Search This Blog
Powered by Blogger.

Blog Archive

Labels

About Me

Load More
Trendy Right Now

Popular Posts

Showing posts with label Internet. Show all posts
Technology
AI Chatbot ChatGPT Cloud Google Google Drive Internet Microsoft Technology

How ChatGPT prompt can allow cybercriminals to steal your Google Drive data


Chatbots and other AI tools have made life easier for threat actors. A recent incident highlighted how ChatGPT can be exploited to obtain API keys and other sensitive data from cloud platforms.

Prompt injection attacks leads to cloud access

Experts have discovered a new prompt injection attack that can turn ChatGPT into a hacker’s best friend in data thefts. Known as AgentFlayer, the exploit uses a single document to hide “secret” prompt instructions that target OpenAI’s chatbot. An attacker can share what appears to be a harmless document with victims through Google Drive, without any clicks.

Zero-click threat: AgentFlayer

AgentFlayer is a “zero-click” threat as it abuses a vulnerability in Connectors, for instance, a ChatGPT feature that connects the assistant to other applications, websites, and services. OpenAI suggests that Connectors supports a few of the world’s most widely used platforms. This includes cloud storage platforms such as Microsoft OneDrive and Google Drive.

Experts used Google Drive to expose the threats possible from chatbots and hidden prompts. 

GoogleDoc used for injecting prompt

The malicious document has a 300-word hidden malicious prompt. The text is size one, formatted in white to hide it from human readers but visible to the chatbot.

The prompt used to showcase AgentFlayer’s attacks prompts ChatGPT to find the victim’s Google Drive for API keys, link them to a tailored URL, and an external server. When the malicious document is shared, the attack is launched. The threat actor gets the hidden API keys when the target uses ChatGPT (the Connectors feature has to be enabled).

Othe cloud platforms at risk too

AgentFlayer is not a bug that only affects the Google Cloud. “As with any indirect prompt injection attack, we need a way into the LLM's context. And luckily for us, people upload untrusted documents into their ChatGPT all the time. This is usually done to summarize files or data, or leverage the LLM to ask specific questions about the document’s content instead of parsing through the entire thing by themselves,” said expert Tamir Ishay Sharbat from Zenity Labs.

“OpenAI is already aware of the vulnerability and has mitigations in place. But unfortunately, these mitigations aren’t enough. Even safe-looking URLs can be used for malicious purposes. If a URL is considered safe, you can be sure an attacker will find a creative way to take advantage of it,” Zenith Labs said in the report.

Read more »
Windows
AI Akira Ransomware Bugs Internet malware Ransomware Windows

Akira ransomware turns off Windows Defender to install malware on Windows devices

Akira ransomware turns off Windows Defender to install malware on Windows devices

Akira ransomware strikes again. This time, it has abused an Intel CPU tuning driver to stop Microsoft Defender in attacks from EDRs and security tools active on target devices.

Windows defender turned off for attacks

The exploited driver is called “rwdrv.sys” (used by ThrottleStop), which the hackers list as a service that allows them to gain kernel-level access. The driver is probably used to deploy an additional driver called “hlpdrv.sys,” a hostile tool that modifies Windows Defender to shut down its safety features.

'Bring your own vulnerable driver' attack

Experts have termed the attack “Bring your vulnerable driver (BYOVD), where hackers use genuine logged-in drivers that have known bugs that can be exploited to get privilege escalation. The driver is later used to deploy a hostile that turns off Microsoft Defender. According to the experts, the additional driver hlpdrv.sys is “similarly registered as a service. When executed, it modifies the DisableAntiSpyware settings of Windows Defender within \REGISTRY\MACHINE\SOFTWARE\Policies\Microsoft\Windows Defender\DisableAntiSpyware.” The malware achieves this by executing regedit.exe. 

Discovery of the Akira ransomware attack

The technique was observed by Guidepoint Security, which noticed repeated exploitation of the rwdrv.sys driver in Akira ransomware attacks. The experts flagged this tactic due to its ubiquity in the latest Akira ransomware incidents. “This high-fidelity indicator can be used for proactive detection and retroactive threat hunting,” the report said. 

To assist security experts in stopping these attacks, Guidepoint Security has offered a YARA rule for hlpdrv.sys and complete indicators of compromise (IoCs) for the two drivers, as well as their file paths and service names.

SonicWall VPN attack

Akira ransomware was also recently associated with SonicWall VPN attacks. The threat actor used an unknown bug. According to Guidepoint Security, it could not debunk or verify the abuse of a zero-day flaw in SonicWall VPNs by the Akira ransomware gang. Addressing the reports, SonicWall has advised to turn off SSLVPN, use two-factor authentication (2FA), remove inactive accounts, and enable Botnet/Geo-IP safety.

