Category: vishing

Generative AI has the potential to make social engineering attacks much more sophisticated and personalised. The technology can rapidly mine sites for information on a company, individuals, their responsibilities and specific habits to create multi-level campaigns. Through automated gethering of information, the technology can acquire photos, videos and audio recordings which can then be used to craft emails (phishing), voice attacks (vishing) and deep fake videos and images for spear phishing attacks against individuals in positions of power, for instance. 

We’re already seeing evidence of such attacks in action. Back in February the Hong Kong police revealed that a finance worker at Arup, a UK engineering firm, was duped into transferring $25m when he attended a video call in which every attendee, including the CFO, was a deep fake. Similar attacks have been carried out over the WhatsApp platform, with LastPass targeted in April by calls, texts and voicemails which impersonated the company’s CEO and a senior exec at advertising firm, WPP invited to a video call in which they were asked to set up a new business by a clone of the CEO crafted from YouTube videos and voice cloning technology.

Deep fakes go wide

These are no longer isolated incidents either, with the CIO of Arup, Rob Greig, warning in his statement that the number and sophistication of deepfake scams has been rising sharply in recent months. It’s a view substantiated by The State of Information Security 2024 report from ISMS.Online which reveals that 32% of UK businesses experienced deep fake cyber security incidents, with Business Email Compromise (BEC) the most common attack type over the last 12 months. Indeed, reports suggest there was a 780% rise in deep fake attacks across Europe between 2022-23.

GenAI is a gamechanger for crafting deep fakes, because the AI enhances its own production, delivering hyper-realistic content. Physical mannerisms, movements, intonations of voice and other subtleties are processed via an AI encoding algorithm or Generative Antagonistic Network (GAN) to clone individuals. These GANS have significantly lowered the barrier to entry so that creating deepfakes today requires a much lower level of skills and resources, according to the Department for Homeland Security.

Defending against such attacks can prove challenging because users are much more susceptible to phishing which emulates another person. There are giveaways, however, with deep fake technology typically struggling to accurately capture the inside of the mouth, resulting in blurring. There may also be less movement such as blinking, or more screen flashes than you’d expect. Generally speaking, its currently easiest to fake audio, followed by photos while video is the most challenging.

Why we can’t fight fire with fire

While standalone and open source technological solutions are now available that scan and assess the possible manipulation of video, audio and text giving a reliability score as a percentage, success rates are mixed. It’s difficult to verify accuracy because few are transparent about how they arrived at the score, the dataset used or when they were last updated. They vary in approach from those trained on GANs to classifiers that can detect if a piece of content was produced with a specific tool, although even content deemed as authentically created in a piece of software can be manipulated. Many video apps, messaging and collaborative platforms already use AI with respect to filters, making detection even more problematic. 

Given the current technological vacuum, the main form of mitigation today is employee security awareness, with 47% saying they are placing greater emphasis on training in the ISMS.Online survey. However, the survey notes that even well-trained employees can struggle to identify deep fakes and this is being compounded by a lack of policy enforcement; the survey found 34% were not using adequate security on their BYOD devices and 30% were not securing sensitive information. Zero trust initiatives may well help here in limiting access to such sensitive information but few organisations have mature deployments. 

Deloitte makes a number of recommendations on how to mitigate the threat of deep fake attacks in its report The battle against digital manipulation. In addition to training and access controls, it advocates the implementation of a layer of verification in business processes and the clarification of verification protocols when it comes to sanctioning payments. This could be in the form of multiple layers to approve transactions, for example, from code words to token-based systems or live detection verification such as taking a “selfie” or video recording, which is already in use in the banking sector for user verification.

Policy and process

But overarching all of this we need to see a comprehensive security policy covering people, process and technology from an AI-perspective. This should seek to address AI attack detection and response, for example, so that there are channels in place for reporting a suspected Gen-AI attack or if a payment has been made. There are already a number of AI standards that can be used to help here in the governance of AI such as ISO/IEC 42001:2023 as well as the NIST AI Risk Management Framework. 

