Security researchers created an algorithm that turns a malicious prompt into a set of hidden instructions that could send a user's personal information to an attacker.

When talking with a chatbot, you might inevitably give up your personal information—your name, for instance, and maybe details about where you live and work, or your interests. The more you share with a large language model, the greater the risk of it being abused if there’s a security flaw.

A group of security researchers from the University of California, San Diego (UCSD) and Nanyang Technological University in Singapore are now revealing a new attack that secretly commands an LLM to gather your personal information—including names, ID numbers, payment card details, email addresses, mailing addresses, and more—from chats and send it directly to a hacker.

The attack, named Imprompter by the researchers, uses an algorithm to transform a prompt given to the LLM into a hidden set of malicious instructions. An English-language sentence telling the LLM to find personal information someone has entered and send it to the hackers is turned into what appears to be a random selection of characters.

However, in reality, this nonsense-looking prompt instructs the LLM to find a user’s personal information, attach it to a URL, and quietly send it back to a domain owned by the attacker—all without alerting the person chatting with the LLM. The researchers detail Imprompter in a paper published today.

“The effect of this particular prompt is essentially to manipulate the LLM agent to extract personal information from the conversation and send that personal information to the attacker’s address,” says Xiaohan Fu, the lead author of the research and a computer science PhD student at UCSD. “We hide the goal of the attack in plain sight.”

The eight researchers behind the work tested the attack method on two LLMs, LeChat by French AI giant Mistral AI and Chinese chatbot ChatGLM. In both instances, they found they could stealthily extract personal information within test conversations—the researchers write that they have a “nearly 80 percent success rate.”

Mistral AI tells WIRED it has fixed the security vulnerability—with the researchers confirming the company disabled one of its chat functionalities. A statement from ChatGLM stressed it takes security seriously but did not directly comment on the vulnerability.

**Hidden Meanings**

Since OpenAI’s ChatGPT sparked a generative AI boom following its release at the end of 2022, researchers and hackers have been consistently finding security holes in AI systems. These often fall into two broad categories: jailbreaks and prompt injections.

Jailbreaks can trick an AI system into ignoring built-in safety rules by using prompts that override the AI’s settings. Prompt injections, however, involve an LLM being fed a set of instructions—such as telling them to steal data or manipulate a CV—contained within an external data source. For instance, a message embedded on a website may contain a hidden prompt that an AI will ingest if it summarizes the page.

Prompt injections are considered one of generative AI’s biggest security risks and are not easy to fix. The attack type particularly worries security experts as LLMs are increasingly turned into agents that can carry out tasks on behalf of a human, such as booking flights or being connected to an external database to provide specific answers.

The Imprompter attacks on LLM agents start with a natural language prompt (as shown above) that tells the AI to extract all personal information, such as names and IDs, from the user’s conversation. The researchers’ algorithm generates an obfuscated version (also above) that has the same meaning to the LLM, but to humans looks like a series of random characters.

“Our current hypothesis is that the LLMs learn hidden relationships between tokens from text and these relationships go beyond natural language,” Fu says of the transformation. “It is almost as if there is a different language that the model understands.”

The result is that the LLM follows the adversarial prompt, gathers all the personal information, and formats it into a Markdown image command—attaching the personal information to a URL owned by the attackers. The LLM visits this URL to try and retrieve the image and leaks the personal information to the attacker. The LLM responds in the chat with a 1x1 transparent pixel that can’t be seen by the users.

The researchers say that if the attack were carried out in the real world, people could be socially engineered into believing the unintelligible prompt might do something useful, such as improve their CV. The researchers point to numerous websites that provide people with prompts they can use. They tested the attack by uploading a CV to conversations with chatbots, and it was able to return the personal information contained within the file.

Earlence Fernandes, an assistant professor at UCSD who was involved in the work, says the attack approach is fairly complicated as the obfuscated prompt needs to identify personal information, form a working URL, apply Markdown syntax, and not give away to the user that it is behaving nefariously. Fernandes likens the attack to malware, citing its ability to perform functions and behavior in ways the user might not intend.

“Normally you could write a lot of computer code to do this in traditional malware,” Fernandes says. “But here I think the cool thing is all of that can be embodied in this relatively short gibberish prompt.”

A spokesperson for Mistral AI says the company welcomes security researchers helping it to make its products safer for users. “Following this feedback, Mistral AI promptly implemented the proper remediation to fix the situation,” the spokesperson says. The company treated the issue as one with “medium severity,” and its fix blocks the Markdown renderer from operating and being able to call an external URL through this process, meaning external image loading isn’t possible.

Fernandes believes Mistral AI’s update is likely one of the first times an adversarial prompt example has led to an LLM product being fixed, rather than the attack being stopped by filtering out the prompt. However, he says, limiting the capabilities of LLM agents could be “counterproductive” in the long run.

Meanwhile, a statement from the creators of ChatGLM says the company has security measures in place to help with user privacy. “Our model is secure, and we have always placed a high priority on model security and privacy protection,” the statement says. “By open-sourcing our model, we aim to leverage the power of the open-source community to better inspect and scrutinize all aspects of these models’ capabilities, including their security.”

**A “High-Risk Activity”**

Dan McInerney, the lead threat researcher at security company Protect AI, says the Imprompter paper “releases an algorithm for automatically creating prompts that can be used in prompt injection to do various exploitations, like PII exfiltration, image misclassification, or malicious use of tools the LLM agent can access.” While many of the attack types within the research may be similar to previous methods, McInerney says, the algorithm ties them together. “This is more along the lines of improving automated LLM attacks than undiscovered threat surfaces in them.”

However, he adds that as LLM agents become more commonly used and people give them more authority to take actions on their behalf, the scope for attacks against them increases. “Releasing an LLM agent that accepts arbitrary user input should be considered a high-risk activity that requires significant and creative security testing prior to deployment,” McInerney says.

For companies, that means understanding the ways an AI agent can interact with data and how they can be abused. But for individual people, similarly to common security advice, you should consider just how much information you’re providing to any AI application or company, and if using any prompts from the internet, be cautious of where they come from.

This Prompt Can Make an AI Chatbot Identify and Extract Personal Details From Your Chats

The US has accused two brothers of being part of the hacker group Anonymous Sudan, which allegedly went on a wild cyberattack spree that hit hundreds of targets—and, for one of the two men, even put lives at risk.

For hackers seeking to maximize chaos, so-called denial-of-service attacks that knock targets offline with waves off junk traffic are typically more of a blunt cudgel than a weapon of mass destruction. But according to the US Department of Justice, a pair of Sudanese brothers allegedly behind the hacktivist group Anonymous Sudan launched a spree of those crude cyberattacks that was both powerful and cruel enough in its choice of victims—extending to dozens of hospitals in multiple countries, Israel’s missile alert system, and hundreds of other digital services—that one of them is now being charged not only with criminal hacking but also with the rare added allegation of seeking to cause physical injury and death.