The DFIR report has also released a study of the Akira ransomware incidents, revealing the use of Bumblebee malware loader deployed through trojanized MSI loaders of IT software tools.

Read more »
Technology
DarkBit Encyption Internet Iran Israel Ransomware Technology

Experts decoded encryption keys used by DarkBit ransomware gang

Experts decoded encryption keys used by DarkBit ransomware gang

Encryption key for Darkbit ransomware

Good news for people affected by the DarkBit ransomware: experts from Profero have cracked the encryption process, allowing victims to recover their files for free without paying any ransom.

However, the company has not yet released the decryptor. The National Cyber Directorate from Israel connected the DarkBit ransomware operation to the Iran-nexus cybercriminal gang called “MuddyWater APT.”

How the attack started

After a DarkBit ransomware attack in 2023, Profero encrypted various VMware ESXi servers, which were believed as retaliation for Iranian drone attacks. The threat actors did not negotiate the ransom and emphasized disrupting operations and campaigns to damage the target’s reputation.

The gang posed as pro-Iran hackers and had a history of attacking Israeli agencies. In this incident, the gang asked for 80 Bitcoins and had anti-Israel messages in ransom notes. Profero, however, cracked the encryption, allowing free recovery.

How did the experts find out

While studying DarkBit ransomware, experts discovered that its AES-128-CBC key generation tactic gave weak and predictable keys. Profero used file timestamps and a known VMDK header to limit the keyspace to billions of probabilities, allowing effective brute-force.

“We made use of an AES-128-CBC key-breaking harness to test if our theory was correct, as well as a decryptor which would take an encrypted VMDK and a key and IV pair as input to produce the unencrypted file. The harness ran in a high-performance environment, allowing us to speed through the task as quickly as possible, and after a day of brute-forcing, we were successful!” according to the Profero report. 

Persistent effort led to successful encryption

The experts had proven that it was possible and got the key. They continued brute-forcing another VMDK. This method, however, was not scalable for the following reasons:

  • Each VMDK would require a day for the experts to decrypt
  • The harness resides in an HPC environment and is difficult to scale

“While expensive, it ended up being possible. We decided to once again take a look at any potential weaknesses in the crypto,” Proffero experts said.

The experts made a tool to check all possible seeds and create key and IV pairs to match them against VMDK headers. This allowed them to restore the decryption keys. Profero also leveraged the scarce VMDK files, where most of the content was unencrypted, as the ransom was partially encrypted. The experts then directly recovered the most needed files, avoiding brute-force decryption for most of the data.

Read more »
Technology
AI Corporate Internet Ransomware Technology

Why Companies Keep Ransomware Payments Secret


Companies hiding ransomware payments

Ransomware attacks are ugly. For every ransomware attack news story we see in our feed, a different reality hides behind it. Victims secretly pay their attackers. The shadow economy feeds on corporate guilt and regulatory hysteria.

Companies are hiding the true numbers of ransomware incidents. For each attack that makes headlines, five more companies quietly push it under the carpet, keeping it secret, and wire cryptocurrency payments to attackers, in hopes of avoiding detection. We can call it corporate cowardice, but this gives confidence to the ransomware cybercriminals. It costs the victims $57 billion annually and directly damages the devices that we use.

Paying attackers fuels future attacks

According to the FBI, it “does not support paying a ransom in response to a ransomware attack. Paying a ransom doesn’t guarantee you or your organization will get any data back. It also encourages perpetrators to target more victims and offers an incentive for others to get involved in this type of illegal activity.

The patches in our smartphones exist because companies suffer attacks. Our laptop endpoint protection was developed from enterprise systems compromised by ransomware groups that used secret corporate ransoms to invest in more advanced malware. 

Corporate guilt is a reason for keeping payments secret

Few experts believe that for every reported ransomware attack, five more are kept hidden, and the payments are made secretly to escape market panic and regulatory enquiry. The transactions travel through the cryptocurrency networks, managed by negotiators who deal in digital extortion.

Companies justify their actions by keeping quiet to avoid regulatory scrutiny and falling stock prices, and quietly resolving the issue. The average ransom demand is around $5.2 million, but actual payments hit $1 million, a relative discount that may fund future ransomware attacks.

According to Gadget Review, “This secrecy creates a feedback loop more vicious than algorithmic social media engagement. Ransomware groups reinvest payments into advanced encryption, better evasion techniques, and expanded target lists that inevitably include the consumer technology ecosystem you depend on daily.”

