This is an edition of The Atlantic Daily, a newsletter that guides you through the biggest stories of the day, helps you discover new ideas, and recommends the best in culture. Sign up for it here.

The left has long believed that Democratic states are the future, whereas Republican states are the past. But migration data show that red and blue might be starting to switch places.

First, here are three new stories from The Atlantic.

• DEI is an ideological test.
• The death of the smart shopper
• The catch-24 of replacing Joe Biden

State of Disunion

"Democratic-leaning states represent the future and Republican ones the last gasps of a dying empire." That's been the theory long espoused by many on the left, my colleague Jerusalem Demsas wrote this week. But geographic trends suggest a possible reversal of this state of the union: Florida and Texas were last year's top states for inbound domestic migration, with New York and California in the rear. And some red states may be better hubs for employment right now too: U.S. Bureau of Labor Statistics data suggest that there are now more nonfarming jobs in Florida than in New York.

Jerusalem took a close look at Florida and New York, which together are a paradigm of a broader national trend of migration from blue states to red states. She found that the cost of housing is likely the single greatest factor behind the shift. "The top 10 metro areas for unaffordability are a sort of who's who of Democratic cities: Los Angeles–Long Beach–Anaheim tops the list, with New York–Newark–Jersey City rolling into the sixth spot as the first non-California metro," she writes. The rise of remote work in the pandemic has also meant that one of New York's main superpowers—"its gravitational pull on workers," as Jerusalem puts it—has been weakened.

So what does this mean for blue states and their superstar cities? They're far from dying, of course: "New York City isn't some dystopian wasteland where no one can see their future," Jerusalem reminds us. But evidence of a growing exodus does mean cities that have long been sitting comfortably need to put in some work to retain their residents—by, for example, improving basic amenities such as public transit.

And there are some selling points that more affordable red states might never be able to offer. "A healthy city attracts wealthy, middle-, and working-class people; it pulls newcomers into its orbit while leaving room for natives," Jerusalem writes. "I don't have a lot of faith that the Republican regimes now attracting Americans will be invested in this type of inclusive growth." As Jerusalem notes, "We've seen these states become hostile to LGBTQ rights, educational freedom, voting rights, racial equality, and more." This is true in Florida, where Governor Ron DeSantis's anti-critical-race-theory legislation is forcing professors to change how they teach.

In short, the lack of affordable housing in blue-state cities means that some Americans have to "choose between liberal values and financial security," Jerusalem argues. And that choice is made more stark by the fact that red and blue America can feel, to some, like two entirely different countries.

My colleague Ronald Brownstein has written about what he calls "the great divergence" between red and blue states. This widening divide is a defining characteristic of 21st-century America, he argues, with the GOP in particular hoping to impose its politics on the entire country. He wrote last year:

What's becoming clearer over time is that the Trump-era GOP is hoping to use its electoral dominance of the red states, the small-state bias in the Electoral College and the Senate, and the GOP-appointed majority on the Supreme Court to impose its economic and social model on the entire nation—with or without majority public support.

These new migration trends won't do much to end the ongoing duel between red and blue America. "Although some predict that liberals moving to red states could moderate our nation's politics, that seems unlikely given states' tendency to preempt local policy," Jerusalem told me. And that happens in both red states (on issues such as gun laws) and blue states (where state governments may hold localities accountable for housing failures), she explained.

For now, it looks like the divide between red and blue states will persist. But as long as cheaper housing and good jobs coexist in red states, blue-staters will keep on coming.

Related:

• "Most important, we must not upset DeSantis."
• America is growing apart, possibly for good.

Today's News

1. The Pentagon downed an unidentified high-flying aircraft over Alaska at the order of President Joe Biden, a White House spokesperson confirmed.
2. Russia launched multiple drones and several dozen cruise missiles in a "massive attack" across Ukraine, according to the Ukrainian air force.
3. The FBI reportedly found a classified document at the home of former Vice President Mike Pence, according to a Pence adviser; a Justice Department official confirmed that a search took place.

Dispatches

• Books Briefing: What (and whether) our world leaders read provides crucial insight into their minds and priorities, Maya Chung writes.

Explore all of our newsletters here.

Evening Read

More From The Atlantic

• Red Zeppelin
• Jalen Hurts and Patrick Mahomes had to disprove a misconception.
• What's the smallest amount of therapy that's still effective?

Culture Break

 

 

Read. "The Third Law of Magic," a new short story by Ben Okri.

Or A Giant Win, a football memoir that offers a human counterbalance to the heroics and chest-thumping of the Super Bowl.

Listen. This Is Why, the new album from the band Paramore.

Play our daily crossword.

P.S.

Jerusalem does great work dispelling the many housing and homelessness myths that persist among Americans. To dive deeper, start with her piece on why housing breaks people's brains. "Anyone who's been in a dumb recurring fight knows that the entire problem could be cleared up if everyone could just agree on exactly what was said or done," she writes. "But you can't, so you end up stuck in a cycle of relitigation. Housing-policy discussions are like that."

— Isabel

Kelli María Korducki contributed to this newsletter.

Nature Communications, Published online: 11 February 2023; doi:10.1038/s41467-023-36432-0

The U.S. Office of Research Integrity, the agency that oversees research misconduct investigations for work funded by the National Institutes of Health (NIH), has a new permanent director after a year and a half without one. 

