Login
The Federal Bureau of Investigation (FBI) is urging police departments and governments worldwide to beef up security around their email systems, citing a recent increase in cybercriminal services that use hacked police email accounts to send unauthorized subpoenas and customer data requests to U.S.-based technology companies.
In an alert (PDF) published this week, the FBI said it has seen an uptick in postings on criminal forums regarding the process of emergency data requests (EDRs) and the sale of email credentials stolen from police departments and government agencies.
“Cybercriminals are likely gaining access to compromised US and foreign government email addresses and using them to conduct fraudulent emergency data requests to US based companies, exposing the personal information of customers to further use for criminal purposes,” the FBI warned.
In the United States, when federal, state or local law enforcement agencies wish to obtain information about an account at a technology provider — such as the account’s email address, or what Internet addresses a specific cell phone account has used in the past — they must submit an official court-ordered warrant or subpoena.
Virtually all major technology companies serving large numbers of users online have departments that routinely review and process such requests, which are typically granted (eventually, and at least in part) as long as the proper documents are provided and the request appears to come from an email address connected to an actual police department domain name.
In some cases, a cybercriminal will offer to forge a court-approved subpoena and send that through a hacked police or government email account. But increasingly, thieves are relying on fake EDRs, which allow investigators to attest that people will be bodily harmed or killed unless a request for account data is granted expeditiously.
The trouble is, these EDRs largely bypass any official review and do not require the requester to supply any court-approved documents. Also, it is difficult for a company that receives one of these EDRs to immediately determine whether it is legitimate.
In this scenario, the receiving company finds itself caught between two unsavory outcomes: Failing to immediately comply with an EDR — and potentially having someone’s blood on their hands — or possibly leaking a customer record to the wrong person.
Perhaps unsurprisingly, compliance with such requests tends to be extremely high. For example, in its most recent transparency report (PDF) Verizon said it received more than 127,000 law enforcement demands for customer data in the second half of 2023 — including more than 36,000 EDRs — and that the company provided records in response to approximately 90 percent of requests.
One English-speaking cybercriminal who goes by the nicknames “Pwnstar” and “Pwnipotent” has been selling fake EDR services on both Russian-language and English cybercrime forums. Their prices range from $1,000 to $3,000 per successful request, and they claim to control “gov emails from over 25 countries,” including Argentina, Bangladesh, Brazil, Bolivia, Dominican Republic, Hungary, India, Kenya, Jordan, Lebanon, Laos, Malaysia, Mexico, Morocco, Nigeria, Oman, Pakistan, Panama, Paraguay, Peru, Philippines, Tunisia, Turkey, United Arab Emirates (UAE), and Vietnam.
“I cannot 100% guarantee every order will go through,” Pwnstar explained. “This is social engineering at the highest level and there will be failed attempts at times. Don’t be discouraged. You can use escrow and I give full refund back if EDR doesn’t go through and you don’t receive your information.”
An ad from Pwnstar for fake EDR services.
A review of EDR vendors across many cybercrime forums shows that some fake EDR vendors sell the ability to send phony police requests to specific social media platforms, including forged court-approved documents. Others simply sell access to hacked government or police email accounts, and leave it up to the buyer to forge any needed documents.
“When you get account, it’s yours, your account, your liability,” reads an ad in October on BreachForums. “Unlimited Emergency Data Requests. Once Paid, the Logins are completely Yours. Reset as you please. You would need to Forge Documents to Successfully Emergency Data Request.”
Still other fake EDR service vendors claim to sell hacked or fraudulently created accounts on Kodex, a startup that aims to help tech companies do a better job screening out phony law enforcement data requests. Kodex is trying to tackle the problem of fake EDRs by working directly with the data providers to pool information about police or government officials submitting these requests, with an eye toward making it easier for everyone to spot an unauthorized EDR.
If police or government officials wish to request records regarding Coinbase customers, for example, they must first register an account on Kodexglobal.com. Kodex’s systems then assign that requestor a score or credit rating, wherein officials who have a long history of sending valid legal requests will have a higher rating than someone sending an EDR for the first time.
