe (MCU) series featuring Samuel L. Jackson as Nick Fury. The storyline centers on a faction of shapeshifting aliens called "Skrulls" that have infiltrated Earth, prompting Fury and his former Skrull ally, Talos, to respond. The plot explores the tension and distrust caused by the Skrulls' shapeshifting abilities, which raises doubts about everyone's real identities.

As reported by Polygon, the show's producers think the melty, flowing, somewhat abstract AI-generated aesthetic (prompted as "Skrull cubism" to the AI model) in the title sequence is appropriate because the show features shapeshifting aliens that are hiding as humans in plain sight.
[...]
As a counterpoint to this common argument that has been surfacing all day on social media, voice actor LittleKuriboh quipped, "Look, if skrulls were as easy to detect as AI art is, the secret invasion would've lasted until suppertime on day zero."

Original Submission

NotSanguine writes:

Ars Technica is reporting on recently published research on "smart" materials using liquid metal.

From the Ars Technica article:

While paper isn't exactly a smart material, it someday could be if it is covered in a new type of liquid metal. This liquid alloy has the potential to turn paper and other materials into gadgets that can do some things on their own.

Liquid metal is already used in smart objects like circuits and wearable sensors—but not as a coating. Inspired by origami, a team of scientists led by Bo Yuan of Tsinghua University in China has figured out a way to formulate liquid metal and apply it with a stamp so it sticks to paper without an adhesive, which has never been possible before. In a study recently published in Cell Reports Physical Science, the scientists showed that paper coated in the metal can be crafted into origami shapes and re-fold itself. The metal coating also conducts heat and electricity. It's like magic. Almost.

owl writes:

https://nanochess.org/video_chess.html

Since I discovered Video Chess for the Atari 2600, I found it pretty impressive. However, I believed the game implemented extra RAM memory, but more recently I discovered it worked using only the 128 bytes of memory available, and with a 4K ROM cartridge. That's an impressive achievement for such a small game console, also the rumor of a bug triggered me to reverse engineer this game to see how it works.

Video Chess was developed by Larry Wagner and Bob Whitehead, and released by Atari in 1979. Although the original game is greater than 4K and a bank-switching PCB was created, the final released game was optimized to use only a 4K ROM cartridge.

Video Chess also has the distinction of displaying eight different objects on a screen row using a technique known as Venetian Blinds, where the odd row of the screen would show four chesspieces, and the even row of the screen would show other four chesspieces. This technique allows the Atari 2600 to exceed its own limit of 3 figures per screen row. I won't discuss this technique as it is a video display trick, and in this article I'll concentrate in the AI (Artificial Intelligence) code.

Original Submission

Freeman writes:

https://arstechnica.com/gadgets/2023/06/camera-review-site-dpreview-finds-a-buyer-avoids-shutdown-by-amazon/

Back in March, the editor-in-chief of the 25-year-old, Amazon-owned camera review site DPReview.com announced that the site would be closing in April. The site was the casualty of a round of layoffs at Amazon that will affect a total of about 27,000 employees this year; DPReview was meant to stop publishing new pieces on April 10 and to be available in read-only mode for an undetermined period of time after that.

But then, something odd happened: The site simply kept publishing at a fairly regular clip throughout the entire month of April and continuing until now. A no-update update from EIC Scott Everett published in mid-May merely acknowledged that pieces were still going up and that there was "nothing to share," which wasn't much to go on but also didn't make it sound as though the site were in imminent danger of disappearing.

Yesterday, Everett finally had something to share: DPReview.com and its "current core editorial, tech, and business team[s]" were acquired by Gear Patrol, an independently owned consumer technology site founded by Eric Yang in 2007. The deal had already closed as of yesterday, June 20.
[...]
"You figure this sort of stuff out before shutting down an entire division of your company, not in hindsight after weeks of backlash and leaving freelancers to scramble for new gigs as they're told they won't have a job..." wrote Burgett.
[...]
"And to that, I'll say again, fuck Amazon," said Burgett.