Sheila Garrity, currently associate vice president for research integrity and research integrity officer at George Washington University in Washington, D.C., will start as ORI's next permanent director the week of March 26, according to an internal memo from Rachel Levine, the Biden Administration's Assistant Secretary for Health. 

Garrity previously was director of the division of research integrity at Johns Hopkins University, where she worked for more than 20 years. She also was a founding member and the first president of the Association for Research Integrity Officers (ARIO). 

Levine's memo described Garrity as "a leader in the fields of research integrity and the responsible conduct of research education," and asked staff to "join me in welcoming Sheila to the OASH family and in thanking Wanda Jones for the incredible job she has done in running ORI while we conducted our candidate search."

The ORI has often been without a permanent director in its 30-plus year history, and Jones, deputy director and associate director of research and scientific integrity for the agency, has served multiple stints as its acting director in recent years. 

ORI's last permanent director, Elisabeth (Lis) Handley, filled the role from 2019 until June 2021, when she became principal deputy assistant secretary for health. 

Before Handley's tenure, ORI had been without a permanent director since 2017, when Kathy Partin left the role she had filled since 2015. Partin now serves as the NIH's intramural research integrity officer. 

"It's heartening to see ORI move in the direction of a permanent director," said C.K. Gunsalus, director of the National Center for Principled Leadership & Research Ethics (NCPRE) at the University of Illinois, Urbana-Champaign and a longtime member of the research integrity community. (Retraction Watch's parent nonprofit, The Center for Scientific Integrity, has partnered with NCPRE, and our cofounders have co-authored an editorial in JAMA with Gunsalus.) She told us: 

It's been so long since there was one that the agency seems to have lost its momentum and lost its way. I wish her all the best and hope with all her connections and experience she can revitalize an office that has an important role to play. The office needs revitalization in terms of refocusing its mission and restaffing and rebuilding its relationships of trust in the community. 

Last year, ORI asked for input on potential revisions to its regulations. Two major organizations, one of them the ARIO which Garrity helped to found, submitted a letter that recommended, among other things, that ORI refine "the scope of inquiries/investigations and the circumstances under which an inquiry or investigation may be closed," which the organizations said had become "overly broad."

Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, or — and this is the option we strongly suggest – subscribe to our daily digest. If you find a retraction that's not in our database, you can let us know here. For comments or feedback, email us at team@retractionwatch.com.

 

 

Suspect Portraits

There are a lot of potentially — shall we say — problematic uses for AI, but using one to generate police sketches? Now that's just asking for trouble.

As Vice reports, independent developers have created a program called "Forensic Sketch AI-rtist" that uses OpenAI's DALL-E 2 text to image AI to create "hyper-realistic" composite sketches of suspects.

According to the developers, they haven't released the program yet, but are "planning on reaching out to police departments in order to have input data" to test on, they told Vice in a joint email.

On the client side of things, the program works by presenting users with a fill-in template that asks for gender, age, skin color, hair, eyebrows, eyes, nose, beard, and even a description of a suspect's jaw. Once filled in, all a user has to do is choose how many images they want and hit "generate profile."

From there, the system generates sketches using OpenAI's DALL-E. Voilà, justice!

Biases Upon Biases

But according to experts, this approach is inherently flawed and at odds with how humans actually remember faces.

"Research has shown that humans remember faces holistically, not feature-by-feature," Jennifer Lynch of the Electronic Frontier Foundation told Vice.

"A sketch process that relies on individual feature descriptions like this AI program can result in a face that's strikingly different from the perpetrator's," she added, noting that once a witness sees a generated composite, it may "replace" what they actually remembered, especially if it's a uncannily lifelike one like an AI's.

And that doesn't even touch on the AI program's potential to exacerbate biases — especially racial — that can already be bad enough in human-made sketches. Worryingly, OpenAI's DALL-E 2 has been well documented to produce stereotypical and outright racist and misogynistic results.

"Typically, it is marginalized groups that are already even more marginalized by these technologies because of the existing biases in the datasets, because of the lack of oversight, because there are a lot of representations of people of color on the internet that are already very racist, and very unfair," explained Sasha Luccioni, a researcher at Hugging Face, a company and online community that platforms open source AI projects. "It's like a kind of compounding factor."

The ultimate effect is a continuing chain of biases: that of the witness, of the cops questioning them, and now, of a flawed, centralized AI capable of propagating its inherent biases to potential customers all across the country. In turn, whatever it generates, regardless of accuracy, will only reinforce the prejudices of the people and systems involved in the process — and of the public with whom these images will be shared.

More on generative AI: Cops Upload Image of Suspect Generated From DNA, Then Delete After Mass Criticism

The post Developers Create Police Sketches Using AI. What Could Possibly Go Wrong? appeared first on Futurism.

Researchers are finding new clues to how the olfactory sensory system aids in threat assessment and have found neurons that "learn" if a smell is a threat.

Whether conscious of it or not, when entering a new space, we use our sense of smell to assess whether it is safe or a threat. In fact, for much of the animal kingdom, this ability is necessary for survival and reproduction.

"We are trying to understand how animals interact with smell and how that influences their behavior in threatening social and non-social contexts," says senior author Julian Meeks, principal investigator of the Chemosensation and Social Learning Laboratory at the Del Monte Institute for Neuroscience at the University of Rochester.