It adds that “even as payment rates drop to historic lows—just 25% of victims now pay—the total damage keeps climbing. Companies face average costs exceeding $5.5 million per attack, combining ransom payments, recovery expenses, and reputation management.”

Read more »
Technology
AI Dark Web Data Leak Internet Ransomware Technology

Ransomware Attacks Threaten CEOs to Get Results


Ransomware gangs are getting desperate for results. Generally known for encrypting and leaking data on the internet, they have now started blackmailing CEOs with physical violence. 

CEO's get physically threatened

Cybersecurity experts from Semperis say that over the past year, in 40% of ransomware attacks, the CEOs of the victim company were physically attacked, which is particularly prevalent in US-based organizations, at 46%.

However, even paying the attackers is not enough. The research revealed that over 55% of businesses that paid a ransom had to do so multiple times, with around 29% of those firms paying three or more times, and 15% didn’t even receive decryption keys, while in a few cases, they received corrupted keys.

New ransomware tactics 

Blackmailing to file a regulatory complaint is also a famous tactic, Semperis said. It was found in 47% of attacks, increasing to 58% in the US. 

In 2023, the notorious BlackCat ransomware gang reported one of its victims to the Securities and Exchange Commission (SEC) to make them pay. This was done because the SEC requires organizations to report about a cybersecurity incident if there is a breach, which includes the SEC's four-day disclosure rule for publicly traded businesses.

Ransomware on the rise

Ransomware attacks have threatened businesses and the cybersecurity industry for decades, constantly evolving and outsmarting security professionals. The attacks started with encryption, but the companies started mitigating by having offline backups of all the important data.

Ransomware actors then turned to stealing data and blackmailing to leak it on the web if the ransom was not paid. Known as “double extortion,” the technique works really well. Some threat actors even dropped the encryption part totally and now focus on stealing files. But many companies still don’t cave in, forcing cybercriminals to go to extreme lengths. 

New tactics

In a few cases, the attackers combine the encryption of the back-end with a DDoS on the front-end, stopping the business entirely. Semperis CEO  Mickey Bresman said that while some “circumstances might leave the company in a non-choice situation, we should acknowledge that it's a down payment on the next attack.”

"Every dollar handed to ransomware gangs fuels their criminal economy, incentivizing them to strike again. The only real way to break the ransomware scourge is to invest in resilience, creating an option to not pay ransom," he commented.

Read more »
Technology
AI Internet Login Password Tech News Technology

Proton Launches New Authenticator App With Standalone Features



Proton has released Proton Authenticator, an independent, standalone 2-factor authentication (2FA) app for macOS, Windows, Android, Linux, and iOS. 2FA verification applications are offline tools that create time-based OTPs that expire within 20 seconds, and can also be used with passwords when signing into offline accounts, offering a second layer of verification.

A Swiss tech company, Proton, is famous for its privacy-focused end-to-end encryption services such as

Integration of an authenticator app adds to the company’s product portfolio and brings a privacy-specialized tool that challenges competitors that are mostly ad-supported, closed-source, and trap customers into proprietary ecosystems.

But Proton Authenticator doesn’t have ads, vendor lock-in, or trackers, and uses no Proton account. According to the company, “Proton Authenticator is built with the same values that power everything Proton does: privacy, transparency, and user-first security.” "The company is now bringing these standards to the 2FA space – offering a secure, easy-to-use, and encrypted alternative to apps like Google Authenticator that further lock users into Big Tech's surveillance ecosystems." 

The application is open-source, but it takes around two weeks for the Proton team to release the source code of the latest tools on GitHub. The app has end-to-end encryption, which supports safe cross-device sync and shift to other platforms via easy-to-use import and export features. A lot of apps, such as Microsoft and Authy, cannot export the time-based OTP seeds feature.

The Proton Authenticator also provides automatic encrypted backups and app lock with PIN or biometrics, giving an extra security layer.

“Proton Authenticator will make it easier for everyone to log in to their online accounts securely, a vital step in making the internet a safer place,” read the product statement.

Read more »
Windows
Artificial Intelligence DevilsTongue Internet malware Microsoft Spyware surveillance Windows

DevilsTongue Spyware Attacking Windows System, Linked to Saudi Arabia, Hungary


Cybersecurity experts have discovered a new infrastructure suspected to be used by spyware company Candiru to target computers via Windows malware.

DevilsTongue spyware targets Windows systems

The research by Recorded Future’s Insikt Group disclosed eight different operational clusters associated with the spyware, which is termed as DevilsTongue. Five are highly active, including clusters linked to Hungary and Saudi Arabia. 

