The US Cybersecurity and Infrastructure Security Agency (CISA) has found that compromise of valid accounts and spear phishing attacks were the two most common vectors of initial access in 2022, Decipher reports. Valid accounts were compromised in 54% of successful attacks. “Valid accounts can be former employee accounts that have not been removed from the active directory or default administrator accounts,” CISA said.

We've reported on several Amazon scams, but for once, there is positive news. Amazon sent an email Thursday morning highlighting the top scams your users should watch out for: Prime Membership Scams Per Amazon, "These are unexpected calls/texts/emails that refer to a costly membership fee or an issue with your membership and ask you to confirm or cancel the charge.

Researchers from Carnegie Mellon University and the Center for A.I. Safety have discovered a new prompt injection method to override the guardrails of large language models (LLMs). These guardrails are safety measures designed to prevent AI from generating harmful content. This discovery poses a significant risk to the deployment of LLMs in public-facing applications, as it could potentially allow these models to be used for malicious purposes.

What happened? The SEC (Securities and Exchange Commission) has introduced new rules that require public companies to be more transparent about their cybersecurity risks and any breaches they experience. This means companies will need to regularly share information about how they're managing cybersecurity risks and any significant cybersecurity incidents they've had. If a company experiences a significant cybersecurity incident, they'll need to report it within four business days.

Social engineering is involved in 70% to 90% of successful compromises. It is the number one way that hackers and malware successfully attack devices and networks. No other initial root cause comes close (unpatched software and firmware are a distant second, being involved in about 33% of attacks). A particular type of social engineering is responsible for more successful compromises than any other type of attack: spear phishing.

New data focused on the first half of the year shows some anomalies. Phishing attacks are slowing down… that is, until you dive into the details. I can’t remember the last time I posted a headline stating that phishing numbers were down; that’s because we haven’t seen this trend occur in a number of years. But new data from Vade Secure’s H1 2023 Phishing and Malware Report shows an interesting outlier that skews a high-level view of the data.

Amid potential concerns by governments, customers, and prospects about ties with the Russian government, the cybersecurity vendor Group-IB continues in its promise to separate itself from Russia. You can understand how an organization may look at a Russia-based company these days; it’s not the fault of the Russian company, but of the negative posture many feel towards the Russian government.

Fraudsters are spreading scams on Facebook that pose as ads for legitimate AI tools, according to researchers at Check Point. The Facebook pages impersonate ChatGPT, Google Bard, Midjourney, Jasper, and more.

Scammers are taking advantage of the popularity of the Barbie movie, according to researchers at McAfee. “In the last 3 weeks, we’ve seen 100 new instances of malware that have Barbie-related filenames,” the researchers write. “Once again, this shows how attackers have latched onto the movie’s hype, hoping the people will click the malicious files because the Barbie name is trending. The types of files varied but included typical types such as.html and.exe.

IBM Security has released its annual Cost of a Data Breach Report, revealing that the global average cost of a data breach reached $4.45 million in 2023. This marks a significant increase of 15% over the past 3 years, making it the highest recorded cost in the history of the report. Notably, detection and escalation costs have seen a substantial rise of 42% during the same period, indicating a shift towards more complex breach investigations.

For over 25 years, a technology utilized for vital data and voice radio communications globally has remained under wraps, preventing in-depth testing for potential vulnerabilities. However, a small group of researchers in the Netherlands has now shed light on it, uncovering significant flaws, including a deliberate backdoor.

A new social engineering campaign tracked as “FakeSG” is distributing the NetSupport remote access Trojan (RAT) via phony browser updates, according to researchers at Malwarebytes. The campaign is similar but distinct from the widespread “SocGholish” campaign, which also uses fake browser updates to deliver NetSupport.

The US Federal Bureau of Investigation (FBI) has warned of an increase in tech support scams that attempt to trick users into sending cash via snail mail.

Microsoft was the most impersonated brand in phishing attacks during Q2, 2023, according to Check Point’s latest Brand Phishing Report.