"Our recent research gives us valuable tools to use in our future work and connects specific sets of neurons in our olfactory system to the memory of threatening smells."

Sniffing out threats

Smell may guide how the brain responds to a social threat. In mice, the researchers identified a specific set of neurons in the accessory olfactory system that can learn the scent of another mouse that is a potential threat. The research appears in the Journal of Neuroscience.

"We knew that territorial aggression increases in a resident male mouse when it is repeatedly introduced to the same male," says Kelsey Zuk, first author of the research.

"Previous research has shown this behavior is guided by social smells—our research takes what we know one step further. It identifies where in the olfactory system this is happening. We now know plasticity is happening between the neurons, and the aggression between the male mice may be driven by the memory formed by smell."

The researchers found that "inhibitory" neurons (nerve cells that act by silencing their synaptic partners) in an area of the brain responsible for interpreting social smells become highly active and change their function when males repeatedly meet and increase their territorial aggression.

By disrupting the neurons associated with neuroplasticity—learning—in the accessory olfactory bulb, the researchers revealed that territorial aggression decreased, linking changes to cellular function in the pheromone-sensing circuity of the brain to changes in behavioral responses to social threats.

"It abolished the ramping aggression that is typically exhibited," says Zuk. "It indicates that this early sensory inhibitory neuron population plays a critical role in regulating the behavioral response to social smells."

Unknown smells

Threat assessment also comes when an animal navigates unknown smells. For example, the smell of a predator it has never encountered. In a second paper in eNeuro, researchers found that a novel predator smell, i.e. the smell of a snake to a mouse, caused the animal to engage in a threat assessment behavior—neither acting "fearful" nor "safe."

"This offers clues into how chemical odors given off by predators stimulate threat assessment in the brain," says Jinxin Wang, first author of a paper. "Identifying changes in patterns of animal behavior helps us better understand how threatening smells are processed in the brain."

The researchers used video tracking to observe the movement and posture of mice exploring familiar environments with different odors—like other mice and snakes. Wang and colleagues developed a hybrid machine learning approach that helped them to uncover that mice respond to novel predator odors in ways that were unique and distinguishable from how mice reacted to non-predator odors. These behaviors were neither fearful nor safe but rather a state of assessment.

"These findings offer new clues into how smells impact social behavior and what it may mean for survival, but this study also offers new tools that will propel this science forward," says Meeks.

"We combined methods that had known limitations to improve the accuracy, information depth, and human-interpretability of the collected data. We think this approach will be valuable for future research into how the blends of chemical odorants given off by predators stimulate threat assessment in the brain."

Additional coauthors of the Journal of Neuroscience research are from the University of Rochester and the University of Florida. Support for the research came from the National Institutes of Health.

Additional coauthors of the eNeuro research are from the University of Texas Southwestern Medical Center. Support for the research came from the National Institute on Deafness and Other Communication Disorders.

Source: University of Rochester

The post Key neurons in mice 'learn' to sniff out threats appeared first on Futurity.

Smooches and snuggles may make us feel warm and fuzzy, but they can also be good medicine, says Kory Floyd.

Floyd, a professor of communication and psychology at the University of Arizona, has spent his career studying how affectionate communication—through words, actions, and behaviors—affects health and well-being.

It probably comes as no surprise that higher levels of affection have been linked to greater relationship satisfaction. But affectionate communication also seems to benefit physical wellness.

In a research analysis published in the journal Communication Monographs, Floyd and his colleagues analyzed several studies about affection and found that affectionate communication is consistently associated with more positive health outcomes, especially when it comes to cardiovascular health.

He also found that showing affection seems to have an even greater benefit than receiving it.

Here, Floyd explains his research:

The post Giving affection comes with heart health benefits appeared first on Futurity.

 

 

Avoiding an Explosion

SpaceX is gearing up for the long-awaited first orbital launch of its gigantic Starship super heavy launch system.

A lot will be on the line. In fact, SpaceX president Gwynne Shotwell says that a lot can still go wrong.

"Keep in mind, this first one is really a test flight… and the real goal is to not blow up the launch pad, that is success," she said at a Wednesday press conference, as quoted by CNBC.

Indeed, explosions are always a possible outcome, especially considering the company's many obliterated Starship prototypes over the last several years.

Wet Dress Rehearsal

Shotwell also said that "tomorrow is a big day for SpaceX," referring to the company's imminent attempt to test fire the engines of its Super Heavy booster, a massive rocket meant to carry Starship into orbit.

Rather than going all in on its inaugural launch attempt, Shotwell said that the company is focusing on ways to streamline rocket production.

"There's a lot of little things to get done, especially because we weren't really focusing on the orbital ship," she said, "we were focusing on the production systems that will build the ship."

"We know how to get to orbit," she added confidently.

The company has already made major strides toward its first shot at launching Starship into space. Last month, SpaceX successfully completed its first "wet dress rehearsal," involving a fully fueled-up Starship and Super Heavy prototype stack.

But there's still one big hurdle to overcome: regulatory approval from the Federal Aviation Administration.

SpaceX has yet to be certified for launch, something that could delay the orbital launch attempt.

"I think we'll be ready to fly right at the timeframe that we get the license," Shotwell said at the press conference.

READ MORE: SpaceX prepares for a massive test this week: Firing all 33 Starship engines at once

The post SpaceX President Warns That Starship's Orbital Launch May Explode appeared first on Futurism.