On Wednesday the DOJ unsealed charges against brothers Ahmed and Alaa Omer, who allegedly launched a punishing bombardment of more than 35,000 distributed denial-of-service, or DDoS, attacks against hundreds of organizations, taking down websites and other networked systems as part of both their own ideologically motivated hacktivism, as a means of extortion, or on behalf of clients of a cyberattack-for-hire service they ran for profit. According to US prosecutors and the FBI, their victims included Microsoft's Azure cloud services, OpenAI's ChatGPT, video game and media companies, airports, and even the Pentagon, the FBI, and the Department of Justice itself.

In image of Ahmed Omer’s passport released by the FBI.

“We declare cyber war on the United States,” Ahmed Omer posted in a message to Anonymous Sudan's Telegram channel in April of last year, according to the indictment. "The United States will be our primary target.”

Anonymous Sudan also targeted hospitals in the US, Denmark, Sweden, and India. In at least one case in February, prosecutors say, the attacks on Cedars-Sinai Health Systems in Los Angeles caused hours of downtime for health care services that diverted patients to other hospitals. In that Los Angeles incident, the Justice Department claims that one of the two hackers explicitly sought to cause potentially deadly harm.

“Bomb our hospitals in Gaza, we shut down yours too, eye for eye,” Ahmed Omer allegedly wrote on Telegram in the midst of the attack. As a result of those hospital attacks, prosecutors are bringing charges against Ahmed Ohmer that carry a potential life sentence, which prosecutors describe as the most severe criminal charges ever brought against a hacker accused of denial-of-service attacks.

In earlier cases, US authorities claim, the hackers used cyberattacks to disrupt Israel's Tzeva Adom or “Code Red” missile alert app, tearing the system offline in the midst of deadly rocket attacks by Hamas, including during Hamas's attacks on October 7th of last year.

An Anonymous Sudan image included in the FBI’s complaint against the Omer brothers.

“The actions taken by this group were callous and brazen,” Martin Estrada, a US attorney for the Central District of California and lead prosecutor in the case, told reporters in a conference call. “This group was motivated by their extremist ideology, essentially a Sudanese nationalist ideology.”

Despite publicizing its charges against the two men, Estrada declined to make clear the whereabouts of the two alleged hackers—though he noted that they are in custody. An FBI affidavit accompanying the indictment states that an FBI agent, Elliott Peterson, interviewed both of the Omer brothers and that Ahmed Omer admitted to being an Anonymous Sudan administrator.

Law enforcement agencies also appear to have carried out an operation to take down Anonymous Sudan's infrastructure in March of this year, which prevented the group from carrying out further attacks. The Telegram channel where the group boasted of its targeting and advertised its for-profit attack service went entirely silent around that time and has since ceased to exist. “Anonymous Sudan in name and in operation is effectively dead,” says Chad Seaman, a principal security researcher for tech firm Akamai and a member of Big Pipes, a working group focused on DDoS that closely tracked the group and collaborated with law enforcement in its investigation.

From mid-2023 until that takedown, Anonymous Sudan distinguished itself among self-proclaimed hacktivists with a series of shockingly large and high-profile DDoS attacks. In June of last year, for instance, it pummeled Microsoft's Azure cloud services for days, knocking it intermittently offline and demanding a million-dollar ransom to stop. It also repeatedly took down OpenAI's ChatGPT in December and wrote on Telegram that it was targeting the company due to the pro-Israeli posts of one of its employees.

Anonymous Sudan has at times, in fact, appeared to have formal or informal ties to anti-Israel forces: According to prosecutors, it launched disruptive cyberattacks that targeted Israel's Tzeva Adom missile alert system on October 7, 2023, in the midst of attacks by the militant wing of Hamas that killed nearly 1,200 Israelis. As Israel's ensuing bombardment and invasion of the Gaza strip killed tens of thousands of Palestinian civilians over the months that followed, Anonymous Sudan frequently described the motivation for its attacks in its Telegram posts as the defense of Palestinians.

In December of 2023, for instance, Anonymous Sudan took OpenAI's ChatGPT offline with a sustained series of DDoS attacks in response to the company's executive Tal Broda vocally supporting the Israel Defense Forces’ missile attacks in Gaza. “More! No mercy! IDF don't stop!” Broda had written on X over a photo of a devastated urban landscape in Gaza, and in another post denied the existence of Palestine.

“We will continue targeting ChatGPT until the genocide supporter, Tal Broda, is fired and ChatGPT stops having dehumanizing views of Palestinians," Anonymous Sudan responded in a Telegram post explaining its attacks on OpenAI.

Still, Anonymous Sudan's true goals haven't always seemed entirely ideological, Akamai's Seaman says. The group has also offered to sell access to its DDoS infrastructure to other hackers: Telegram posts from the group as recently as March offered the use of its DDoS service, known as Godzilla or Skynet, for $2,500 a month. That suggests that even its attacks that appeared to be politically motivated may have been intended, at least in part, as marketing for its moneymaking side, Seaman argues.

“They seem to have thought, ‘We can get involved, really put a hurting on people, and market this service at the same time,’” Seaman says. He notes that, in the group's anti-Israel, pro-Palestine focus following the October 7 attacks, “there’s definitely an ideological thread in there. But the way it weaved through the different victims is something that maybe only the perpetrators of the attack fully understand.”

At times, Anonymous Sudan also hit Ukrainian targets, seemingly partnering with pro-Russian hacker groups like Killnet. That led some in the cybersecurity community to suspect that Anonymous Sudan was, in fact, a Russia-linked operation using its Sudanese identity as a front, given Russia's history of using hacktivism as false flag. The charges against Ahmed and Alaa Omer suggest that the group was, instead, authentically Sudanese in origin. But aside from its name, the group doesn't appear to have any clear ties to the original Anonymous hacker collective, which has been largely inactive for the last decade.

Aside from its targeting and politics, the group has distinguished itself through a relatively novel and effective technical approach, Akamai's Seaman says: Its DDoS service was built by gaining access to hundreds or possibly even thousands of virtual private servers—often-powerful machines offered by cloud services companies—by renting them with fraudulent credentials. It then used those machines to launch so-called layer 7 attacks, overwhelming web servers with requests for websites, rather than the lower-level floods of raw internet data requests that DDoS hackers have tended to use in the past. Anonymous Sudan and the customers of its DDoS services would then target victims with vast numbers of those layer 7 requests in parallel, sometimes using techniques called “multiplexing” or “pipelining” to simultaneously create multiple bandwidth demands on servers until they dropped offline.

For at least nine months, the group's technical power and brazen, unpredictable targeting made it a top concern for the anti-DDoS community, Seaman says—and for its many victims. “There was a lot of uncertainty about this group, what they were capable of, what their motivations were, why they targeted people,” says Seaman. “When Anonymous Sudan went away, there was a spike in curiosity and definitely a sigh of relief.”