A new report focused on the healthcare sector sheds light on the state of cyber attacks in the European Union, including the types of attacks, who’s targeted, motivations, and who’s responsible. The newly-released Health Threat Landscape report from the European Union Agency for Cybersecurity (ENISA) is the first analysis completed by the agency and covers incidents from 2021 through March of 2023.

The latest data from the FBI’s Internet Crime Complaint Center (IC3) ups the estimate for the cost of losses and exposure through business email compromise (BEC) attacks from 2013 through 2023. In the latest advisory from the IC3 entitled “Business Email Compromise: The $50 Billion Scam,” there was a 17% increase in losses from BEC attacks in 2022.

Cybercriminals are exploiting the introduction of “.ZIP” as a new generic Top-Level Domain (gTLD) to launch phishing attacks, according to researchers at Fortinet. “Cybercriminals are always on the lookout for new opportunities and techniques to exploit, and the recent availability of '.ZIP' domains for public purchase has unfortunately created such an opportunity,” the researchers write.

CyberWire wrote: "Researchers at SlashNext describe a generative AI cybercrime tool called “WormGPT,” which is being advertised on underground forums as “a blackhat alternative to GPT models, designed specifically for malicious activities.” The tool can generate output that legitimate AI models try to prevent, such as malware code or phishing templates.

A new scam aimed at stealing your credit card and banking information has reared its’ ugly head as a completely legitimate ad that is likely to be clicked based on the corresponding search term. If you type in “USPS Tracking” in Google, you probably want to enter a U.S. Postal Service tracking number so you can see where your package is, right? So, if you saw the following result, would you give it a second thought? Source: Malwarebytes.

The quantity of emails involved in scams and cyber attacks continues to grow as credential theft and response-based phishing persist as top attack variants. The ripple effect from cybercrime-as-a-service launching a few years back has reached critical mass, where we’re seeing significant increases in the percentage of emails that are either clearly determined to be malicious (7.7%) as well as those suspicious enough that users are recommended to not engage with (15.9%).

New insight from blockchain analysis company, Chainalysis, shows that activity involving known ransomware crypto addresses has grown over the last 18 months, despite a downfall of other malicious activity. When I cover reports, there’s an understanding that the accuracy of the data provided is dependent on the number of organizations responding to a survey, the geos and industries represented, etc.

Researchers at Check Point outline various forms of tailgating attacks. These attacks can allow threat actors to bypass physical security measures via social engineering. “Tailgating is a common form of social engineering attack,” the researchers write. “Social engineering attacks use trickery, deception, or coercion to induce someone to take actions that are not in the best interests of themselves or the organization.

PoisonGPT works completely normally, until you ask it who the first person to walk on the moon was. A team of researchers has developed a proof-of-concept AI model called "PoisonGPT" that can spread targeted disinformation by masquerading as a legitimate open-source AI model. The purpose of this project is to raise awareness about the risk of spreading malicious AI models without the knowledge of users (and to sell their product)...

On July 10th, the EU Commission adopted an adequacy decision for the proposed EU-U.S. Data Privacy Framework. This is exciting news for organizations, as many have been stuck in privacy "limbo" since the annulment of the previous EU-U.S. Data transfer mechanism, Privacy Shield, which was annulled due to challenges in court by privacy activist Max Schrems.

Researchers at Veriti have observed hundreds of spoofed domains following Meta’s launch of its Threads social media platform. “In recent weeks, we have observed a surge in the creation of suspicious domains, with over 700 domains related to Threads being registered daily,” the researchers write.

Using a new twist to bypass detection from security solutions, cyber attacks are now employing what will be construed as a benign image whose malicious intent can’t be traced. Threat actors need some means of getting a user to engage with malicious content – whether an attachment, link, or phone call, there needs to be some content within an email that provides the victim user with their next step.

As the rate of ransomware attacks steadily increased over time, there are clear indicators as to how these attacks are starting and, therefore, what can be done to stop them. With the exception of the Verizon Data Breach Investigations Report, we rarely get insight into specific industry verticals.