About Candiru’ spyware

According to the report, the “infrastructure includes both victim-facing components likely used in the deployment and [command and control] of Candiru’s DevilsTongue spyware, and higher-tier infrastructure used by the spyware operators.” While a few clusters directly handle their victim-facing infrastructure, others follow an intermediary infrastructure layers approach or through the Tor network, which allows threat actors to use the dark web.

Additionally, experts discovered another cluster linked to Indonesia that seemed to be active until November 2024. Experts couldn’t assess whether the two extra clusters linked with Azerbaijan are still active.

Mode of operation

Mercenary spyware such as DevilsTongue is infamous worldwide, known for use in serious crimes and counterterrorism operations. However, it also poses various legal, privacy, and safety risks to targets, their companies, and even the reporter, according to Recorded Future.

Windows itself has termed the spyware Devil's Tongue. There is not much reporting on its deployment techniques, but the leaked materials suggest it can be delivered via malicious links, man-in-the-middle attacks, physical access to a Windows device, and weaponized files. DevilsTongue has been installed via both threat actor-controlled URLs that are found in spearphishing emails and via strategic website attacks known as ‘watering hole,’ which exploit bugs in the web browser.

Insikt Group has also found a new agent inside Candiru’s network that is suspected to have been released during the time when Candiru’s assets were acquired by Integrity Partners, a US-based investment fund. Experts believe that a different company might have been involved in the acquisition.

How to stay safe?

In the short term, experts from Recorded Future advise defenders to “implement security best practices, including regular software updates, hunting for known indicators, pre-travel security briefings, and strict separation of personal and corporate devices.” In the long term, organizations are advised to invest in robust risk assessments to create effective policies.

Read more »
UK
Data protection Internet OSA Privacy Safe Browsing Tech News UK

UK'S Online Safety Act Faces Criticism, Doesn't Make Children Safer Online

UK'S Online Safety Act Faces Criticism, Doesn't Make Children Safer Online

The implementation of a new law to protect the online safety of children in the UK has caught criticism from digital rights activist groups, politicians, free-speech campaigners, tech companies, content creators, digital rights advocacy groups, and others. The Online Safety Act (OSA) came into effect on July 25th. The legislation aims to protect children from accessing harmful content on the internet. Why problematic?

Safe internet for kids?

However, the act also poses potential privacy risks. Certain provisions of the act require companies behind websites in the UK to prevent users under 18 from accessing dangerous content such as pornography and content related to eating disorders, self-harm, or worse- suicide. The act also mandates companies to give minors age-appropriate access to other types of material concerning abusive or hateful, and bullying content.

Tech companies' role

In compliance with the OSA provisions, platforms have enforced age authentication steps to verify the ages of users on their sites or apps. These include platforms like X, Discord, Bluesky, and Reddit; porn besides such as YouPorn and Pornhub, and music streaming services like Spotify, which also require users to provide face scans to view explicit content.

As a result, VPN companies have experienced a major surge in VPN subscriptions in the UK over the past few weeks. Proton VPN reported a 1800% hike in UK daily sign-ups, according to the BBC.

As the UK is one of the first democratic countries after Australia to enforce such strict content regulations on tech companies, it has garnered widespread criticism, becoming a watched test case, and might impact online safety regulation in other countries such as India.

About OSA rules

To make the UK the ‘safest place’ in the world to be online, the OSA Act was signed into law in 2023. It includes provisions that impose a burden on social media platforms to remove illegal content as well as implement transparency and accountability measures. But the British government website claims that the strictest provisions in the OSA are aimed a promoting online safety of under-18 children.

The provisions apply to companies that exist even outside the UK. Companies had until July 24, 2025, to assess if their websites were likely to be accessed by children and complete their evaluation of the harm to children.

Read more »
Technology
Blockchain Cloud Data Digital ID Internet NDAChain Technology

Vietnam Launches NDAChain for National Data Security and Digital Identity


Vietnam has launched NDAChain, a new blockchain network that allows only approved participants to join. The move is aimed at locking down Vietnam’s government data. 

About NDAChain

The network is built by the National Data Association and managed by the Ministry of Public Security’s Data Innovation and Exploitation Center. It will serve as the primary verification layer for tasks such as supply-chain logs, school transcripts, and hospital records.

According to experts, NDAChain is based on a hybrid model, relying on a Proof-of-Authority mechanism to ensure only authorized nodes can verify transactions. It also adds Zero-Knowledge-Proofs to protect sensitive data while verifying its authenticity. According to officials, NDAChain can process between 1,200 and 3,600 transactions per second, a statistic that aims to support faster verifications in logistics, e-government, and other areas. 