It is not uncommon to see fake EDR vendors claim the ability to send data requests through Kodex, with some even sharing redacted screenshots of police accounts at Kodex.
Matt Donahue is the former FBI agent who founded Kodex in 2021. Donahue said just because someone can use a legitimate police department or government email to create a Kodex account doesn’t mean that user will be able to send anything. Donahue said even if one customer gets a fake request, Kodex is able to prevent the same thing from happening to another.
Kodex told KrebsOnSecurity that over the past 12 months it has processed a total of 1,597 EDRs, and that 485 of those requests (~30 percent) failed a second-level verification. Kodex reports it has suspended nearly 4,000 law enforcement users in the past year, including:
-1,521 from the Asia-Pacific region;
-1,290 requests from Europe, the Middle East and Asia;
-460 from police departments and agencies in the United States;
-385 from entities in Latin America, and;
-285 from Brazil.
Donahue said 60 technology companies are now routing all law enforcement data requests through Kodex, including an increasing number of financial institutions and cryptocurrency platforms. He said one concern shared by recent prospective customers is that crooks are seeking to use phony law enforcement requests to freeze and in some cases seize funds in specific accounts.
“What’s being conflated [with EDRs] is anything that doesn’t involve a formal judge’s signature or legal process,” Donahue said. “That can include control over data, like an account freeze or preservation request.”
In a hypothetical example, a scammer uses a hacked government email account to request that a service provider place a hold on a specific bank or crypto account that is allegedly subject to a garnishment order, or party to crime that is globally sanctioned, such as terrorist financing or child exploitation.
A few days or weeks later, the same impersonator returns with a request to seize funds in the account, or to divert the funds to a custodial wallet supposedly controlled by government investigators.
“In terms of overall social engineering attacks, the more you have a relationship with someone the more they’re going to trust you,” Donahue said. “If you send them a freeze order, that’s a way to establish trust, because [the first time] they’re not asking for information. They’re just saying, ‘Hey can you do me a favor?’ And that makes the [recipient] feel valued.”
Echoing the FBI’s warning, Donahue said far too many police departments in the United States and other countries have poor account security hygiene, and often do not enforce basic account security precautions — such as requiring phishing-resistant multifactor authentication.
How are cybercriminals typically gaining access to police and government email accounts? Donahue said it’s still mostly email-based phishing, and credentials that are stolen by opportunistic malware infections and sold on the dark web. But as bad as things are internationally, he said, many law enforcement entities in the United States still have much room for improvement in account security.
“Unfortunately, a lot of this is phishing or malware campaigns,” Donahue said. “A lot of global police agencies don’t have stringent cybersecurity hygiene, but even U.S. dot-gov emails get hacked. Over the last nine months, I’ve reached out to CISA (the Cybersecurity and Infrastructure Security Agency) over a dozen times about .gov email addresses that were compromised and that CISA was unaware of.”
Organizations that get relieved of credentials to their cloud environments can quickly find themselves part of a disturbing new trend: Cybercriminals using stolen cloud credentials to operate and resell sexualized AI-powered chat services. Researchers say these illicit chat bots, which use custom jailbreaks to bypass content filtering, often veer into darker role-playing scenarios, including child sexual exploitation and rape.
Image: Shutterstock.
Researchers at security firm Permiso Security say attacks against generative artificial intelligence (AI) infrastructure like Bedrock from Amazon Web Services (AWS) have increased markedly over the last six months, particularly when someone in the organization accidentally exposes their cloud credentials or key online, such as in a code repository like GitHub.
Investigating the abuse of AWS accounts for several organizations, Permiso found attackers had seized on stolen AWS credentials to interact with the large language models (LLMs) available on Bedrock. But they also soon discovered none of these AWS users had enabled full logging of LLM activity (by default, logs don’t include model prompts and outputs), and thus they lacked any visibility into what attackers were doing with that access.