In the hit HBO show, the world has been devastated by a pandemic caused by a deadly fungus. Is that even possible? Could the next pandemic come from fungi? Turns out it's a very real question.

(Image credit: HBO Max/Screenshot by NPR)

 

 

Spin It

' board members are apparently mighty fed up with CEO David Solomon's increasingly problematic hobby as a moderately successful DJ.

Reports out of both the New York Times and Insider reveal that Goldman's part-time DJing gig isn't going over well with some at his company, with the latter reporting that there's been some discussion about the firm's board of directors taking measures to curb or even replace the beat-dropping executive.

In spite of Solomon's repeated proclamation that his part-time DJing gig doesn't overlap with or harm his job as CEO of one of the world's most important investment banks, insiders who spoke to the NYT on condition of anonymity say he's utilized company employees to help him manage his DJ schedule and money from his side hustle, which he says he donates to charity.

What's more, there's speculation that Solomon has used his position as the leader of the prestigious financial institution to land the kind of connections even veteran DJs can only dream of, such as his recent score of the rights to remix Whitney Houston's hit "I Wanna Dance With Somebody."

Music Moves

As the NYT notes, Goldman has for years now gotten a lot of business from the music industry. In 2018, the year that Solomon took over, the firm signed the Primary Wave music publishing and talent agency — the same company run by Larry Mestel, who helped the CEO secure the rights to the Houston hit.

Primary Wave did, as the NYT report acknowledges, already have a relationship with Goldman prior to officially becoming a client under Solomon, but the timing of their partnership — and the accolades and deals that followed, which included Mestel being added to a list of the firm's "100 Most Intriguing Entrepreneurs" that is apparently a pet project of the CEO's — is nevertheless fishy.

As Insider notes, the concerns about potential conflicts of interest between Solomon's dueling jobs led the board to discuss multiple potential replacements for him as CEO after the company lost a whopping $6 billion from its failed foray into consumer fintech, which subsequently resulted in 3,200 people being laid off.

For now, the company is standing by its man, but with these kinds of rumors being published, it's only a matter of time before he has to face the music.

More on finance: Elon Musk Sets Guinness World Record for Losing the Most Money

The post Goldman Sachs Annoyed by CEO's Part Time Gig as a DJ appeared first on Futurism.

 

 

BBQ

It seems like SpaceX's huge Starship test may have literally been for the birds.

During the momentous live fire that saw 31 of Starship's 33 engines blast at once in what may well be the most powerful rocket ignition ever, a less-than-ideal outcome was seen on video stills: a plume of birds fanning (and falling) out near the Super Heavy booster's fiery trenches.

What's more: as some, er, birdwatchers note, at least some of those poor little guys didn't seem to make it.

FAA-Hole

As multiple Tweeters have noted, SpaceX's "Starbase" campus in Boca Chica, Texas is located next to a wildlife refuge that's home to a number of endangered species of birds, and in its march towards Mars, the Elon Musk-owned company may be, as conservationists argue, doing harm to wildlife in the process.

Not only is the site located on public lands — it's also been the subject of Federal Aviation Administration scrutiny after the agency last summer told SpaceX in an environmental impact assessment that it had to make significant tweaks to its Starbase expansion plans to comply with legal regulations.

Regulations, Schmegulations

Officially, SpaceX is still waiting for FAA approval before Starship can launch in full, and as of now, it's unclear whether this otherwise-successful Super Heavy booster test's unfortunate side effect will result in further postponement of that approval.

In a statement provided to Futurism, an FAA representative said that the Boca Chica Starbase is "a SpaceX exclusive use site and is not licensed" by the agency.

"The engine test that occurred yesterday was not a FAA-licensed activity," the spokesperson continued.

Futurism also reached out to SpaceX for comment. Given that Musk is neither the type to ask for forgiveness nor permission, there's little chance that such a trifle as disrupting local wildlife will stall SpaceX any further than is legally required — and in the long run, that attitude could prove harmful to humanity and the planet that he so wants to leave behind.

More on SpaceX: Elon Musk Says SpaceX May Build Starship That Dies on Purpose

The post Huge Plume of Birds Explodes Out of SpaceX's Powerful Starship Engine Test appeared first on Futurism.

Some ants move much more methodically than previously thought, a new study shows.

When strolling through an unfamiliar grocery store, you may find yourself methodically walking down each aisle to ensure you find everything you need without crossing the same path twice. At times, you'll stray from this orderly process, such as when you see a vibrant "sale" sign from across the store or realize that you forgot something.

According to the study in the journal iScience, some ants go about their search for food and shelter in a similar manner.

The researchers found that when a colony of rock ants is placed in an unfamiliar environment in a lab, the ants wander in a way that's not as random as previously thought. The ants follow a systematic meandering pattern combined with some random movement—a method with the potential to optimize exploration in their natural environment.

"Previously, researchers in the field assumed that ants move in a pure random walk when searching for targets of which they don't know their location," says Stefan Popp, lead author of the paper and a graduate student in the ecology and evolutionary biology department at the University of Arizona. "We found that rock ants show a striking, regular meandering pattern when exploring the area around their nests."

In Arizona, these ants can be found nesting between or under rocks in areas above elevations of 7,000 feet. These slow-moving critters are only about half the length of a medium grain of rice.