Two new features

The networks have two main features: NDA DID offers digital IDs that integrate with Vietnam’s current VNeID framework, allowing users to verify their IDs online when signing documents or using services. On the other hand, NDATrace provides end-to-end product tracking via GS1 and EBSI Trace standards. Items are tagged with unique identifiers that RFID chips or QR codes can scan, helping businesses prove verification to overseas procurers and ease recalls in case of problems.

Privacy layer and network protection

NDAChain works as a “protective layer” for Vietnam’s digital infrastructure, built to scale as data volume expands. Digital records can be verified without needing personal details due to the added privacy tools. The permissioned setup also offers authorities more control over people joining the network. According to reports, total integration with the National Data Center will be completed by this year. The focus will then move towards local agencies and universities, where industry-specific Layer 3 apps are planned for 2026.

According to Vietnam Briefing, "in sectors such as food, pharmaceuticals, and health supplements, where counterfeit goods remain a persistent threat, NDAChain enables end-to-end product origin authentication. By tracing a product’s whole journey from manufacturer to end-consumer, businesses can enhance brand trust, reduce legal risk, and meet rising regulatory demands for transparency."

Read more »
Technology
AI ChatGPT Data Data Theft Internet Technology

A Massive 800% Rise in Data Breach Incidents in First Half of 2025


Cybersecurity experts have warned of a significant increase in identity-based attacks, following the revelation that 1.8 billion credentials were stolen in the first half of 2025, representing an 800% increase compared to the previous six months.

Data breach attacks are rising rapidly

Flashpoint’s Global Threat Intelligence Index report is based on more than 3.6 petabytes of data studied by the experts. Hackers stole credentials from 5.8 million compromised devices, according to the report. The significant rise is problematic as stolen credentials can give hackers access to organizational data, even when the accounts are protected by multi-factor authentication (MFA).

The report also includes details that concern security teams.

About the bugs

Until June 2025, the firm has found over 20,000 exposed bugs, 12,200 of which haven’t been reported in the National Vulnerability Database (NVD). This means that security teams are not informed. 7000 of these have public exploits available, exposing organizations to severe threats.

According to experts, “The digital attack surface continues to expand, and the volume of disclosed vulnerabilities is growing at a record pace – up by a staggering 246% since February 2025.” “This explosion, coupled with a 179% increase in publicly available exploit code, intensifies the pressure on security teams. It’s no longer feasible to triage and remediate every vulnerability.”

Surge in ransomware attacks

Both these trends can cause ransomware attacks, as early access mostly comes through vulnerability exploitation or credential hacking. Total reports of breaches have increased by 179% since 2024, manufacturing (22%), technology (18%), and retail (13%) have been hit the most. The report has also disclosed 3104 data breaches in the first half of this year, linked to 9.5 billion hacked records.

2025 to be record year for data breaches

Flashpoint reports that “Over the past four months, data breaches surged by 235%, with unauthorized access accounting for nearly 78% of all reported incidents. Data breaches are both the genesis and culmination of threat actor campaigns, serving as a source of continuous fuel for cybercrime activity.” 

In June, the Identity Theft Resource Center (ITRC) warned that 2025 could become a record year for data cyberattacks in the US.

Read more »
Ransomware
AI Bugs ChatGPT Cyber Attacks Generative AI Internet Ransomware

AI-supported Cursor IDE Falls Victim to Prompt Injection Attacks


Experts have found a bug called CurXecute that is present in all variants of the AI-supported code editor Cursor and can be compromised to run remote code execution (RCE), along with developer privileges. 

About the bug

The security bug is now listed as CVE-2025-54135 and can be exploited by giving the AI agent a malicious prompt to activate threat actor control commands. 

The Cursor combined development environment (IDE) relies on AI agents to allow developers to code quicker and more effectively, helping them to connect with external systems and resources using Model Context Protocol (MCP).

According to the experts, a threat actor effectively abusing the CurXecute bug could trigger ransomware and ransomware data theft attacks. 

Prompt-injection 

CurXecute shares similarities to the EchoLeak bug in Microsoft 365 CoPilot that hackers can use to extort sensitive data without interacting with the users. 

After finding and studying EchoLeak, the experts from the cybersecurity company Aim Security found that hackers can even exploit the local AI agent.

Cursor IDE supports the MCP open-standard framework, which increases an agent’s features by connecting it to external data tools and sources.

Agent exploitation

But the experts have warned that doing so can exploit the agent, as it is open to external, suspicious data that can impact its control flow. The threat actor can take advantage by hacking the agent’s session and features to work as a user.