Original Submission

owl writes:

https://sfconservancy.org/blog/2023/jun/23/rhel-gpl-analysis/

This article was originally published primarily as a response to IBM's Red Hat's change to no longer publish complete, corresponding source (CCS) for RHEL and the prior discontinuation of CentOS Linux (which are related events, as described below). We hope that this will serve as a comprehensive document that discusses the history of Red Hat's RHEL business model, the related source code provisioning, and the GPL compliance issues with RHEL.

For approximately twenty years, Red Hat (now a fully owned subsidiary of IBM) has experimented with building a business model for operating system deployment and distribution that looks, feels, and acts like a proprietary one, but nonetheless complies with the GPL and other standard copyleft terms. Software rights activists, including SFC, have spent decades talking to Red Hat and its attorneys about how the Red Hat Enterprise Linux (RHEL) business model courts disaster and is actively unfriendly to community-oriented Free and Open Source Software (FOSS). These pleadings, discussions, and encouragements have, as far as we can tell, been heard and seriously listened to by key members of Red Hat's legal and OSPO departments, and even by key C-level executives, but they have ultimately been rejected and ignored — sometimes even with a "fine, then sue us for GPL violations" attitude. Activists have found this discussion frustrating, but kept the nature and tenure of these discussions as an "open secret" until now because we all had hoped that Red Hat's behavior would improve. Recent events show that the behavior has simply gotten worse, and is likely to get even worse.

Original Submission

DECbot writes:

Tech reporters don their sports reporting hats as Elon Musk and Mark Zuckerberg exchange words, both agreeing to battle in a cage match. Widely reported in articles in from tech sites like The Guardian, The Verge, and Engadget the two moguls banter and publicly agree to a fight in Vegas. It is to be seen if their agents will book the venue to make this a reality.
 
From The Verge:

After Elon Musk recently tweeted that he would be "up for a cage fight" with Zuckerberg, the Meta CEO shot back by posting a screenshot of Musk's tweet with the caption "send me location."
 
I've confirmed that Zuckerberg's post on his Instagram account is, in fact, not a joke, which means the ball is now in Musk's court. "The story speaks for itself," Meta spokesperson Iska Saric told me.
 
After this story was published, Musk responded with two words: "Vegas Octagon." He then tweeted: "I have this great move that I call 'The Walrus,' where I just lie on top of my opponent and do nothing."
...
In terms of tech billionaire CEOs literally fighting, Musk versus Zuckerberg would be as good as it gets. Musk, 51, has the upper hand on Zuckerberg in terms of sheer physical size, and he has talked about being in "real hard-core street fights" when he was growing up in South Africa. Meanwhile, Zuckerberg, 39, is an aspirational MMA fighter who is already winning Jiu-Jitsu tournaments. He also claims to have recently completed the grueling "Murph Challenge" workout in just under 40 minutes.
 
Regardless of who would win, I think we can all agree that a Musk-versus-Zuckerberg match would be one of the most entertaining fights of all time. It needs to happen. Don't back down now, Musk.

 
As a tech enthusiast, I'm disappointed the battle will be resolved by elbows and fists rather than APIs and technical specs.

Original Submission

upstart writes:

LexisNexis Is Selling Your Personal Data to ICE So It Can Try to Predict Crimes:

The legal research and public records data broker LexisNexis is providing U.S. Immigration and Customs Enforcement with tools to target people who may potentially commit a crime — before any actual crime takes place, according to a contract document obtained by The Intercept. LexisNexis data then helps ICE to track the purported pre-criminals' movements.

The unredacted contract overview provides a rare look at the controversial $16.8 million agreement between LexisNexis and ICE, a federal law enforcement agency whose surveillance of and raids against migrant communities are widely criticized as brutal, unconstitutional, and inhumane.

"The purpose of this program is mass surveillance at its core," said Julie Mao, an attorney and co-founder of Just Futures Law, which is suing LexisNexis over allegations it illegally buys and sells personal data. Mao told The Intercept the ICE contract document, which she reviewed for The Intercept, is "an admission and indication that ICE aims to surveil individuals where no crime has been committed and no criminal warrant or evidence of probable cause."