Defending against deep fakes will therefore require a three-pronged approach that sees awareness training combined with security controls including access and user verification, as well as frameworks to govern how GenAI is used within the business and remediation and response. Ironically, it’s a problem that is likely to be addressed best by people and process rather than technology.  

Looking to the future, some are suggesting that deep fakes could see senior execs decide to adopt a lower profile online in a bid to limit the capture of their likeness. Yet conversely there are some, such as the CEO of Zoom, who believe we will instead go to the opposite extreme and embrace the technology to create digital clones of ourselves that will then attend meetings on our behalf. These will learn from individual recordings to reproduce our mannerisms, be briefed by us, and report back with a call summary and actions. If that approach is widely adopted then detection technologies will prove to be something of a non-starter, making the primary methods of defence the verification processes and an effective AI policy.

 

 

The post Cyborg Social Engineering: Defending against personalised attacks appeared first on Cybersecurity Insiders.

Web hosting giant GoDaddy made headlines this month when it disclosed that a multi-year breach allowed intruders to steal company source code, siphon customer and employee login credentials, and foist malware on customer websites. Media coverage understandably focused on GoDaddy’s admission that it suffered three different cyberattacks over as many years at the hands of the same hacking group.  But it’s worth revisiting how this group typically got in to targeted companies: By calling employees and tricking them into navigating to a phishing website.

In a filing with the U.S. Securities and Exchange Commission (SEC), GoDaddy said it determined that the same “sophisticated threat actor group” was responsible for three separate intrusions, including:

-March 2020: A spear-phishing attack on a GoDaddy employee compromised the hosting login credentials of approximately 28,000 GoDaddy customers, as well as login credentials for a small number employees;

-November 2021: A compromised GoDaddy password let attackers steal source code and information tied to 1.2 million customers, including website administrator passwords, sFTP credentials, and private SSL keys;

-December 2022: Hackers gained access to and installed malware on GoDaddy’s cPanel hosting servers that “intermittently redirected random customer websites to malicious sites.”

“Based on our investigation, we believe these incidents are part of a multi-year campaign by a sophisticated threat actor group that, among other things, installed malware on our systems and obtained pieces of code related to some services within GoDaddy,” the company stated in its SEC filing.

What else do we know about the cause of these incidents? We don’t know much about the source of the November 2021 incident, other than GoDaddy’s statement that it involved a compromised password, and that it took about two months for the company to detect the intrusion. GoDaddy has not disclosed the source of the breach in December 2022 that led to malware on some customer websites.

But we do know the March 2020 attack was precipitated by a spear-phishing attack against a GoDaddy employee. GoDaddy described the incident at the time in general terms as a social engineering attack, but one of its customers affected by that March 2020 breach actually spoke to one of the hackers involved.

The hackers were able to change the Domain Name System (DNS) records for the transaction brokering site escrow.com so that it pointed to an address in Malaysia that was host to just a few other domains, including the then brand-new phishing domain servicenow-godaddy[.]com.

The general manager of Escrow.com found himself on the phone with one of the GoDaddy hackers, after someone who claimed they worked at GoDaddy called and said they needed him to authorize some changes to the account.

In reality, the caller had just tricked a GoDaddy employee into giving away their credentials, and he could see from the employee’s account that Escrow.com required a specific security procedure to complete a domain transfer.

The general manager of Escrow.com said he suspected the call was a scam, but decided to play along for about an hour — all the while recording the call and coaxing information out of the scammer.

“This guy had access to the notes, and knew the number to call,” to make changes to the account, the CEO of Escrow.com told KrebsOnSecurity. “He was literally reading off the tickets to the notes of the admin panel inside GoDaddy.”

About halfway through this conversation — after being called out by the general manager as an imposter — the hacker admitted that he was not a GoDaddy employee, and that he was in fact part of a group that enjoyed repeated success with social engineering employees at targeted companies over the phone.