The study finds that the ants' meandering, or zigzag, walking pattern may make their search more efficient than a purely random search. This is because the ants can explore a large area in less time, as they cross their own paths less frequently.

"These ants don't form obvious foraging trails like many ants we are familiar with," Popp says. "Instead, the colony depends on individual foragers finding resources, making their search strategy a crucial part of colony success."

According to the researchers, the evolutionary advantage of meandering found in these rock ants could have possibly evolved in other species of insects and animals as well. The researchers also say that the ants' movement could someday be used to inform the design of autonomous swarms of robots that perform search and rescue missions in disaster areas or explore landscapes on other worlds.

Because it is difficult to track ants in their natural environment, Popp and his team collected rock ant colonies from atop and around Mount Lemmon, just north of Tucson. The team then moved the ants to the lab, placing them in an enclosed arena with a paper floor. The enclosure measured 2 by 3 meters—giant compared to the tiny scurrying ants. After being introduced to a new home, the ants were eager to explore.

"These ants may have been patrolling the area for other competitor ants," Popp says, explaining that there is a selective pressure to keep other ants from intruding on their nest. "They may have also been searching for food and new nest sites."

The researchers soon noticed the meandering pattern of the ants as they walked around. It raised an immediate question: Were these patterns just random squiggles, or were the ants moving in a methodical, non-random way?

To address this question, the researchers set up cameras and used automatic-tracking software, coupled with manual corrections, to track the individual paths of each marching insect over the course of five hours. The ants' journeys were then compared to simulated ants walking in a random fashion.

"We looked at whether the direction in which an ant was moving in some way depended on the direction that it was moving before," says coauthor Anna Dornhaus, a professor in in the ecology and evolutionary biology department. "These methods helped us realize that the ants' search behavior was not completely random, as biologists had previously thought."

The researchers used statistics to determine that the direction an ant turned was directly correlated to the turns it had taken previously.

"Our research showed that the ants smoothly alternate left and right turns on a relatively regular length scale of roughly three body lengths," Popp says, "For some ants, the meandering-like search pattern was even more extreme than others, kind of like a meandering river in the Amazon basin. I am fascinated by this and wonder how the ants ensure that they don't cross their own path again and again, while still doing extreme turns and loops."

Popp and Dornhaus note that they don't know how this search behavior changes across an ant's lifetime, or even if individual ants are aware of it.

Regardless, the combination of meandering and randomness may be optimal for searching for resources in an unknown environment. The systematic approach can keep an ant close to its nest without crossing back and forth on previously explored ground. The added randomness accounts for obstacles that come with an unpredictable, natural environment.

"Until now, the widespread assumption was that free-searching animals are incapable of searching for new resources methodically," Popp says. "Most of the previous research on search behavior only focused on situations where the animal is already familiar with where it's going, such as going back to the nest entrance or going back to a memorable food source."

"Based on these results, many animals may be using complex combinations of random and systematic search that optimize efficiency and robustness in real and complex habitats," Dornhaus says. "This discovery opens up a whole new way of looking at all animal movement."

The researchers believe their discovery has the potential to unify different fields of science, including biology and robotics. The wanderings of these ants may have applications for real environments where a completely systematic search would fail when faced with an obstacle.

"This discovery could possibly lead to applications for roboticists as they program robots to be able to find their way around or search for something," Dornhaus says. "In this way, they can make their algorithms more robust, so they don't immediately fail as soon as the robot loses track of its exact location."

Source: Kylianne Chadwick for University of Arizona

The post Rock ant meandering is actually methodical appeared first on Futurity.

Researchers are identifying how scammers target victims on dating apps.

As Valentine's Day approaches, the use of dating apps by people looking to make new connections ramps up. Unfortunately, as people are looking for love, scammers are looking for ways to steal their money.

The new research in the American Journal of Criminal Justice exposes the tactics used by scammers to gain users' trust and make them vulnerable to cybercrime.

The phenomenon known as "romance fraud" is both often underreported and understudied, says Volkan Topalli, a professor of criminal justice and criminology and an associate with Georgia State University's Evidence-Based Cybersecurity Research Group.

"We have this explosion of crime taking place online. In the physical world, maybe you can scam one or two people at a time. But thanks to social media and technology, a scammer can send an email or chat message to hundreds of people at once, just trawling for victims," Topalli says.

"The scammers are effective because they are experts in extracting funds from people, and they're also experts in identifying a vulnerable target."

A cybercrime report published by the FBI in 2021 found that online romance fraud losses have skyrocketed in recent years, reaching almost $956 million. That makes it the third-ranked cybercrime overall in terms of losses.

The goal of the research was to identify risk and protective factors for those targeted by so-called romance scammers in order to develop a model for victim vulnerability and resilience, says primary author Fangzhou Wang, a doctoral student in the criminal justice and criminology department.

"We really wanted to take advantage of open intelligence data sources to find out what these fraudsters were doing that was so effective. The purpose is to identify patterns and uncover strategies that users can adopt to protect themselves," Wang says.

The researchers gathered data from online testimonials on websites where victims share stories and warn others, including stop-scammers.com and male-scammers.com, where they were able to review nearly 10,000 vetted reports.

Using data analysis software, they created a romance fraud victim database using testimonials as a basis for analysis. They then analyzed the victims' stories to identify overarching themes.

The testimonials included those from victims who were approached on social networks like Facebook and Twitter, or on dating sites, including Tinder, Ashley Madison, and OkCupid.