[...] The federal government allows the general Homeland Security apparatus so much legal latitude, [Georgetown Law School's Center on Privacy and Technology Executive Director Emily] Tucker explained, that an agency like ICE is the perfect vehicle for indiscriminate surveillance of the general public, regardless of immigration status.

Original Submission

Booga1 writes:

MIT News reports: MIT engineers have developed a soft, printable, metal-free electrode.

A new Jell-O-like material could replace metals as electrical interfaces for pacemakers, cochlear implants, and other electronic implants.

Implantable electrodes are predominantly made from rigid metals that are electrically conductive by nature. But over time, metals can aggravate tissues, causing scarring and inflammation that in turn can degrade an implant's performance.

Now, MIT engineers have developed a metal-free, Jell-O-like material that is as soft and tough as biological tissue and can conduct electricity similarly to conventional metals. The material can be made into a printable ink, which the researchers patterned into flexible, rubbery electrodes. The new material, which is a type of high-performance conducting polymer hydrogel, may one day replace metals as functional, gel-based electrodes, with the look and feel of biological tissue.

"This material operates like metal electrodes but is made from gels that are similar to our bodies, and with similar water content," says Hyunwoo Yuk SM '16, PhD '21, co-founder of SanaHeal, a medical device startup. "It's like an artificial tissue or nerve."

"We believe that for the first time, we have a tough, robust, Jell-O-like electrode that can potentially replace metal to stimulate nerves and interface with the heart, brain, and other organs in the body," adds Xuanhe Zhao, professor of mechanical engineering and of civil and environmental engineering at MIT.

hubie writes:

Researchers outline steps to keep shelves stocked in crises:

We all remember 2020. At the grocery store, toilet paper shelves were empty. Cleaning supplies and disinfectants were treasured finds. Rattled consumers, concerned that they would run out of essential items, swiftly stockpiled products until they disappeared from shelves. In the media, it was referred to as "panic buying."

[...] "Disaster-related buying behaviors or DRBBs is a more accurate phrase than panic buying," Holguin-Veras said. "Panic refers to extreme fear that causes someone to act irrationally, which negates the possibility to reason with these individuals to change their purchasing behaviors."

Instead, in research recently published, Holguín-Veras and team collected surveys in dozens of countries and found that, in the U.S., the top three themes were: precaution (25.79%), the anticipation of needs (22.63%) to avoid the possibility of regret if they didn't make the purchases (16.39%), and their interpretation of the actions of others, or social cues (13.44%).

They found that the key to altering consumer behavior to maintain adequate supply levels is trusted change agents, or TCAs. TCAs refer to the representatives of groups active in disaster perceived to be knowledgeable and trustworthy such as the Red Cross, national/state/local emergency responders, firefighters, and national/state/local health officials, that can reach large numbers of people, understand the need to influence the individuals that enact "panic buying," and are willing to act. A whopping 89% of people said that they would limit their purchases if asked by a TCA.

[...] Peoples' objectives in DRBBs ran the gamut. First, self-preservation was a primary concern. Second, altruism factored in as consumers wanted to provide for their loved ones and their communities. Third, opportunism reared its head, as people tried to cash in on opportunities to sell hard-to-find items at a premium. With this in mind, a one-size-fits-all approach will not work.

"Managing DRBBs requires effective public-private-humanitarian collaboration," said Holguin-Veras. "The public sector has the legal authority to intervene when needed, the private sector has control over the access and supplies, and the humanitarian sector has the deep community connections."

The sectors must work together in preparation, response, and assessment. Building trust and relationships, monitoring events and identifying situations that could prompt DRBBs, and examining legal frameworks are priorities during the preparation phase. During response, demand should be managed by working with private- and humanitarian- sector representatives, TCAs should be engaged, opportunistic purchases should be controlled to foster equitable access, and rationing should be instituted if needed. Appeals for the donations of stockpiled supplies are another measure that can be taken. Finally, a comprehensive assessment would facilitate any needed improvements.

Journal Reference:
José Holguín-Veras, Trilce Encarnación, Sofía Pérez-Guzmán, et al., The role and potential of trusted change agents and freight demand management in mitigating "Panic Buying" shortages, Transportation Research Interdisciplinary Perspectives 19 (2023) DOI: https://doi.org/10.1016/j.trip.2023.100792

Original Submission

hubie writes:

Could it mean we're judging those who aren't like us too harshly?