Absent from GoDaddy’s SEC statement is another spate of attacks in November 2020, in which unknown intruders redirected email and web traffic for multiple cryptocurrency services that used GoDaddy in some capacity.

It is possible this incident was not mentioned because it was the work of yet another group of intruders. But in response to questions from KrebsOnSecurity at the time, GoDaddy said that incident also stemmed from a “limited” number of GoDaddy employees falling for a sophisticated social engineering scam.

“As threat actors become increasingly sophisticated and aggressive in their attacks, we are constantly educating employees about new tactics that might be used against them and adopting new security measures to prevent future attacks,” GoDaddy said in a written statement back in 2020.

Voice phishing or “vishing” attacks typically target employees who work remotely. The phishers will usually claim that they’re calling from the employer’s IT department, supposedly to help troubleshoot some issue. The goal is to convince the target to enter their credentials at a website set up by the attackers that mimics the organization’s corporate email or VPN portal.

Experts interviewed for an August 2020 story on a steep rise in successful voice phishing attacks said there are generally at least two people involved in each vishing scam: One who is social engineering the target over the phone, and another co-conspirator who takes any credentials entered at the phishing page — including multi-factor authentication codes shared by the victim — and quickly uses them to log in to the company’s website.

The attackers are usually careful to do nothing with the phishing domain until they are ready to initiate a vishing call to a potential victim. And when the attack or call is complete, they disable the website tied to the domain.

This is key because many domain registrars will only respond to external requests to take down a phishing website if the site is live at the time of the abuse complaint. This tactic also can stymie efforts by companies that focus on identifying newly-registered phishing domains before they can be used for fraud.

A U2F device made by Yubikey.

GoDaddy’s latest SEC filing indicates the company had nearly 7,000 employees as of December 2022. In addition, GoDaddy contracts with another 3,000 people who work full-time for the company via business process outsourcing companies based primarily in India, the Philippines and Colombia.

Many companies now require employees to supply a one-time password — such as one sent via SMS or produced by a mobile authenticator app — in addition to their username and password when logging in to company assets online. But both SMS and app-based codes can be undermined by phishing attacks that simply request this information in addition to the user’s password.

One multifactor option — physical security keys — appears to be immune to these advanced scams. The most commonly used security keys are inexpensive USB-based devices. A security key implements a form of multi-factor authentication known as Universal 2nd Factor (U2F), which allows the user to complete the login process simply by inserting the USB device and pressing a button on the device. The key works without the need for any special software drivers.

The allure of U2F devices for multi-factor authentication is that even if an employee who has enrolled a security key for authentication tries to log in at an impostor site, the company’s systems simply refuse to request the security key if the user isn’t on their employer’s legitimate website, and the login attempt fails. Thus, the second factor cannot be phished, either over the phone or Internet.

In July 2018, Google disclosed that it had not had any of its 85,000+ employees successfully phished on their work-related accounts since early 2017, when it began requiring all employees to use physical security keys in place of one-time codes.

Many organizations are already struggling to combat cybersecurity threats from ransomware purveyors and state-sponsored hacking groups, both of which tend to take days or weeks to pivot from an opportunistic malware infection to a full blown data breach. But few organizations have a playbook for responding to the kinds of virtual “smash and grab” attacks we’ve seen recently from LAPSUS$, a juvenile data extortion group whose short-lived, low-tech and remarkably effective tactics have put some of the world’s biggest corporations on edge.

Since surfacing in late 2021, LAPSUS$ has gained access to the networks or contractors for some of the world’s largest technology companies, including Microsoft, NVIDIA, Okta and Samsung. LAPSUS$ typically threatens to release sensitive data unless paid a ransom, but with most victims the hackers ended up publishing any information they stole (mainly computer source code).