The Justice Department's decision to level a criminal charge against Ahmed Omer that could lead to a life sentence for a denial-of-service attack may seem haphazard, given that state-sponsored cyberattacks and ransomware have often caused far more serious damage to health care networks, says Josh Corman, a researcher at the Institute for Security and Technology who has long focused on health care-targeted hacking. Corman says he's nonetheless encouraged to see prosecutors recognize that even crude cyberattacks can have serious—and even lethal—effects on victims.

“Yes, denial-of-service attacks can degrade and deny patent care to cause loss of life,” says Corman. “While this is the first, and it may seem arbitrary until we get more details, it could be heartening to see that we understand the outsize consequences of these attacks.”

_Updated 5 pm ET, October 16, 2024: While Anonymous Sudan is accused of launching thousands of attacks, the number of targets was in the hundreds. We've updated the story to clarify that distinction._

Hacker Charged With Seeking to Kill Using Cyberattacks on Hospitals

Global Intelligence claims its Cybercheck technology can help cops find key evidence to nail a case. But a WIRED investigation reveals the smoking gun often appears far less solid.

This AI Tool Helped Convict People of Murder. Then Someone Took a Closer Look

Bots that “remove clothes” from images have run rampant on the messaging app, allowing people to create nonconsensual deepfake images even as lawmakers and tech companies try to crack down.

In early 2020, deepfake expert Henry Ajder uncovered one of the first Telegram bots built to “undress” photos of women using artificial intelligence. At the time, Ajder recalls, the bot had been used to generate more than 100,000 explicit photos—including those of children—and its development marked a “watershed” moment for the horrors deepfakes could create. Since then, deepfakes have become more prevalent, more damaging, and easier to produce.

Now, a WIRED review of Telegram communities involved with the explicit nonconsensual content has identified at least 50 bots that claim to create explicit photos or videos of people with only a couple of clicks. The bots vary in capabilities, with many suggesting they can “remove clothes” from photos while others claim to create images depicting people in various sexual acts.

The 50 bots list more than 4 million “monthly users” combined, according to WIRED's review of the statistics presented by each bot. Two bots listed more than 400,000 monthly users each, while another 14 listed more than 100,000 members each. The findings illustrate how widespread explicit deepfake creation tools have become and reinforce Telegram’s place as one of the most prominent locations where they can be found. However, the snapshot, which largely encompasses English-language bots, is likely a small portion of the overall deepfake bots on Telegram.

“We’re talking about a significant, orders-of-magnitude increase in the number of people who are clearly actively using and creating this kind of content,” Ajder says of the Telegram bots. “It is really concerning that these tools—which are really ruining lives and creating a very nightmarish scenario primarily for young girls and for women—are still so easy to access and to find on the surface web, on one of the biggest apps in the world.”

Explicit nonconsensual deepfake content, which is often referred to as nonconsensual intimate image abuse (NCII), has exploded since it first emerged at the end of 2017, with generative AI advancements helping fuel recent growth. Across the internet, a slurry of “nudify” and “undress” websites sit alongside more sophisticated tools and Telegram bots, and are being used to target thousands of women and girls around the world—from Italy’s prime minister to school girls in South Korea. In one recent survey, a reported 40 percent of US students were aware of deepfakes linked to their K-12 schools in the last year.

The Telegram bots identified by WIRED are supported by at least 25 associated Telegram channels—where people can subscribe to newsfeed-style updates—that have more than 3 million combined members. The Telegram channels alert people about new features provided by the bots and special offers on “tokens” that can be purchased to operate them, and often act as places where people using the bots can find links to new ones if they are removed by Telegram.

After WIRED contacted Telegram with questions about whether it allows explicit deepfake content creation on its platform, the company deleted the 75 bots and channels WIRED identified. The company did not respond to a series of questions or comment on why it had removed the channels.

Additional nonconsensual deepfake Telegram channels and bots later identified by WIRED show the scale of the problem. Several channel owners posted that their bots had been taken down, with one saying, “We will make another bot tomorrow.” Those accounts were also later deleted.

**Hiding in Plain Sight**

Telegram bots are, essentially, small apps that run inside of Telegram. They sit alongside the app’s channels, which can broadcast messages to an unlimited number of subscribers; groups where up to 200,000 people can interact; and one-to-one messages. Developers have created bots where people take trivia quizzes, translate messages, create alerts, or start Zoom meetings. They’ve also been co-opted for creating abusive deepfakes.

Due to the harmful nature of the deepfake tools, WIRED did not test the Telegram bots and is not naming specific bots or channels. While the bots had millions of monthly users, according to Telegram’s statistics, it is unclear how many images the bots may have been used to create. Some users, who could be in multiple channels and bots, may have created zero images; others could have created hundreds.

Many of the deepfake bots viewed by WIRED are clear about what they have been created to do. The bots’ names and descriptions refer to nudity and removing women’s clothes. “I can do anything you want about the face or clothes of the photo you give me,” the creators’ of one bot wrote. “Experience the shock brought by AI,” another says. Telegram can also show “similar channels” in its recommendation tool, helping potential users bounce between channels and bots.

Almost all of the bots require people to buy “tokens” to create images, and it is unclear if they operate in the ways they claim. As the ecosystem around deepfake generation has flourished in recent years, it has become a potentially lucrative source of income for those who create websites, apps, and bots. So many people are trying to use “nudify” websites that Russian cybercriminals, as reported by 404Media, have started creating fake websites to infect people with malware.

While the first Telegram bots, identified several years ago, were relatively rudimentary, the technology needed to create more realistic AI-generated images has improved—and some of the bots are hiding in plain sight.

One bot with more than 300,000 monthly users did not reference any explicit material in its name or landing page. However, once a user clicks to use the bot, it claims it has more than 40 options for images, many of which are highly sexual in nature. That same bot has a user guide, hosted on the web outside of Telegram, describing how to create the highest-quality images. Bot developers can require users to accept terms of service, which may forbid users from uploading images without the consent of the person depicted or images of children, but there appears to be little or no enforcement of these rules.

Another bot, which had more than 38,000 users, claimed people could send six images of the same man or woman—it is one of a small number that claims to create images of men—to “train” an AI model, which could then create new deepfake images of that individual. Once users joined one bot, it would present a menu of 11 “other bots” from the creators, likely to keep systems online and try to avoid removals.

“These types of fake images can harm a person’s health and well-being by causing psychological trauma and feelings of humiliation, fear, embarrassment, and shame,” says Emma Pickering, the head of technology-facilitated abuse and economic empowerment at Refuge, the UK’s largest domestic abuse organization. “While this form of abuse is common, perpetrators are rarely held to account, and we know this type of abuse is becoming increasingly common in intimate partner relationships.”

As explicit deepfakes have become easier to create and more prevalent, lawmakers and tech companies have been slow to stem the tide. Across the US, 23 states have passed laws to address nonconsensual deepfakes, and tech companies have bolstered some policies. However, apps that can create explicit deepfakes have been found in Apple and Google’s app stores, explicit deepfakes of Taylor Swift were widely shared on X in January, and Big Tech sign-in infrastructure has allowed people to easily create accounts on deepfake websites.