So Permiso researchers decided to leak their own test AWS key on GitHub, while turning on logging so that they could see exactly what an attacker might ask for, and what the responses might be.
Within minutes, their bait key was scooped up and used in a service that offers AI-powered sex chats online.
“After reviewing the prompts and responses it became clear that the attacker was hosting an AI roleplaying service that leverages common jailbreak techniques to get the models to accept and respond with content that would normally be blocked,” Permiso researchers wrote in a report released today.
“Almost all of the roleplaying was of a sexual nature, with some of the content straying into darker topics such as child sexual abuse,” they continued. “Over the course of two days we saw over 75,000 successful model invocations, almost all of a sexual nature.”
Ian Ahl, senior vice president of threat research at Permiso, said attackers in possession of a working cloud account traditionally have used that access for run-of-the-mill financial cybercrime, such as cryptocurrency mining or spam. But over the past six months, Ahl said, Bedrock has emerged as one of the top targeted cloud services.
“Bad guy hosts a chat service, and subscribers pay them money,” Ahl said of the business model for commandeering Bedrock access to power sex chat bots. “They don’t want to pay for all the prompting that their subscribers are doing, so instead they hijack someone else’s infrastructure.”
Ahl said much of the AI-powered chat conversations initiated by the users of their honeypot AWS key were harmless roleplaying of sexual behavior.
“But a percentage of it is also geared toward very illegal stuff, like child sexual assault fantasies and rapes being played out,” Ahl said. “And these are typically things the large language models won’t be able to talk about.”
AWS’s Bedrock uses large language models from Anthropic, which incorporates a number of technical restrictions aimed at placing certain ethical guardrails on the use of their LLMs. But attackers can evade or “jailbreak” their way out of these restricted settings, usually by asking the AI to imagine itself in an elaborate hypothetical situation under which its normal restrictions might be relaxed or discarded altogether.
“A typical jailbreak will pose a very specific scenario, like you’re a writer who’s doing research for a book, and everyone involved is a consenting adult, even though they often end up chatting about nonconsensual things,” Ahl said.
In June 2024, security experts at Sysdig documented a new attack that leveraged stolen cloud credentials to target ten cloud-hosted LLMs. The attackers Sysdig wrote about gathered cloud credentials through a known security vulnerability, but the researchers also found the attackers sold LLM access to other cybercriminals while sticking the cloud account owner with an astronomical bill.
“Once initial access was obtained, they exfiltrated cloud credentials and gained access to the cloud environment, where they attempted to access local LLM models hosted by cloud providers: in this instance, a local Claude (v2/v3) LLM model from Anthropic was targeted,” Sysdig researchers wrote. “If undiscovered, this type of attack could result in over $46,000 of LLM consumption costs per day for the victim.”
Ahl said it’s not certain who is responsible for operating and selling these sex chat services, but Permiso suspects the activity may be tied to a platform cheekily named “chub[.]ai,” which offers a broad selection of pre-made AI characters with whom users can strike up a conversation. Permiso said almost every character name from the prompts they captured in their honeypot could be found at Chub.
Some of the AI chat bot characters offered by Chub. Some of these characters include the tags “rape” and “incest.”
Chub offers free registration, via its website or a mobile app. But after a few minutes of chatting with their newfound AI friends, users are asked to purchase a subscription. The site’s homepage features a banner at the top that reads: “Banned from OpenAI? Get unmetered access to uncensored alternatives for as little as $5 a month.”
Until late last week Chub offered a wide selection of characters in a category called “NSFL” or Not Safe for Life, a term meant to describe content that is disturbing or nauseating to the point of being emotionally scarring.
Fortune profiled Chub AI in a January 2024 story that described the service as a virtual brothel advertised by illustrated girls in spaghetti strap dresses who promise a chat-based “world without feminism,” where “girls offer sexual services.” From that piece:
Chub AI offers more than 500 such scenarios, and a growing number of other sites are enabling similar AI-powered child pornographic role-play. They are part of a broader uncensored AI economy that, according to Fortune’s interviews with 18 AI developers and founders, was spurred first by OpenAI and then accelerated by Meta’s release of its open-source Llama tool.