Microsoft blogged about its attack at the hands of LAPSUS$, and about the group targeting its customers. It found LAPSUS$ used a variety of old-fashioned techniques that seldom show up in any corporate breach post-mortems, such as:

-targeting employees at their personal email addresses and phone numbers;
-offering to pay $20,000 a week to employees who give up remote access credentials;
-social engineering help desk and customer support employees at targeted companies;
-bribing/tricking employees at mobile phone stores to hijack a target’s phone number;
-intruding on their victims’ crisis communications calls post-breach.

If these tactics sound like something you might sooner expect from spooky, state-sponsored “Advanced Persistent Threat” or APT groups, consider that the core LAPSUS$ members are thought to range in age from 15 to 21. Also, LAPSUS$ operates on a shoestring budget and is anything but stealthy: According to Microsoft, LAPSUS$ doesn’t seem to cover its tracks or hide its activity. In fact, the group often announces its hacks on social media.

ADVANCED PERSISTENT TEENAGERS

This unusual combination makes LAPSUS$ something of an aberration that is probably more aptly referred to as “Advanced Persistent Teenagers,” said one CXO at a large organization that recently had a run-in with LAPSUS$.

“There is a lot of speculation about how good they are, tactics et cetera, but I think it’s more than that,” said the CXO, who spoke about the incident on condition of anonymity. “They put together an approach that industry thought suboptimal and unlikely. So it’s their golden hour.”

LAPSUS$ seems to have conjured some worst-case scenarios in the minds of many security experts, who worry what will happen when more organized cybercriminal groups start adopting these techniques.

“LAPSUS$ has shown that with only $25,000, a group of teenagers could get into organizations with mature cybersecurity practices,” said Amit Yoran, CEO of security firm Tenable and a former federal cybersecurity czar, testifying last week before the House Homeland Security Committee. “With much deeper pockets, focus, and mission, targeting critical infrastructure. That should be a sobering, if not terrifying, call to action.”

My CXO source said LAPSUS$ succeeds because they simply refuse to give up, and just keep trying until someone lets them in.

“They would just keep jamming a few individuals to get [remote] access, read some onboarding documents, enroll a new 2FA [two-factor authentication method] and exfiltrate code or secrets, like a smash-and-grab,” the CXO said. “These guys were not leet, just damn persistent.”

HOW DID WE GET HERE?

The smash-and-grab attacks by LAPSUS$ obscure some of the group’s less public activities, which according to Microsoft include targeting individual user accounts at cryptocurrency exchanges to drain crypto holdings.

In some ways, the attacks from LAPSUS$ recall the July 2020 intrusion at Twitter, wherein the accounts for Apple, Bill Gates, Jeff Bezos, Kanye West, Uber and others were made to tweet messages inviting the world to participate in a cryptocurrency scam that promised to double any amount sent to specific wallets. The flash scam netted the perpetrators more than $100,000 in the ensuing hours.

The group of teenagers who hacked Twitter hailed from a community that traded in hacked social media accounts. This community places a special premium on accounts with short “OG” usernames, and some of its most successful and notorious members were known to use all of the methods Microsoft attributed to LAPSUS$ in the service of hijacking prized OG accounts.

The Twitter hackers largely pulled it off by brute force, writes Wired on the July 15, 2020 hack.

“Someone was trying to phish employee credentials, and they were good at it,” Wired reported. “They were calling up consumer service and tech support personnel, instructing them to reset their passwords. Many employees passed the messages onto the security team and went back to business. But a few gullible ones—maybe four, maybe six, maybe eight—were more accommodating. They went to a dummy site controlled by the hackers and entered their credentials in a way that served up their usernames and passwords as well as multifactor authentication codes.”

Twitter revealed that a key tactic of the group was “phone spear phishing” (a.k.a. “voice phishing” a.k.a. “vishing”). This involved calling up Twitter staffers using false identities, and tricking them into giving up credentials for an internal company tool that let the hackers reset passwords and multi-factor authentication setups for targeted users.