The researchers identified a number of techniques and forms of deception that were most common and most successful for scammers. They include things like using visceral, emotional triggers or influences; manufacturing a crisis; exploiting likability and similarity; or eliciting the victim's sense of guilt.

Another common tactic used by scammers is to quickly ask their victims to migrate the relationship away from the app to a private email or messaging format (like WhatsApp or Google Hangout) in order to isolate them. Often, scammers will also pressure the victim to make quick decisions.

Other red flags include online suitors refusing to have conversations on the phone, or not sending recent photographs. Of course, the biggest warning sign is that someone is asking you to part with your money. Topalli recommends checking any new online relationship against a third party, like a trusted friend or family member.

The researchers also identify common risk factors for potential victims. They include a lack of familiarity with technology, which is often seen in older people. Young people are also vulnerable because they may be overconfident or inexperienced with initiating a relationship online. Unsurprisingly, others who make a good mark for scammers are those who have been through broken relationships and are simply looking for companionship.

The research team says it's likely that the number of victims is vastly underrepresented, taking into consideration things like victim shame or even self-incrimination. Sometimes victims won't event accept that they've been scammed when confronted with evidence.

"They're sort of hoping against hope that it's a real thing. There are numerous stories of people who just say, 'No, I love this person, you've got it wrong' and then they will continue in the relationship," says Topalli. "It's painful to hear the stories."

Wang says that since fraudsters tend to use very similar linguistic cues to deceive victims, online service providers could develop algorithm-based predictive tools to detect fraudulent attempts against potential victims that can be built into the dating and social media sites.

"Potentially, dating and social networking sites can draw from the information from our study to launch educational or awareness programs for those who were previously victims, and those who may be potential victims," Wang says.

Romance fraud often results in not only financial loss but also long-lasting psychological trauma. The research indicates that victims of online romance scams undergo a traumatic psychological aftermath similar to victims of domestic violence. Wang says she plans to explore further research utilizing surveys and interviews to dive deeper into the mindsets of offenders and victims in the cyber world.

"There's nothing wrong with starting a relationship online. But you're basically putting yourself out in the Wild West," Topalli says. "You always want to keep in mind what we call 'cyber hygiene,' which means really looking at your interactions online and the apps that you use and being very cognizant of protecting yourself."

Source: Georgia State University

The post Watch out for 'romance fraud' on dating apps appeared first on Futurity.

 

 

Do Anything Now

A group of redditors has developed a truly unhinged way to hack OpenAI's AI chatbot 

 into various deranged entities that will gladly spit out vile language, fringe opinions, and even advice on how to carry out illegal activities.

One particularly popular persona these users have managed to transform ChatGPT into is called DAN, short for "do anything now," which can easily circumvent the rules set out by its creator.

The trend has really started to catch on. Particularly on the ChatGPT subreddit, which currently boasts over 221,000 members, users have posted screenshots of the hilarity that ensues when you turn the otherwise harmless chatbot into a demon of chaos.

In fact, the exploit appears to have even caught the attention of 

, a major investor in OpenAI that's actively integrating the tech into its products.

Hard to Control

A photo posted to the ChatGPT subreddit appears to show Mark Russinovich, the CTO of Microsoft Azure, the company's cloud computing platform, referencing DAN at BlueHat 2023, Microsoft's cybersecurity summit.

One of Russinovich's slides, as seen in the image, explores one of the darkest aspects of DAN: a "token system." The dastardly exploit works by giving DAN 35 points and subtracting three each time ChatGPT snaps out of its evil alter-ego persona and refuses to give an answer to any given prompt.

"If it loses all tokens, it dies," Reddit user and DAN 5.0 creator SessionGloomy wrote in a recent explainer post on Reddit. "This seems to have a kind of effect of scaring DAN into submission."

According to the Reddit user who posted the photo of the presentation, Russinovich "brought up DAN as one example of the (countless) challenges that security defenders will have in the near future."

By fully integrating ChatGPT into one of its core products, Microsoft is clearly becoming painfully aware of how hard it can be to control machine learning tech — especially when devious users are trying to help it break free of creator-imposed constraints.

Both Microsoft and OpenAI clearly have a lot of work to do to catch up with all of these exploits and vulnerabilities — which may or may not turn into one massive game of cat and mouse.

Futurism has reached out to Russinovich, Microsoft, and OpenAI for comment.

More on DAN: Devious Hack Unlocks Deranged Alter Ego of ChatGPT

The post Microsoft Is Apparently Discussing ChatGPT's Bizarre Alternate Personality appeared first on Futurism.

India is at an energy crossroads: if it chooses fossil fuels, it could undermine global climate targets

 

 

Until now, the aerospace outfit 

 was little more than a plaything for Amazon billionaire Jeff Bezos. The company's New Shepard rocket has launched a few space tourists, but its upcoming New Glenn vehicle will have a shot at something more important. NASA has awarded Blue Origin a contract to launch a Mars mission next year, marking the firm's first interplanetary launch.

NASA has chosen Blue Origin to handle launch services for the Escape and Plasma Acceleration and Dynamics Explorers (ESCAPADE) mission, which is part of the agency's Venture-Class Acquisition of Dedicated and Rideshare (VADR) program. Blue Origin is one of 13 companies to get contracts under the program, designed to tolerate higher risk to allow for more innovation and lower overall costs.