A phishing campaign is targeting the tourism and hospitality industries, according to researchers at Votiro. “In this instance, the hacker booked a room at an international hotel and submitted a request for the hotel to get in touch with them immediately via WhatsApp about an urgent issue,” the researchers write. “Once the hotel employee engaged the customer over WhatsApp, the hacker responded with their request.

Job scams are a rising form of socially engineered cybercrime. And while it’s easy to imagine the trouble they cause individuals who innocently fall for them (lost opportunities, identity theft, financial loss, and so on) this form of fraud also affects businesses.

Australia officially launched their National Anti-Scam Centre this week. With more than AUD $3.1 billion lost each year, Australians need support. With representatives from the banks, telecommunications industries and digital platforms, the intent of the center is to identify methods to disrupt all kinds of scams and reduce scam losses. While I completely support this initiative, it would be remiss of me not to highlight that the prevention of scams is perhaps as important as the cure.

If your organization uses Microsoft Teams, then you definitely want to hear about a new way bad actors are exploiting this newly discovered cyber attack tool. "TeamsPhisher," a new tool recently discovered on GitHub, gives cybercriminals a new way to deliver malicious files directly to any Teams user. The genesis of this new cyber attack tool was published by the US Navy Red Team due to a recently discovered vulnerability in Microsoft Teams.

An abnormally massive focus on business email compromise attacks in Europe has fueled an equally large growth in overall email attacks there, with the U.S. also seeing significant attack growth. We’ve seen a lot of recent reports looking back at 2022 to help provide insight into what to continue to expect this year. But new data from security analysts at Abnormal Security takes us well into June of this year, highlighting the problem of email-based attacks.

Apparently expanding efforts outside of Southeast Asian countries, this threat group’s known malware has shown up in a European healthcare facility, raising concerns for USB-based attacks. You’d think that literally no one uses USB drives anymore, making them a very improbable attack vector. And yet, the Camaro Dragon APT group has been tracked by security researchers at Check Point for well over a year, with them finding instances of attacks throughout all of last year and into this year.

A new threat alert from ConsumerAffairs and TrendMicro proves more than just shoppers will take advantage of Amazon’s upcoming Prime Day. As Prime Day approaches on July 11-12, ConsumerAffairs reports on a variety of scams bad actors are expected to use to take advantage of online shoppers looking for a good deal: Trend Micro's research team identified.

Attackers are increasingly using images in phishing to evade text-based security filters, according to researchers at INKY.

Email-based social engineering attacks have risen by 464% this year compared to the first half of 2022, according to a report by Acronis. Business email compromise (BEC) attacks have also increased significantly. “One out of 76, or 1.3%, of the received emails were malicious,” the researchers write. “Phishing remains the number one threat, with these attacks making up 73% of the total.

We are thrilled to announce that KnowBe4 has been named a leader in the latest G2 Grid Report that compares Security Awareness Training (SAT) vendors based on user reviews, customer satisfaction, popularity and market presence. The latest G2 Grid Report compares Security Awareness Training (SAT) vendors based on user reviews, customer satisfaction, popularity and market presence.

The largest port in Japan, Nagoya, is now the most recent victim of a ransomware attack. The attack impacts the operation of container terminals, as the port handles over two million containers each year. This port is also used by the Toyota Motor Corporation, one of the world’s largest automakers, to export most of its cars.

We are excited to announce that KnowBe4 has been named a leader in the Summer 2023 G2 Grid Report for Security Orchestration, Automation, and Response (SOAR) for the PhishER platform for the ninth consecutive quarter! The latest G2 Grid Report compares Security Orchestration, Automation, and Response (SOAR) Software vendors based on user reviews, customer satisfaction, popularity and market presence. Based on 202 G2 customer reviews, KnowBe4’s PhishER platform is the top ranked SOAR software.

The Better Business Bureau (BBB) has warned of a scam in which attackers pose as process servers in order to steal information and commit identity theft. “You receive a call from an unknown or blocked number from a person claiming to be a process server,” the Bureau says. “They might say there is a lien on your home or someone is taking you to court over unpaid medical bills.