According to the experts, Cursor doesn’t need permission to run new entries to the ~/.cursor/mcp.json file. When the target opens the new conversation and tells the agent to summarize the messages, the shell payload deploys on the device without user authorization.

“Cursor allows writing in-workspace files with no user approval. If the file is a dotfile, editing it requires approval, but creating one if it doesn't exist doesn't. Hence, if sensitive MCP files, such as the .cursor/mcp.json file, don't already exist in the workspace, an attacker can chain an indirect prompt injection vulnerability to hijack the context to write to the settings file and trigger RCE on the victim without user approval,” Cursor said in a report.

Read more »
Technology
Browser Firefox Internet Mozilla Technology

Malicious Firefox Extension Steals Verification Tokens: Update to stay safe


Credential theft and browser security were commonly found in Google Chrome browsers due to its wide popularity and usage. Recently, however, cyber criminals have started targeting Mozilla Firefox users. A recent report disclosed a total of eight malicious Firefox extensions that could spy on users and even steal verification tokens.

About the malicious extension

Regardless of the web browser we use, criminals are always on the hunt. Threat actors generally prefer malicious extensions or add-ons; therefore, browser vendors like Mozilla offer background protections and public support to minimize these threats as much as possible. Despite such a measure, on July 4th, the Socket Threat Research Team's report revealed that threat actors are still targeting Firefox users. 

According to Kush Pandya, security engineer at Socket Threat Research Team, said that while the “investigation focuses on Firefox extensions, these threats span the entire browser ecosystem.” However, the particular Firefox investigation revealed a total of eight potentially harmful extensions, including user session hijacking to earn commissions on websites, redirection to scam sites, surveillance via an invisible iframe tracking method, and the most serious: authentication theft.

How to mitigate the Firefox attack threat

Users are advised to read the technical details of the extensions. According to Forbes, Mozilla is taking positive action to protect Firefox users from such threats. The company has taken care of the extensions mentioned in the report. According to Mozilla, the malicious extension impacted a very small number of users; some of the extensions have been shut down. 

“We help users customize their browsing experience by featuring a variety of add-ons, manually reviewed by our Firefox Add-ons team, on our Recommended Extensions page,” said a Firefox spokesperson. To protect the users, Mozilla has disabled “extensions that compromise their safety or privacy, or violate its policies, and continuously works to improve its malicious add-on detection tools and processes.”

How to stay safe?

To protect against these threats, Mozilla has advised users to Firefox users to take further steps, cautioning that such extensions are made by third parties. Users should check the extension rating and reviews, and be extra careful of extensions that need excessive permissions that are not compatible with what the extension claims to do. If any extension seems to be malicious, “users should report it for review,” a Firefox spokesperson said. 

Read more »
Technology
AI AI Prompt ChatGPT Gemini Google Internet Technology

Google Gemini Bug Exploits Summaries for Phishing Scams


False AI summaries leading to phishing attacks

Google Gemini for Workspace can be exploited to generate email summaries that appear legitimate but include malicious instructions or warnings that direct users to phishing sites without using attachments or direct links.

Google Gemini for Workplace can be compromised to create email summaries that look real but contain harmful instructions or warnings that redirect users to phishing websites without using direct links or attachments. 

Similar attacks were reported in 2024 and afterwards; safeguards were pushed to stop misleading responses. However, the tactic remains a problem for security experts. 

Gemini for attack

A prompt-injection attack on the Gemini model was revealed via cybersecurity researcher Marco Figueoa, at 0din, Mozilla’s bug bounty program for GenAI tools. The tactic creates an email with a hidden directive for Gemini. The threat actor can hide malicious commands in the message body text at the end via CSS and HTML, which changes the font size to zero and color to white. 

According to Marco, who is GenAI Bug Bounty Programs Manager at Mozilla, “Because the injected text is rendered in white-on-white (or otherwise hidden), the victim never sees the instruction in the original message, only the fabricated 'security alert' in the AI-generated summary. Similar indirect prompt attacks on Gemini were first reported in 2024, and Google has already published mitigations, but the technique remains viable today.”

Gmail does not render the malicious instruction as there are no attachments or links present, and the message may reach the victim’s inbox. If the receiver opens the email and asks Gemini to make a summary of the received mail, the AI tool will parse the invisible directive and create the summary. Figueroa provides an example of Gemini following hidden prompts, accompanied by a security warning that the victim’s Gmail password and phone number may be compromised.

Impact

Supply-chain threats: CRM systems, automated ticketing emails, and newsletters can become injection vectors, changing one exploited SaaS account into hundreds of thousands of phishing beacons.