Sometimes life's most meaningful relationships grow from the briefest of connections. Like when you go to a party and meet someone wearing your favorite band's T-shirt, or who laughs at the same jokes as you, or who grabs that unpopular snack you alone (or so you thought) love. One small, shared interest sparks a conversation—that's my favorite, too!—and blossoms into lasting affection.

This is called the similarity-attraction effect: we generally like people who are like us. Now, new findings from a Boston University researcher have uncovered one reason why.

In a series of studies, Charles Chu, a BU Questrom School of Business assistant professor of management and organizations, tested the conditions that shape whether we feel attracted to—or turned off by—each other. He found one crucial factor was what psychologists call self-essentialist reasoning, where people imagine they have some deep inner core or essence that shapes who they are. Chu discovered that when someone believes an essence drives their interests, likes, and dislikes, they assume it's the same for others, too; if they find someone with one matching interest, they reason that person will share their broader worldview. [...]

[...] But Chu's research suggests this rush to embrace an indefinable, fundamental similarity with someone because of one or two shared interests may be based on flawed thinking—and that it could restrict who we find a connection with. Working alongside the pull of the similarity-attraction effect is a countering push: we dislike those who we don't think are like us, often because of one small thing—they like that politician, or band, or book, or TV show we loathe.

"We are all so complex," says Chu. "But we only have full insight into our own thoughts and feelings, and the minds of others are often a mystery to us. What this work suggests is that we often fill in the blanks of others' minds with our own sense of self and that can sometimes lead us into some unwarranted assumptions."

[...] "When you hear a single fact or opinion being expressed that you either agree or disagree with, it really warrants taking an additional breath and just slowing down," he says. "Not necessarily taking that single piece of information and extrapolating on it, using this type of thinking to go to the very end, that this person is fundamentally good and like me or fundamentally bad and not like me."

[...] But in a time when political division has invaded just about every sphere of our lives, including workplaces, the applications of Chu's findings go way beyond corporate horse trading. Managing staff, collaborating on projects, team bonding—all are shaped by the judgments we make about each other. Self-essentialist reasoning may even influence society's distribution of resources, says Chu: who we consider worthy of support, who gets funds and who doesn't, could be driven by "this belief that people's outcomes are caused by something deep inside of them." That's why he advocates pushing pause before judging someone who, at first blush, doesn't seem like you.

"There are ways for us to go through life and meet other people, and form impressions of other people, without constantly referencing ourselves," he says. "If we're constantly going around trying to figure out, who's like me, who's not like me?, that's not always the most productive way of trying to form impressions of other people. People are a lot more complex than we give them credit for."

Journal Reference:
Charles Chu and Brian S. Lowery, Self-Essentialist Reasoning Underlies the Similarity-Attraction Effect, PSPI, 2023. DOI: https://doi.org/10.1037/pspi0000425

Original Submission

JoeMerchant writes:

Orca rams into yacht off Shetland in first such incident in northern waters:

An orca repeatedly rammed a yacht in the North Sea off Shetland on Monday, in a concerning development following previous interactions between the cetaceans and vessels in the strait of Gibraltar and Portugal...

Experts believe this could be play among juvenile whales. Dr Alfredo López, of the Grupo de Trabajo Orca Atlántica in Portugal, said: "We know that many boats use fishing lines from the stern to fish and it is a motivation for orcas, they come to examine them." But the focus on boats' rudders may come from adult whales who have developed an aversion towards boats, perhaps because they "had a bad experience and try to stop the boat so as not to repeat it".

Most surprising is the fact that this learned behaviour should have appeared nearly 3,000 miles (4,800km) from Gibraltar. Dr Conor Ryan, a scientific adviser to the Hebridean Whale and Dolphin Trust, who has studied orca pods off the Scottish coast, said: "I'd be reluctant to say it cannot be learned from [the southern population]. It's possible that this 'fad' is leapfrogging through the various pods/communities."

How long before "whale watching" tour boats have to harden their propulsion / steering systems to survive Orca attacks?

Original Submission