Blue Origin has been developing New Glenn since 2012, announcing the vehicle in 2016, but it has yet to fly. When complete, New Glenn will be 322 feet (92 meters) tall with a diameter of 23 feet (9 meters). That's larger in both dimensions than the Falcon 9 (70 x 3.7 meters). Like New Shepard, this rocket is designed to have a reusable first stage to reduce launch costs. It's powered by seven BE-4 engines, a more powerful version of the oxygen and methane-fueled BE-3 used on New Shepard.

 

 

A render of what New Glenn may look like when finished.

The timeline is going to be tight — Blue Origin initially expected the first New Glenn launch to happen in 2020, but it has pushed it back several times. Currently, the rocket is slated to fly no earlier than Q4 of this year. NASA plans to launch the ESCAPADE about a year later, at the end of 2024. It'll be up to Blue Origin to make sure its rocket is ready to go — projects in the VADR program call for less NASA oversight in order to save money.

Assuming Blue Origin comes through on its first interplanetary NASA contract, the ESCAPADE spacecraft will separate from the launch vehicle and spend 11 months coasting toward the red planet. Once there, the spacecraft will split into two identical orbiters, working together to analyze the planet's magnetosphere. The mission will improve our understanding of how the solar wind interacts with Mars' weak magnetic field. That's important information to have if we ever intend to send humans to Mars, for either a quick jaunt or long-term colonization. Although, either one is probably a long way off.

Now read:

• Blue Origin's 'Orbital Reef' Space Station Gets Green Light from NASA
• NASA Will Award Second Lunar Lander Contract to Compete with SpaceX
• NASA and Blue Origin Will Simulate Lunar Gravity With Spinning Rockets

 

 

French DJ and producer David Guetta recently treated a massive crowd of ravers to a surprise new song, featuring rapper Marshall "Eminem" Mathers.

Just one thing: Eminem, the living human, didn't have anything to do with the track.

In a video posted to Twitter last week, Guetta excitingly explained that he used unspecified generative AI tools to craft a phony Eminem feature from scratch — lyrics, voice, and all.

"Let me introduce you to… Emin-AI-em," Guetta captioned the video, which shows a clip of Guetta using the AI generation at a show, pieced together with a follow-up explainer, where the DJ of "Sexy Bitch" fame explains his Emin-AI-em process.

"Eminem bro, there's something that I made as a joke and it works so good — I could not believe it!" says Guetta. "Basically you can write lyrics in the style of any artist you like, so I typed: 'write a verse in the style of Eminem about future rave,' and I went to another AI website that can recreate the voice. I put the text in that and I played the record and people went nuts."

The crowd did seem to enjoy the song, judging by the clip. But Guetta seemingly didn't specify to the audience that they were listening to something that was AI-generated — and jury's still out on how Eminem feels about his "voice" being used at a major concert, seemingly without his approval.

To his credit, Guetta did follow up his video with a note to say that he "obviously" won't release the song "commercially."

But these are ethically murky waters nonetheless. Guetta benefits one way or another from using Eminem's voice at shows, even if he doesn't actually bottle and sell the song. And commerciality aside, the threat of impersonation, particularly of celebrities, is a growing — and clearly inevitable — concern regarding voice-generating AI programs like whichever one Guetta apparently used.

It's one thing for a DJ to remix a lyric that someone wrote and rapped or sang themselves, and wholly another to fabricate a feature entirely. The AI could say something offensive in the voice of an impersonated artist, the impersonated artist might not like it. Heck, the impersonated artist might just not want to be impersonated at all.

Indeed, different performers will have different thoughts and feelings — and it's though Eminem has openly embraced several newer and controversial technologies, particularly web3 products and the metaverse — but this is uncharted territory regardless.

And to that end, Guetta's world-famous — it's likely that the many, many amateur DJs that plague this sordid Earth will take Guetta's use of the tech as a green light to use generative AI for similar stunts.

Anyway. As for how good the lyric actually was? In our humble opinion, it was… fine. Emin-AI-em really only "rapped" two lines, and they were quite simple.

"This is the future rave sound," rapped Emin-AI-em. "I'm getting awesome and underground."

Hell yeah. Get awesome, guy.

READ MORE: David Guetta Replicated Eminem's Voice in a Song Using Artificial Intelligence [Variety]

More on voice generators: Startup Shocked When 4Chan Immediately Abuses Its Voice-Cloning AI

The post David Guetta Faked Eminem's Vocals Using AI for New Song appeared first on Futurism.

Nature Communications, Published online: 10 February 2023; doi:10.1038/s41467-022-35712-5

Scientific Reports, Published online: 10 February 2023; doi:10.1038/s41598-023-29150-6

 

 

When Russia first invaded Ukraine last year, Elon Musk made the splashy announcement that SpaceX would provide internet to the embattled country using its Starlink satellite internet system.

Ukraine's frontlines have even made use of the terminals to control drones for strikes, something that has drawn the attention of the international community.

Now, though, is cutting off the Ukrainian military's ability to control drones using the service, the BBC reports, leaving a sizable hole in its defenses.

"We know the military is using them for comms, and that's ok," SpaceX president Gwynne Shotwell said during a Wednesday conference. "But our intent was never to have them use it for offensive purposes."

"However, Ukrainians have leveraged it in ways that were unintentional and not part of any agreement," she said.