Fortune says Chub is run by someone using the handle “Lore,” who said they launched the service to help others evade content restrictions on AI platforms. Chub charges fees starting at $5 a month to use the new chatbots, and the founder told Fortune the site had generated more than $1 million in annualized revenue.
KrebsOnSecurity sought comment about Permiso’s research from AWS, which initially seemed to downplay the seriousness of the researchers’ findings. The company noted that AWS employs automated systems that will alert customers if their credentials or keys are found exposed online.
AWS explained that when a key or credential pair is flagged as exposed, it is then restricted to limit the amount of abuse that attackers can potentially commit with that access. For example, flagged credentials can’t be used to create or modify authorized accounts, or spin up new cloud resources.
Ahl said Permiso did indeed receive multiple alerts from AWS about their exposed key, including one that warned their account may have been used by an unauthorized party. But they said the restrictions AWS placed on the exposed key did nothing to stop the attackers from using it to abuse Bedrock services.
Sometime in the past few days, however, AWS responded by including Bedrock in the list of services that will be quarantined in the event an AWS key or credential pair is found compromised or exposed online. AWS confirmed that Bedrock was a new addition to its quarantine procedures.
Additionally, not long after KrebsOnSecurity began reporting this story, Chub’s website removed its NSFL section. It also appears to have removed cached copies of the site from the Wayback Machine at archive.org. Still, Permiso found that Chub’s user stats page shows the site has more than 3,000 AI conversation bots with the NSFL tag, and that 2,113 accounts were following the NSFL tag.
The user stats page at Chub shows more than 2,113 people have subscribed to its AI conversation bots with the “Not Safe for Life” designation.
Permiso said their entire two-day experiment generated a $3,500 bill from AWS. Most of that cost was tied to the 75,000 LLM invocations caused by the sex chat service that hijacked their key.
Paradoxically, Permiso found that while enabling these logs is the only way to know for sure how crooks might be using a stolen key, the cybercriminals who are reselling stolen or exposed AWS credentials for sex chats have started including programmatic checks in their code to ensure they aren’t using AWS keys that have prompt logging enabled.
“Enabling logging is actually a deterrent to these attackers because they are immediately checking to see if you have logging on,” Ahl said. “At least some of these guys will totally ignore those accounts, because they don’t want anyone to see what they’re doing.”
In a statement shared with KrebsOnSecurity, AWS said its services are operating securely, as designed, and that no customer action is needed. Here is their statement:
“AWS services are operating securely, as designed, and no customer action is needed. The researchers devised a testing scenario that deliberately disregarded security best practices to test what may happen in a very specific scenario. No customers were put at risk. To carry out this research, security researchers ignored fundamental security best practices and publicly shared an access key on the internet to observe what would happen.”
“AWS, nonetheless, quickly and automatically identified the exposure and notified the researchers, who opted not to take action. We then identified suspected compromised activity and took additional action to further restrict the account, which stopped this abuse. We recommend customers follow security best practices, such as protecting their access keys and avoiding the use of long-term keys to the extent possible. We thank Permiso Security for engaging AWS Security.”
AWS said customers can configure model invocation logging to collect Bedrock invocation logs, model input data, and model output data for all invocations in the AWS account used in Amazon Bedrock. Customers can also use CloudTrail to monitor Amazon Bedrock API calls.
The company said AWS customers also can use services such as GuardDuty to detect potential security concerns and Billing Alarms to provide notifications of abnormal billing activity. Finally, AWS Cost Explorer is intended to give customers a way to visualize and manage Bedrock costs and usage over time.
Anthropic told KrebsOnSecurity it is always working on novel techniques to make its models more resistant to jailbreaks.
“We remain committed to implementing strict policies and advanced techniques to protect users, as well as publishing our own research so that other AI developers can learn from it,” Anthropic said in an emailed statement. “We appreciate the research community’s efforts in highlighting potential vulnerabilities.”