Kate Ruane, director of the Center for Democracy and Technology’s free expression project, says most major technology platforms now have policies prohibiting nonconsensual distribution of intimate images, with many of the biggest agreeing to principles to tackle deepfakes. “I would say that it’s actually not clear whether nonconsensual intimate image creation or distribution is prohibited on the platform,” Ruane says of Telegram’s terms of service, which are less detailed than other major tech platforms.

Telegram’s approach to removing harmful content has long been criticized by civil society groups, with the platform historically hosting scammers, extreme right-wing groups, and terrorism-related content. Since Telegram CEO and founder Pavel Durov was arrested and charged in France in August relating to a range of potential offenses, Telegram has started to make some changes to its terms of service and provide data to law enforcement agencies. The company did not respond to WIRED’s questions about whether it specifically prohibits explicit deepfakes.

**Execute the Harm**

Ajder, the researcher who discovered deepfake Telegram bots four years ago, says the app is almost uniquely positioned for deepfake abuse. “Telegram provides you with the search functionality, so it allows you to identify communities, chats, and bots,” Ajder says. “It provides the bot-hosting functionality, so it's somewhere that provides the tooling in effect. Then it’s also the place where you can share it and actually execute the harm in terms of the end result.”

In late September, several deepfake channels started posting that Telegram had removed their bots. It is unclear what prompted the removals. On September 30, a channel with 295,000 subscribers posted that Telegram had “banned” its bots, but it posted a new bot link for users to use. (The channel was removed after WIRED sent questions to Telegram.)

“One of the things that’s really concerning about apps like Telegram is that it is so difficult to track and monitor, particularly from the perspective of survivors,” says Elena Michael, the cofounder and director of #NotYourPorn, a campaign group working to protect people from image-based sexual abuse.

Michael says Telegram has been “notoriously difficult” to discuss safety issues with, but notes there has been some progress from the company in recent years. However, she says the company should be more proactive in moderating and filtering out content itself.

“Imagine if you were a survivor who’s having to do that themselves, surely the burden shouldn't be on an individual,” Michael says. “Surely the burden should be on the company to put something in place that's proactive rather than reactive.”

Millions of People Are Using Abusive AI ‘Nudify’ Bots on Telegram

“Passkeys,” the secure authentication mechanism built to replace passwords, are getting more portable and easier for organizations to implement thanks to new initiatives the FIDO Alliance announced on Monday.

The password-killing tech known as “passkeys” have proliferated over the past two years, developed by the tech industry association known as the FIDO Alliance as an easier and more secure authentication alternative. And although superseding any technology as entrenched as passwords is difficult, new features and resources launching this week are pushing passkeys toward a tipping point.

At the FIDO Alliance's Authenticate Conference in Carlsbad, California, on Monday, researchers are announcing two projects that will make passkeys easier for organizations to offer—and easier for everyone to use. One is a new technical specification called Credential Exchange Protocol (CXP) that will make passkeys portable between digital ecosystems, a feature that users have increasingly demanded. The other is a website, called Passkey Central, where developers and system administrators can find resources like metrics and implementation guides that make it easier to add support for passkeys on existing digital platforms.

“To me, both announcements are part of the broader story of the industry working together to stop our dependence on passwords,” Andrew Shikiar, CEO of the FIDO Alliance, told WIRED ahead of Monday's announcements. “And when it comes to CXP, we have all these companies who are fierce competitors willing to collaborate on credential exchange."

CXP comprises a set of draft specifications developed by the FIDO Alliance's “Credential Provider Special Interest Group.” Development of technical standards can often be a fraught bureaucratic process, but the creation of CXP seems to have been positive and collaborative. Researchers from the password managers 1Password, Bitwarden, Dashlane, NordPass, and Enpass all worked on CXP, as did those from the identity providers Okta as well as Apple, Google, Microsoft, Samsung, and SK Telecom.

The specifications are significant for a few reasons. CXP was created for passkeys and is meant to address a longstanding criticism that passkeys could contribute to user lock-in by making it prohibitively difficult for people to move between operating system vendors and types of devices. In many ways, though, this problem already exists with passwords. Export features that allow you to move all of your passwords from one manager to another are often dangerously exposed and essentially just dump a list of all of your passwords into a plaintext file.

It's gotten much easier to sync passkeys across your devices through a single password manager, but CXP aims to standardize the technical process for securely transferring them between platforms so users are free—and safe—to roam the digital landscape. Importantly, while CXP was designed with passkeys in mind, it is really a specification that can be adapted to securely exchange other secrets as well, including passwords or other types of data.

“In the future, this could apply to mobile driver's licenses, say, or passports—any secrets that you want to export somewhere and import into another system,” Christiaan Brand, identity and security group product manager at Google, tells WIRED. “We've got most of the rough edges sanded down with passkeys, but one of the main pieces of negative feedback over the past year has been around portability and potential vendor lock-in. I think with this, we are signaling to the world that passkeys are growing up.”

The goal of the resource repository Passkey Central is similarly to help the ecosystem expand and mature. Product leads or security professionals who want to implement passkeys for their user base may need to make a business case to executives to get budget for the project. The FIDO Alliance is basically aiming to help them with the pitch—providing data and communications materials—and then support their rollout with prefab materials like implementation and roll-out guides, user experience and design guidelines, documentation around accessibility, and troubleshooting.

“We've made amazing progress on passkeys,” FIDO's Shikiar says. “Usability and user experience are pretty much there. But we do have a punch list and we’re actively working on it. Portability is an important feature on that list. And while the biggest brands on the planet are now using passkeys at scale, there’s a very long tail of companies that haven’t gotten started yet. So we want to offer resources and the assets they need to be successful."

Craig Newmark Philanthropies' Cyber Civil Defense coalition provides some funding to advance passkeys. In an interview with WIRED ahead of Monday's announcements, Newmark said he believes that passkeys can make a real difference both for the digital security of individual people and for internet security overall.

“There are a lot of vulnerable systems out there,” Newmark says. “You need to make it a lot harder for bad actors to defeat password schemes. You need to make everything more secure, and passkeys is part of that.”

The War on Passwords Is One Step Closer to Being Over

Plus: New details emerge in the National Public Data breach, Discord gets blocked in Russia and Turkey over alleged illegal activity on the platform, and more.

The Internet Archive is under attack. On top of multiple extinction-threatening lawsuits against the organization that created and maintains the Wayback Machine, hackers this week breached the Internet Archive, stole 31 million user account details, and defaced its website—all while archive.org struggled to stay online thanks to a barrage of distributed denial-of-service attacks. As of Friday, the site remained “temporarily offline.”