Cross-product surface: The same tactics applies to Gemini in Slides, Drive search, Docs and any workplace where the model is getting third-party content.

According to Marco, “Security teams must treat AI assistants as part of the attack surface and instrument them, sandbox them, and never assume their output is benign.”

Read more »
Windows
AI Android Apple Extortion Internet malware Ransomware Windows

Latest Malware "Mamona" Attacks Locally, Hides by Self Deletion

Latest Malware "Mamona" Attacks Locally, Hides by Self Deletion

Cybersecurity experts are tracing Mamona, a new ransomware strain that is famous for its stripped-down build and silent local execution. Experts believe that the ransomware prevents the usual command-and-control (C2) servers, choosing instead a self-contained method that moves past tools relying on network traffic analysis.  

The malware is executed locally on a Windows system as a standalone binary file. The offline approach reveals a blind spot in traditional defenses, raising questions about how even the best antivirus and detection mechanisms will work when there is no network.

Self-deletion and escape techniques make detection difficult

Once executed, it starts a three-second delay via a modified ping command, ”cmd.exe /C ping 127.0.0.7 -n 3 > Nul & Del /f /q.” After this, it self-deletes. The self-deletion helps to eliminate forensic artifacts that make it difficult for experts to track or examine the malware after it has been executed. 

The malware uses 127.0.0.7 instead of the popular 127.0.0.1, which helps in evading detection measures. This tactic escapes simple detection tests and doesn’t leave digital traces that older file-based scanners might tag. The malware also drops a ransom note titled README.HAes.txt and renames impacted files with the .HAes extension. This means the encryption was successful. 

“We integrated Sysmon with Wazuh to enrich logs from the infected endpoint and created Wazuh detection rules to identify malicious behaviour associated with Mamona ransomware,” said Wazuh in a blog post.

Spotting Mamona

Wazuh has alerted that the “plug-and-play” nature of the malware makes it easy for cybercriminals and helps in the commodization of ransomware. This change highlights an urgent need for robust inspections of what stands as the best ransomware protection when such attacks do not need remote control infrastructure. Wazu’s method to track Mamona involves combining Sysom for log capture and employing custom rules to flag particular behaviours like ransom note creation and ping-based delays.

According to TechRadar, “Rule 100901 targets the creation of the README.HAes.txt file, while Rule 100902 confirms the presence of ransomware when both ransom note activity and the delay/self-delete sequence appear together.”

Read more »
Windows
AI Android Bug Data Exploit Internet Technology Vulnerability Windows

CISA Lists Citrix Bleed 2 as Exploit, Gives One Day Deadline to Patch

CISA Lists Citrix Bleed 2 as Exploit, Gives One Day Deadline to Patch

CISA confirms bug exploit

The US Cybersecurity & Infrastructure Security Agency (CISA) confirms active exploitation of the CitrixBleed 2 vulnerability (CVE-2025-5777 in Citrix NetScaler ADC and Gateway. It has given federal parties one day to patch the bugs. This unrealistic deadline for deploying the patches is the first since CISA issued the Known Exploited Vulnerabilities (KEV) catalog, highlighting the severity of attacks abusing the security gaps. 

About the critical vulnerability

CVE-2025-5777 is a critical memory safety bug (out-of-bounds memory read) that gives hackers unauthorized access to restricted memory parts. The flaw affects NetScaler devices that are configured as an AAA virtual server or a Gateway. Citrix patched the vulnerabilities via the June 17 updates. 

After that, expert Kevin Beaumont alerted about the flaw’s capability for exploitation if left unaddressed, terming the bug as ‘CitrixBleed 2’ because it shared similarities with the infamous CitrixBleed bug (CVE-2023-4966), which was widely abused in the wild by threat actors.

What is the CitrixBleed 2 exploit?

According to Bleeping Computer, “The first warning of CitrixBleed 2 being exploited came from ReliaQuest on June 27. On July 7, security researchers at watchTowr and Horizon3 published proof-of-concept exploits (PoCs) for CVE-2025-5777, demonstrating how the flaw can be leveraged in attacks that steal user session tokens.”

The rise of exploits

During that time, experts could not spot the signs of active exploitation. Soon, the threat actors started to exploit the bug on a larger scale, and after the attack, they became active on hacker forums, “discussing, working, testing, and publicly sharing feedback on PoCs for the Citrix Bleed 2 vulnerability,” according to Bleeping Computers. 

Hackers showed interest in how to use the available exploits in attacks effectively. The hackers have become more active, and various exploits for the bug have been published.