In August 2020, KrebsOnSecurity warned that crooks were using voice phishing to target new hires at major companies, impersonating IT employees and asking them to update their VPN client or log in at a phishing website that mimicked their employer’s VPN login page.

Two days after that story ran, the FBI and the Cybersecurity & Infrastructure Security Agency (CISA) issued their own warning on vishing, saying the attackers typically compiled dossiers on employees at specific companies by mass-scraping public profiles on social media platforms, recruiter and marketing tools, publicly available background check services, and open-source research. The joint FBI/CISA alert continued:

“Actors first began using unattributed Voice over Internet Protocol (VoIP) numbers to call targeted employees on their personal cellphones, and later began incorporating spoofed numbers of other offices and employees in the victim company. The actors used social engineering techniques and, in some cases, posed as members of the victim company’s IT help desk, using their knowledge of the employee’s personally identifiable information—including name, position, duration at company, and home address—to gain the trust of the targeted employee.”

“The actors then convinced the targeted employee that a new VPN link would be sent and required their login, including any 2FA [2-factor authentication] or OTP [one-time passwords]. The actor logged the information provided by the employee and used it in real-time to gain access to corporate tools using the employee’s account.”

Like LAPSUS$, these vishers just kept up their social engineering attacks until they succeeded. As KrebsOnSecurity wrote about the vishers back in 2020:

“It matters little to the attackers if the first few social engineering attempts fail. Most targeted employees are working from home or can be reached on a mobile device. If at first the attackers don’t succeed, they simply try again with a different employee.”

“And with each passing attempt, the phishers can glean important details from employees about the target’s operations, such as company-specific lingo used to describe its various online assets, or its corporate hierarchy.”

“Thus, each unsuccessful attempt actually teaches the fraudsters how to refine their social engineering approach with the next mark within the targeted organization.”

SMASH & GRAB

The primary danger with smash-and-grab groups like LAPSUS$ is not just their persistence but their ability to extract the maximum amount of sensitive information from their victims using compromised user accounts that typically have a short lifespan. After all, in many attacks, the stolen credentials are useful only so long as the impersonated employee isn’t also trying to use them.

This dynamic puts tremendous pressure on cyber incident response teams, which suddenly are faced with insiders who are trying frantically to steal everything of perceived value within a short window of time. On top of that, LAPSUS$ has a habit of posting screenshots on social media touting its access to internal corporate tools. These images and claims quickly go viral and create a public relations nightmare for the victim organization.

Single sign-on provider Okta experienced this firsthand last month, when LAPSUS$ posted screenshots that appeared to show Okta’s Slack channels and another with a Cloudflare interface. Cloudflare responded by resetting its employees’ Okta credentials.

Okta quickly came under fire for posting only a brief statement that said the screenshots LAPSUS$ shared were connected to a January 2022 incident involving the compromise of “a third-party customer support engineer working for one of our subprocessors,” and that “the matter was investigated and contained by the subprocessor.”

This assurance apparently did not sit well with many Okta customers, especially after LAPSUS$ began posting statements that disputed some of Okta’s claims. On March 25, Okta issued an apology for its handling of the January breach at a third-party support provider, which ultimately affected hundreds of its customers.

My CXO source said the lesson from LAPSUS$ is that even short-lived intrusions can have a long-term negative impact on victim organizations — especially when victims are not immediately forthcoming about the details of a security incident that affects customers.

“It does force us to think about insider access differently,” the CXO told KrebsOnSecurity. “Nation states have typically wanted longer, more strategic access; ransomware groups want large lateral movement. LAPSUS$ doesn’t care, it’s more about, ‘What can these 2-3 accounts get me in the next 6 hours?’ We haven’t optimized to defend that.”

Any organizations wondering what they can do to harden their systems against attacks from groups like LAPSUS$ should consult Microsoft’s recent blog post on the group’s activities, tactics and tools. Microsoft’s guidance includes recommendations that can help prevent account takeovers or at least mitigate the impact from stolen employee credentials.