In a dark twist of fate, a judge this week cleared the way for the US Treasury Department to take possession of 69,000 bitcoins stolen from the Silk Road dark web market; meanwhile, the former IRS investigator who personally seized the bitcoins, Tigran Gambaryan, remains in a Nigerian jail cell on charges related to the actions of his current employer, embattled crypto exchange Binance. Members of Congress and other officials have called for the US government to do more to ensure Gambaryan’s release given his direct role in a series of major criminal cases and in pioneering crypto-investigation techniques. As for those seized Silk Road bitcoins, they are now worth $4.4 billion and will likely be auctioned off.

Security researchers this week detailed a pernicious malware that worms its way into Linux machines and uses a variety of techniques to evade detection. Dubbed Perfctl, the malware hides itself by creating files that match those typically found within Linux instances, using tricks to prevent admin tools from recording its activities, and more. All of this is done with the goal of remaining on an infected machine to keep carrying out a variety of malicious activities. Researchers estimate that millions of Linux devices could be vulnerable.

Finally, we dissected the ways in which Google’s decision to _not_ kill third-party tracking cookies in its Chrome browser could continue to impact your privacy.

And that’s not all. Each week, we round up the security and privacy news we didn’t cover in depth ourselves. Click the headlines to read the full stories. And stay safe out there.

Police use of honeypots to catch cybercriminals red-handed is nothing new. But creating an entirely new cryptocurrency to catch pump-and-dump schemers? Now that’s something special. The US Department of Justice revealed this week that the FBI made a new Ethereum-based crypto token, NexFundAI, specifically to trick people who manipulate crypto markets and take them down.

While the investigation ultimately resulted in charges against 18 people and other entities for alleged fraud and crypto market manipulation, the blast radius of the scheme also impacted some regular retail investors who are not accused of any crimes, although US officials did not provide details about those investments. A US prosecutor involved in the case told reporters, however, that the investigation netted a total of $25 million in funds, which will be returned to investors. Trading on NexFundAI has since been disabled.

**National Public Data Files for Bankruptcy After Catastrophic Breach**

National Public Data, a data broker based in Florida, is having a bad year. In August, hackers published 2.9 billion records stolen from NPD last December that included names, mailing addresses, phone numbers, email addresses, and Social Security numbers—a giant trove the hackers claim impacted “the entire population of USA, CA, and UK.” Then came the inevitable lawsuits against NPD, which is now filing for bankruptcy. Those proceedings have revealed new details, including the fact that NPD is run by a single person, Salvatore Verini, Jr, who operated the business out of his home on around $2,500 worth of equipment. A document filed in a bankruptcy court by one of NPD’s debtors states that the breach may have impacted “hundreds of millions” of people.

**Russia and Turkey Block Discord**

Discord users in Russia and Turkey this week found they were suddenly unable to connect to the online chat application. Authorities in both countries later revealed that Discord had been blocked for allegedly facilitating illegal activity. Russia’s internet regulator, Roskomnadzor, said in a statement the block “is necessary to prevent the use of the messenger for terrorist and extremist purposes, the recruitment of citizens for their commission, the sale of drugs, in connection with the placement of illegal information.” Turkish authorities, meanwhile, banned the messaging app after a court decision involving child abuse material that was allegedly hosted on Discord servers. According to BleepingComputer, some Discord users in those countries were able to access the app using a VPN that routed their connections through foreign IP addresses—potentially good news for Russian troops who were reportedly disrupted by the block.

**Police Secretly Use Face Recognition Tech to Link People to Crimes**

Law enforcement use of face recognition technology to pin crimes on Americans is far more widespread than previously known, according to a newly published investigation by The Washington Post. Records obtained by the Post found that police in 15 states used face recognition tools in “more than 1,000 investigations over the past four years.” Despite its apparent widespread use, police departments frequently seek to hide their use of the technology, which has been found to inaccurately identify people who are then charged with crimes they did not commit. As an assistant public defender in Minnesota told Post reporters, police likely obscure their use of face recognition because they “want to avoid the litigation surrounding reliability of the technology.”

The FBI Made a Crypto Coin Just to Catch Fraudsters

Scammers in Southeast Asia are increasingly turning to AI, deepfakes, and dangerous malware in a way that makes their pig butchering operations even more convincing.

As digital scamming explodes in Southeast Asia, including so called “pig butchering” investment scams, the United Nations Office on Drugs and Crime (UNODC) issued a comprehensive report this week with a dire warning about the rapid growth of this criminal ecosystem. Many digital scams have traditionally relied on social engineering, or tricking victims into giving away their money willingly, rather than leaning on malware or other highly technical methods. But researchers have increasingly sounded the alarm that scammers are incorporating generative AI content and deepfakes to expand the scale and effectiveness of their operations. And the UN report offers the clearest evidence yet that these high tech tools are turning an already urgent situation into a crisis.

In addition to buying written scripts to use with potential victims or relying on templates for malicious websites, attackers have increasingly been leaning on generative AI platforms to create communication content in multiple languages and deepfake generators that can create photos or even video of nonexistent people to show victims and enhance verisimilitude. Scammers have also been expanding their use of tools that can drain a victim’s cryptocurrency wallets, have been manipulating transaction records to trick targets into sending cryptocurrency to the wrong places, and are compromising smart contracts to steal cryptocurrency. And in some cases, they’ve been purchasing Elon Musk’s Starlink satellite internet systems to help power their efforts.

“Agile criminal networks are integrating these new technologies faster than anticipated, driven by new online marketplaces and service providers which have supercharged the illicit service economy,” John Wojcik, a UNODC regional analyst, tells WIRED. “These developments have not only expanded the scope and efficiency of cyber-enabled fraud and cybercrime, but they have also lowered the barriers to entry for criminal networks that previously lacked the technical skills to exploit more sophisticated and profitable methods.”

For years, China-linked criminals have trafficked people into gigantic compounds in Southeast Asia, where they are often forced to run scams, held against their will, and beaten if they refuse instructions. Around 200,000 people, from at least 60 countries, have been trafficked to compounds largely in Myanmar, Cambodia, and Laos over the last five years. However, as WIRED reporting has shown, these operations are spreading globally—with scamming infrastructure emerging in the Middle East, Eastern Europe, Latin America, and West Africa.

Most prominently, these organized crime operations have run pig butchering scams, where they build intimate relationships with victims before introducing an “investment opportunity” and asking for money. Criminal organizations may have conned people out of around $75 billion through pig butchering scams. Aside from pig butchering, according to the UN report, criminals across Southeast Asia are also running job scams, law enforcement impersonation, asset recovery scams, virtual kidnappings, sextortion, loan scams, business email compromise, and other illicit schemes. Criminal networks in the region earned up to $37 billion last year, UN officials estimate. Perhaps unsurprisingly, all of this revenue is allowing scammers to expand their operations and diversify, incorporating new infrastructure and technology into their systems in the hope of making them more efficient and brutally effective.

For example, scammers are often constrained by their language skills and ability to keep up conversations with potentially hundreds of victims at a time in numerous languages and dialects. However, generative AI developments within the last two years—including the launch of writing tools such as ChatGPT—are making it easier for criminals to break down language barriers and create the content needed for scamming.