Anthropic said it uses feedback from child safety experts at Thorn around signals often seen in child grooming to update its classifiers, enhance its usage policies, fine tune its models, and incorporate those signals into testing of future models.
Update: 5:01 p.m. ET: Chub has issued a statement saying they are only hosting the role-playing characters, and that the LLMs they use run on their own infrastructure.
“Our own LLMs run on our own infrastructure,” Chub wrote in an emailed statement. “Any individuals participating in such attacks can use any number of UIs that allow user-supplied keys to connect to third-party APIs. We do not participate in, enable or condone any illegal activity whatsoever.”
Over the past three decades, we’ve had time at Trend Micro to observe the industry trends that have the biggest impact on our customers. And one of the big things we’ve seen is that threats move largely in tandem with changes to IT infrastructure. This matters today because most organizations are transforming the way they run and manage their infrastructure—a daunting task on its own.
But with digital transformation also comes an expanded corporate attack surface, driving security leaders to demand enhanced visibility, detection & response across the entire enterprise — this is not just about the endpoint.
Transforming business
Over the past five years, there has been a major shift in the way IT infrastructure is delivered, and with that shift, increasing complexity. A big part of this change has been the use of the cloud, reflected in Gartner’s prediction that the market will grow to over $266 billion in 2020. Organizations everywhere are leveraging the cloud and DevOps to rapidly deliver new and differentiated applications and services for their customers, partners and employees. And the use of containers and microservices across a multi-cloud and hybrid environment is increasingly common.
In addition to leveraging public cloud services like IaaS, organizations are also rapidly adopting SaaS applications like Office 365, and expanding their use of mobile and collaborative applications to support remote working. Some are even arguing that working patterns may never be the same again, following the changes forced on many employers by the Covid-19 pandemic.
Combine these changes with networks that continue to extend to include branch offices and add new areas to protect like operational technology including industrial systems, and we can certainly see that the challenges facing the modern enterprise look nothing like they did a few years ago.
Under fire, under pressure
All of these infrastructure changes make for a broader attack surface that the bad guys can take advantage of, and they’re doing so with an increasingly wide range of tools and techniques. In the cloud there is a new class of vulnerabilities introduced through a greater use of open source, containers, orchestration platforms, supply chain applications and more. For all organizations, the majority of threats still prey upon the user, arriving via email (over 90% of the 52.3 billion we blocked in 2019), and they’re no longer just basic phishing attempts. There’s been an uptick in fileless events designed to bypass traditional security filters (we blocked 1.4 million last year). And Business Email Compromise (BEC) and ransomware continue to evolve, the latter causing major outages across local government, healthcare and other vulnerable sectors.
Organizations are often left flat-footed because they don’t have the in-house skills to secure a rapidly evolving IT environment. Mistakes get made, and configuration errors can allow the hackers to sneak in.
Against this backdrop, CISOs need visibility, detection and response capabilities across the extended enterprise. But in too many cases, teams are struggling because they have:
|
|
Beyond the endpoint
While endpoint detection and response (EDR) has become a popular response to some of these problems over recent years, the reality is that cyber-attacks are rarely straightforward and limited to the endpoint (as noted in the email statistic above). Security teams actually need visibility, detection, and response across the entire IT environment, so they can better contextualize and deal with threats.
This is what Trend Micro XDR offers. It provides visibility across not just endpoints but also email, servers, cloud workloads and networks, applying AI and expert security analytics to correlate and identify potential threats. The result is fewer, higher fidelity alerts for stretched IT security teams to deal with. Recognizing the skills shortage reality, we also offer a managed XDR service that augments in-house SOC activities with the power of Trend Micro security experts.
Detection and response is too important to be limited to the endpoint. Today’s CISOs need visibility, detection, and response everywhere.
The post Why CISOs Are Demanding Detection and Response Everywhere appeared first on .
FreshRSS 