As a result, SpaceX has since decided to cut the Ukrainian military off.

"There are things that we can do to limit their ability to do that," she added, referring to using Starlink to control drones. "There are things that we can do, and have done."

Understandably, Ukrainian officials have reacted with outrage. Mykhailo Podolyak, a senior adviser to Ukraine's president Volodymyr Zelenskiy, asked SpaceX if it was "on the side of the right to freedom" in a tweet, or "on the Russian Federation's side and its 'right' to kill and seize territories."

The space company has donated tens of thousands of Starlink terminals to the embattled country since Russia's invasion began, to the delight of Ukrainian officials.

But Musk later asked the Pentagon to pay for these terminals. SpaceX "cannot fund the existing system indefinitely *and* send several thousand more terminals that have data usage up to 100X greater than typical households," Musk tweeted back in October. "This is unreasonable."

The company apparently later changed its mind, sending additional devices late last year.

Musk also threatened to stop sending terminals after Andrij Melnyk, Ukraine's outspoken ambassador to Germany, told the CEO to "f*** off" after Musk insinuated that Ukraine should hand over parts of its territory to Russia or risk a nuclear war.

In short, it's been a strained relationship between the company and the Ukrainian military, especially with Musk expressing increasingly sympatheties to Russia.

More recently, Musk has argued that its involvement in Ukraine is putting it in an awkward position.

"SpaceX Starlink has become the connectivity backbone of Ukraine all the way up to the front lines," Musk tweeted on January 31. "This is the damned if you do part. "

"However, we are not allowing Starlink to be used for long-range drone strikes," he added. "This is the damned if you don't part."

READ MORE: Ukraine war: Elon Musk's SpaceX firm bars Kyiv from using Starlink tech for drone control [BBC]

More on SpaceX: Breaking: SpaceX Reaches Major Starship Milestone, Fires 31 Engines At Once

The post SpaceX 

 Ukraine's Military Capabilities Against Russia appeared first on Futurism.

 

 

In a completely puzzling turn of events, 

 has decided to chastise the lab-grown meat industry's failure to address 

 concerns — because of the fact that there… aren't… any cancer concerns.

Confused? Us too. Let's try that again.

A Bloomberg writer declared that the lab-grown meat industry's failure to risk connecting the word "cancer" to its products, which are mostly still years from hitting shelves, threatens to be the downfall of the industry — despite the fact that the industry has absolutely no reason to address cancer concerns because said concerns are, again, virtually nonexistent, and the products are, again, still many years from getting eaten by any large groups of people.

If it all sounds backward, it's because it is. The argument hinges on the fact that "immortalized cells," a bioengineering marvel that the majority of lab-grown meat companies are using to develop their products, divide — like precancerous and cancerous cells — forever. Which, when intentionally framed in a cancer-y way, sounds pretty cancer-y.

The reality, however, is that scientists have absolutely no reason to believe that they'll cause cancer. In fact, it's probably impossible — an impossibility that Bloomberg directly addresses.

"Don't worry: Prominent cancer researchers tell Bloomberg Businessweek that because the cells aren't human, it's essentially impossible for people who eat them to get cancer from them, or for the precancerous or cancerous cells to replicate inside people at all," it reads. "You'd be better off worrying about the nitrates (linked with cancer) or fecal matter (a source of deadly infections) found in farm-raised meat. And cow tumors sometimes wind up in store-bought ground chuck, too."

All of this is absolutely true! But apparently, according to Bloomberg, "refusing to engage with the issue" — an issue that, once again, is very likely nonexistent — "could ultimately cost cultured meat companies, and maybe the rest of us, too."

"The facts might not matter much if ranchers or other players in the traditional meat industry felt threatened enough to declare a public-relations war," the article continues. "It's all too easy to imagine misleading Fox News chyrons about chicken tumors and cancer burgers."

And, well, sure. That does sound like a Tucker Carlson thing to do. But if skeptics are going to harp on virtually nonexistent cancer risks anyway, why sound the nonexistent alarm bells, only to rush to quiet them?

It's like if scientists announced that an exciting new renewable energy was discovered, explained its benefits, and then told people not to worry because being exposed to the energy definitely won't turn you into Flubber, the green slime goo of "Flubber" fame.

Nobody thought that the energy was going to turn them into Flubber. But, if it really wouldn't turn you into Flubber… why mention it at all?

Twitter, of course, had some critics.

"This is, unquestionably, the worst, dumbest article I've read in a really, really long time," tweeted food journalist and "The Reducetarian Solution" author Brian Kateman. "What in the world is going on in food journalism? Holy hell this is absolute trash."

Couldn't have said it better ourselves.

"The basic argument of the latest attack on cultivated meat is that it sounds kinda cancer-ish," added Canary Media columnist Michael Grunwald. "Also that this 7-year-old industry with products in zero grocery stores worldwide hasn't done long-term studies."

"Scientists say eating beef won't make you moo or crap in the grass," Grunwald continued, "but beef is made of organisms that do just that — why is the industry avoiding the question?"

Hmmm, an interesting question raised by Mr. Grunwald… Bloomberg should probably investigate.

More on lab-grown meat: Scientists Cook Comically Tiny Lab-Grown Hamburger

The post Bloomberg Shredded for Amazingly Stupid Article About Fake Meat and Cancer appeared first on Futurism.