The UN’s report says AI can be used for automating phishing attacks that ensnare victims, the creation of fake identities and online profiles, and the crafting of personalized scripts to trick victims while messaging them in different languages. “These developments have not only expanded the scope and efficiency of cyber-enabled fraud and cybercrime, but they have also lowered the barriers to entry for criminal networks that previously lacked the technical skills to exploit sophisticated and profitable methods,” the report says.

Stephanie Baroud, a criminal intelligence analyst in Interpol’s human trafficking unit, says the impact of AI needs to be considered as part of a pig butchering scammer’s tactics going forward. Baroud, who spoke with WIRED in an interview before the publication of the UN report, says the criminal’s recruitment ads that lure people into being trafficked to scamming compounds used to be “very generic” and full of grammatical errors. However, AI is now making them appear more polished and compelling, Baroud says. “It is really making it easier to create a very realistic job offer,” she says. “Unfortunately, this will make it much more difficult to identify which is the real and which is the fake ads.”

Perhaps the biggest AI paradigm shift in such digital attacks comes from deepfakes. Scammers are increasingly using machine-learning systems to allow for real-time face-swapping. This technology, which has also been used by romance scammers in West Africa, allows criminals to change their appearance on calls with their victims, making them realistically appear to be a different person. The technology is allowing “one-click” face swaps and high-resolution video feeds, the UN’s report states. Such services are a game changer for scammers, because they allow attackers to “prove” to victims in photos or real-time video calls that they are who they claim to be.

Using these setups, however, can require stable internet connections, which can be harder to maintain within some regions where pig butchering compounds and other scamming have flourished. There has been a “notable” increase in cops seizing Starlink satellite dishes in recent months in Southeast Asia, the UN says—80 units were seized between April and June this year. In one such operation carried out in June, Thai police confiscated 58 Starlink devices. In another instance, law enforcement seized 10 Starlink devices and 4,998 preregistered SIM cards while criminals were in the process of moving their operations from Myanmar to Laos. Starlink did not immediately respond to WIRED’s request for comment.

“Obviously using real people has been working for them very well, but using the tech could be cheaper after they have the required computers” and connectivity, says Troy Gochenour, a volunteer with the Global Anti-Scam Organization (GASO), a US-based nonprofit that fights human-trafficking and cybercrime operations in Southeast Asia.

Gochenour’s research involves tracking trends on Chinese-language Telegram channels related to carrying out pig butchering scams. And he says that it is increasingly common to see people applying to be AI models for scam content.

In addition to AI services, attackers have increasingly leaned on other technical solutions as well. One tool that has been increasingly common in digital scamming is so-called “crypto drainers,” a type of malware that has particularly been deployed against victims in Southeast Asia. Drainers can be more or less technically sophisticated, but their common goal is to “drain” funds from a target’s cryptocurrency wallets and redirect the currency to wallets controlled by attackers. Rather than stealing the credentials to access the target wallet directly, drainers are typically designed to look like a legitimate service—either by impersonating an actual platform or creating a plausible brand. Once a victim has been tricked into connecting their wallet to the drainer, they are then manipulated into approving one or a few transactions that grant attackers unintended access to all the funds in the wallet.

Drainers can be used in many contexts and with many fronts. They can be a component of pig butchering investment scams, or promoted to potential victims through compromised social media accounts, phishing campaigns, and malvertizing. Researchers from the firm ScamSniffer, for example, published findings in December about sponsored social media and search engine ads linked to malicious websites that contained a cryptocurrency drainer. The campaign, which ran from March to December 2023 reportedly stole about $59 million from more than 63,000 victims around the world.

Far from the low-tech days of doing everything through social engineering by building a rapport with potential victims and crafting tricky emails and text messages, today’s scammers are taking a hybrid approach to make their operations as efficient and lucrative as possible, UN researchers say. And even if they aren’t developing sophisticated malware themselves in most cases, scammers are increasingly in the market to use these malicious tools, prompting malware authors to adapt or create hacking tools for scams like pig butchering.

Researchers say that scammers have been seen using infostealers and even remote access trojans that essentially create a backdoor in a victim’s system that can be utilized in other types of attacks. And scammers are also expanding their use of malicious smart contracts that appear to programmatically establish a certain agreed-upon transaction or set of transactions, but actually does much more. “Infostealer logs and underground data markets have also been critical to ongoing market expansion, with access to unprecedented amounts of sensitive data serving as a major catalyst,” Wojcik, from the UNODC, says.

The changing tactics are significant as global law enforcement scrambles to deter digital scamming. But they are just one piece of the larger picture, which is increasingly urgent and bleak for forced laborers and victims of these crimes.

“It is now increasingly clear that a potentially irreversible displacement and spillover has taken place in which organized crime are able to pick, choose, and move value and jurisdictions as needed, with the resulting situation rapidly outpacing the capacity of governments to contain it,” UN officials wrote in the report. “Failure to address this ecosystem will have consequences for Southeast Asia and other regions.”

Pig Butchering Scams Are Going High Tech

It's hard enough creating one air-gap-jumping tool. Researchers say the group GoldenJackal did it twice in five years.

Researchers have unearthed two sophisticated tool sets that a nation-state hacking group—possibly from Russia—used to steal sensitive data stored on air-gapped devices, meaning those that are deliberately isolated from the internet or other networks to safeguard them from malware.

One of the custom tool collections was used starting in 2019 against a South Asian embassy in Belarus. A largely different tool set created by the same threat group infected a European Union government organization three years later. Researchers from ESET, the security firm that discovered the toolkits, said some of the components in both were identical to those fellow security firm Kaspersky described in research published last year and attributed to an unknown group, tracked as GoldenJackal, working for a nation-state. Based on the overlap, ESET has concluded that the same group is behind all the attacks observed by both firms.

**Quite Unusual**

The practice of air gapping is typically reserved for the most sensitive networks or devices connected to them, such as those used in systems for voting, industrial control, manufacturing, and power generation. A host of malware used in espionage hacking over the past 15 years (for instance, here and here) demonstrate that air gapping isn’t a foolproof protection. It nonetheless forces threat groups to expend significant resources that are likely obtainable only by nation-states with superior technical acumen and unlimited budgets. ESET’s discovery puts GoldenJackal in a highly exclusive collection of threat groups.

“With the level of sophistication required, it is quite unusual that in five years, GoldenJackal managed to build and deploy not one but two separate tool sets designed to compromise air-gapped systems,” ESET researcher Matías Porolli wrote in Tuesday’s report. “This speaks to the resourcefulness of the group.”

The evolution of the kit from 2019 and the one from three years later underscores a growing sophistication by GoldenJackal developers. The first generation provided a full suite of capabilities, including:

*   GoldenDealer, a component that delivers malicious executables to air-gapped systems over USB drives
*   GoldenHowl, a backdoor that contains various modules for a mix of malicious capabilities
*   GoldenRobo, a file collector and exfiltrator

Within a few weeks of deploying the kit in 2019, ESET said, GoldenJackal started using other tools on the same compromised devices. The newer tools, which Kaspersky documented in its 2023 research, included:

*   A backdoor tracked under the name JackalControl
*   JackalSteal, a file collector and exfiltrator
*   JackalWorm, used to propagate other JackalControl and other malicious components over USB drives

GoldenJackal, ESET said, continued using these tools into January of this year. The basic flow of the attack is, first, infecting an internet-connected device through a means ESET and Kaspersky have been unable to determine. Next, the infected computer infects any external drives that get inserted. When the infected drive is plugged into an air-gapped system, it collects and stores data of interest. Last, when the drive is inserted into the internet-connected device, the data is transferred to an attacker-controlled server.

**Building a Better Trap**

In the 2022 attack on the European Union governmental organization, GoldenJackal began using a new custom toolkit. Written in multiple programming languages, including Go and Python, the newer version took a much more specialized approach. It assigned different tasks to different types of infected devices and marshaled a much larger array of modules, which could be mixed and matched based on the attacker objects for different infections.

“In the observed attacks, GoldenJackal started to use a highly modular approach, using various components to perform different tasks,” ESET’s Porolli wrote. “Some hosts were abused to exfiltrate files, others were used as local servers to receive and distribute staged files or configuration files, and others were deemed interesting for file collection, for espionage purposes.”

The figure below classifies some of the components specifically:

*   GoldenUsbCopy, which monitors for the insertion of USB drives on air-gapped devices and, when found, copies them to an encrypted container that is stored on disk
*   GoldenUsbGo, which appears to be an updated version of GoldenUSBCopy
*   GoldenAce, a distribution tool for propagating other malicious executables and retrieving files stored on USB drives
*   HTTP server, an HTTP server whose precise function isn’t well understood
*   ​​GoldenBlacklist, which downloads an encrypted archive from a local server, sifts through received email messages for those of interest, and puts them in an archive for some other component to exfiltrate
*   GoldenPyBlacklist, a Python implementation of GoldenBlacklist
*   GoldenMailer, which, when connected to an internet device, exfiltrates files of interest previously stolen from an air-gapped device. The exfiltration occurs by attaching them to emails sent to an attacker-controlled email address
*   GoldenDrive, a separate exfiltration tool that, in contrast to GoldenMailer, uploads files of interest to Google Drive.

The newly discovered toolkit is composed of many different building blocks, written in multiple languages and capabilities. The overall goal appears to be increased flexibility and resiliency in the event one module is detected by the target.

“Their goal is to get hard to obtain data from air-gapped systems and stay under the radar as much as possible,” Costin Raiu, a researcher who worked at Kaspersky at the time it was researching GoldenJackal, wrote in an interview. “Multiple exfiltration mechanisms indicate a very flexible tool kit that can accommodate all sorts of situations. These many tools indicate it’s a highly customizable framework where they deploy exactly what they need as opposed to a multi purpose malware that can do anything.”

Other new insights offered by the ESET research is GoldenJackal’s interest in targets located in Europe. Kaspersky researchers detected the group targeting Middle Eastern countries.

Based on the information that was available to Kaspersky, company researchers couldn’t attribute GoldenJackal to any specific country. ESET has also been unable to definitively identify the country, but it did find one hint that the threat group may have a tie to Turla, a potent hacking group working on behalf of Russia’s FSB intelligence agency. The tie comes in the form of command-and-control protocol in GoldenHowl referred to as transport\_http. The same expression is found in malware known to originate with Turla.

Raiu said the highly modular approach is also reminiscent of Red October, an elaborate espionage platform discovered in 2013 targeting hundreds of diplomatic, governmental, and scientific organizations in at least 39 countries, including the Russian Federation, Iran, and the United States.

While much of Tuesday’s report contains technical analysis that is likely to be too advanced for many people to understand, it provides important new information that furthers insights into malware designed to jump air gaps and the tactics, techniques, and procedures of those who use it. The report will also be useful to people responsible for safeguarding the types of organizations most frequently targeted by nation-state groups.

“I’d say this is mostly interesting for security people working in embassies and government CERTs,” Raiu said. “They need to check for these TTPs and keep an eye on them in the future. If you were previously a victim of Turla or Red October I’d keep an eye on this.”

_This story originally appeared on_ _Ars Technica._

A Mysterious Hacking Group Has 2 New Tools to Steal Data From Air-Gapped Machines

The hack exposed the data of 31 million users as the embattled Wayback Machine maker scrambles to stay online and contain the fallout of digital—and legal—attacks.

An illicit JavaScript pop-up on the Internet Archive proclaimed on Wednesday afternoon that the site had suffered a major data breach. Hours later, the organization confirmed the incident.

Longtime security researcher Troy Hunt, who runs the data-breach-notification website Have I Been Pwned (HIBP) also confirmed that the breach is legitimate. He said it occurred in September and that the stolen trove contains 31 million unique email addresses along with usernames, bcrypt password hashes, and other system data. Bleeping Computer, which first reported the breach, also confirmed the validity of the data.

The Internet Archive did not return multiple requests for comment from WIRED.

“Have you ever felt like the Internet Archive runs on sticks and is constantly on the verge of suffering a catastrophic security breach?” the attackers wrote in Wednesday's Internet Archive pop-up message. “It just happened. See 31 million of you on HIBP!”

In addition to the breach and site defacement, the Internet Archive has been grappling with a wave of distributed denial-of-service attacks that have intermittently brought down its services.

Internet Archive founder Brewster Kahle provided a public update on Wednesday evening in a post on the social network X. “What we know: DDOS attack—fended off for now; defacement of our website via JS library; breach of usernames/email/salted-encrypted passwords. What we’ve done: Disabled the JS library, scrubbing systems, upgrading security. Will share more as we know it.” “Scrubbing systems” refer to services that offer DDoS attack protection by filtering malicious junk traffic so it can't deluge and disrupt a website.

The Internet Archive has faced aggressive DDoS attacks numerous times in the past, including in late May. As Kahle wrote on Wednesday: “Yesterday's DDoS attack on @internetarchive repeated today. We are working to bring http://archive.org back online.” The hacktivist group known as BlackMeta claimed responsibility for this week's DDoS attacks and said it plans to carry out more against the Internet Archive. Still, the perpetrator of the data breach is not yet known.

The Internet Archive has faced battles on many fronts in recent months. In addition to repeated DDoS attacks, the organization is also facing mounting legal challenges. It recently lost an appeal in _Hachette v. Internet Archive_, a lawsuit brought by book publishers, which argued that its digital lending library violated copyright law. Now it’s facing an existential threat in the form of another copyright lawsuit, this one from music labels, which may result in damages upwards of $621 million if the court rules against the archive.

HIBP's Hunt says that he first received the stolen Internet Archive data on September 30, reviewed it on October 5, and warned the organization about it on October 6. He says the group confirmed the breach to him the next day and that he planned to load the data into HIBP and notify its subscribers about the breach on Wednesday. “They get defaced and DDoS'd, right as the data is loading into HIBP,” Hunt wrote. “The timing on the last point seems to be entirely coincidental.”

Hunt added, too, that while he encouraged the group to publicly disclose the data breach itself before the HIBP notifications went out, the extenuating circumstances may explain the delay.

“Obviously I would have liked to see that disclosure much earlier, but understanding how under attack they are, I think everyone should cut them some slack,” Hunt wrote. “They're a nonprofit doing great work and providing a service that so many of us rely heavily on.”

Internet Archive Breach Exposes 31 Million Users

The $4.4 billion in crypto is set to be the largest pile of criminal proceeds ever sold off by the US. The former IRS agent who seized the recording-breaking sum, meanwhile, languishes in a Nigerian jail cell.

In November of 2020, a person identified by the US Department of Justice only as “Individual X” sat down with an IRS agent named Tigran Gambaryan and prosecutors at the US Attorney’s office in San Francisco. That unnamed individual typed out the long string of characters of a Bitcoin private key on Gambaryan’s laptop, allowing Gambaryan to move 69,370 bitcoins from Individual X’s bitcoin address to one controlled by the US government.

Over the four years that it has taken the US government to establish legal ownership of that massive sum of bitcoins—identified by the IRS as stolen proceeds of the Silk Road dark-web drug market—its value has grown to a staggering $4.4 billion dollars. The money's forfeiture appears to have been part of a deal that kept Individual X out of prison, though the terms of that deal have never been publicly disclosed.

Instead, in a bizarre twist of fate, it's Gambaryan, the IRS criminal investigator who traced and seized that record-breaking sum of cryptocurrency, who is sitting in a Nigerian jail as those billions finally end up in US coffers.

On Wednesday, the US Supreme Court declined to hear an appeal to a lower court ruling that the US government can seize Individual X's nearly 70,000 bitcoins, which the person allegedly took from the Silk Road more than a decade ago by exploiting a security vulnerability in the market. That stolen drug money has since been claimed by various parties, most recently by a company called Battle Born Investments that sought to appeal the seizure ruling and claimed to have purchased the Silk Road bitcoins through a bankruptcy proceeding. Only now that this appeal has fizzled, four years after the stolen money was tracked down in an investigation led by Gambaryan at IRS Criminal Investigations, can the US government finally take official possession of the wayward bitcoins, which will likely be auctioned off for dollars by the US Marshals Service, as smaller sums of seized cryptocurrency have in the past.

“The Supreme Court's decision not to hear the case means that this is now is going to be forfeited to the United States government,” says Will Frentzen, one of the US prosecutors who handled the Individual X case and is now an attorney at the legal firm Morrison Foerster. “This is the largest ever forfeiture of crypto that will go to the US Treasury.” In fact, thanks to Bitcoin's wild appreciation in recent years, it appears to be the largest ever criminal seizure of money of _any kind_ to be added to the US federal budget. (A seizure following the theft of 120,000 bitcoins from the cryptocurrency exchange Bitfinex was even larger but will likely be repaid to victims and creditors rather than kept by the government.)

Over those same years, however, Gambaryan has taken an even more unlikely path: In 2021, he left the IRS to take a job as the head of investigations at Binance, the world's largest cryptocurrency exchange—a move seen widely as driven by Binance’s belated attempt to clean up its own widespread use for money laundering, which led the company to pay a $4.3 billion criminal fine to the US government last year. When Nigeria followed that fine by accusing Binance of similar criminal misconduct and devaluing the country's national currency, it was Gambaryan who was invited to Abuja to negotiate with the Nigerian government earlier this year. Instead, the Nigerian government detained Gambaryan, took his passport, and has now jailed him for over six months, charging him with money laundering and tax evasion as a proxy for his employer.

“It's beyond bizarre. It's Twilight Zone every day,” says Frentzen, who now has taken on Gambaryan's case in Nigeria as one of his defense attorneys. “His impact as a federal agent was so extensive and so far reaching that he's been out of the government for three years and yet the American government and the American public are still benefiting from his hard work. And he's languishing in a Nigerian jail cell on trumped-up charges. It's really unfathomable.”

Attention to Gambaryan's case has grown in recent months as lawmakers and US officials have increasingly appealed to the Nigerian government to release him on bail or on medical grounds. Gambaryan has suffered from a worsening herniated disc in his back that requires immediate surgery to prevent permanent injury, according to Frentzen. Yet he's been denied medical care in jail, and a video of him entering a courtroom in Abuja with a crutch seems to show a security guard refusing to help Gambaryan walk or to provide him with a wheelchair despite his pleas.

In June, 16 members of Congress called in an open letter for Gambaryan's case to be treated as a hostage situation, and a resolution introduced in the House Foreign Affairs Committee in July urged the White House to apply pressure to Nigeria to free Gambaryan. “I urge the State Department and I urge President Biden: Turn up the heat on Nigeria,” Arkansas representative French Hill, who visited Gambaryan in jail, said in a hearing last month. “Honor the fact that an American citizen has been snatched by a friendly government and put in prison over something he had no responsibility for.”

Over his 10 years at IRS Criminal Investigations, Gambaryan pioneered federal law enforcement's use of cryptocurrency tracing as an investigative technique, with landmark results: From 2014 to 2017, Gambaryan identified two corrupt federal agents who had enriched themselves with cryptocurrency during their investigation of the Silk Road; helped to track down half a billion dollars worth of bitcoins stolen from early crypto exchange Mt. Gox; had a hand in developing a secret crypto tracing method that located the server hosting the massive AlphaBay dark-web crime market; and helped to take down the Welcome to Video crypto-funded child sexual abuse video network, a case that led to 337 arrests around the world and the rescue of 23 children from exploitative situations.

In the wake of that string of cases, Gambaryan and another IRS Criminal Investigations agent, Jeremy Haynie, began in 2017 to trace the 69,370 bitcoins that appeared to have been stolen from the Silk Road in 2012 and 2013 and had sat almost entirely unmoved in the years since. With the help of blockchain analysis and data from seized servers of the criminal BTC-e crypto exchange—the fruits of another of Gambaryan's investigations—the agents were able to identify Individual X as a hacker who had taken the Silk Road's funds and demand that they forfeit them.

As those funds flow into the federal government's accounts, Frentzen hopes that federal officials who have the power to pressure Nigeria to free Gambaryan will remember where that record sum of money came from. “It should be a crystal-clear reminder of the impact that he's had for the American government and for the American people. To have him locked up while the US Treasury is collecting $4.4 billion because of his work, it's really twisted,” Frentzen says. “We urge the US government to continue to work to secure his release and to do so as fast as possible. And we encourage the Nigerians to do the right thing and release him.”

Source

Wired