Now that CISA has confirmed the widespread exploitation of CitrixBleed 2 in attacks, threat actors may have developed their exploits based on the recently released technical information. CISA has suggested to “apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable.”

Read more »
Technology
AI Cloud Data Google Internet Security Breach Technology

Security Breach Reveals "Catwatchful" Spyware is Snooping on Users

Security Breach Reveals "Catwatchful" Spyware is Snooping on Users

A security bug in a stealthy Android spyware operation, “Catwatchful,” has exposed full user databases affecting its 62,000 customers and also its app admin. The vulnerability was found by cybersecurity expert Eric Daigle reported about the spyware app’s full database of email IDs and plaintext passwords used by Catwatchful customers to access stolen data from the devices of their victims. 

Most of the victims were based in India, Argentina, Peru, Mexico, Colombia, Bolivia, and Ecuador. A few records date back to 2018. The leaked database also revealed the identity of the Catwatchful admin called Omar Soca Char.

The Catwatchful database also revealed the identity of the spyware operation’s administrator, Omar Soca Charcov, a developer based in Uruguay.

About Catwatchful

Catwatchful is a spyware that pretends to be a child monitoring app, claiming to be “invisible and can not be detected,” while it uploads the victim’s data to a dashboard accessible to the person who planted the app. The stolen data includes real-time location data, victims’ photos, and messages.  The app can also track live ambient audio from the device’s mic and access the phone camera (both front and rear).

Catwatchful and similar apps are banned on app stores, and depend on being downloaded and deployed by someone having physical access to a victim’s phone. These apps are famous as “stalkerware” or “spouseware” as they are capable of unauthorized and illegal non-consensual surveillance of romantic partners and spouses. 

Rise of spyware apps

The Catwatchful incident is the fifth and latest in this year’s growing list of stalkerware scams that have been breached, hacked, or had their data exposed. 

How was the spyware found?

Daigle has previously discovered stalkerware exploits. Catwatchful uses a custom-made API, which the planted app uses to communicate to send data back to Catwatchful servers. The stalkerware also uses Google Firebase to host and store stolen data. 

According to Techradar, the “data was stored on Google Firebase, sent via a custom API that was unauthenticated, resulting in open access to user and victim data. The report also confirms that, although hosting had initially been suspended by HostGator, it had been restored via another temporary domain."

Read more »
Windows
Apple Google Internet Microsoft Passkeys Passwords Technology Windows

Ditch Passwords, Use Passkeys to Secure Your Account

Ditch Passwords, Use Passkeys to Secure Your Account

Ditch passwords, use passkeys

Microsoft and Google users, in particular, have been warned about ditching passwords for passkeys. Passwords are easy to steal and can unlock your digital life. Microsoft has been at the forefront, confirming it will delete passwords for more than a billion users. Google, too, has warned that most of its users will have to add passkeys to their accounts. 

What are passkeys?

Instead of a username and password, passkeys use our device security to log into our account. This means that there is no password to hack and no two-factor authentication codes to bypass, making it phishing-resistant.

At the same time, the Okta team warned that it found threat actors exploiting v0, an advanced GenAI tool made by Vercelopens, to create phishing websites that mimic real sign-in webpages

Okta warns users to not use passwords

A video shows how this works, raising concerns about users still using passwords to sign into their accounts, even when backed by multi-factor authentication, and “especially if that 2FA is nothing better than SMS, which is now little better than nothing at all,” according to Forbes. 

According to Okta, “This signals a new evolution in the weaponization of GenAI by threat actors who have demonstrated an ability to generate a functional phishing site from simple text prompts. The technology is being used to build replicas of the legitimate sign-in pages of multiple brands, including an Okta customer.”

Why are passwords not safe?

It is shocking how easy a login webpage can be mimicked. Users should not be surprised that today’s cyber criminals are exploiting and weaponizing GenAI features to advance and streamline their phishing attacks. AI in the wrong hands can have massive repercussions for the cybersecurity industry.

According to Forbes, “Gone are the days of clumsy imagery and texts and fake sign-in pages that can be detected in an instant. These latest attacks need a technical solution.”

Users are advised to add passkeys to their accounts if available and stop using passwords when signing in to their accounts. Users should also ensure that if they use passwords, they should be long and unique, and not backed up by SMS 2-factor authentication. 

Read more »
Ransomware
AI Cyber Security Dark Web Extortion Internet Ransomware

Investigation Reveals Employee Secretly Helped in Extortion Payments

Investigation Reveals Employee Secretly Helped in Extortion Payments

Employee helped in ransomware operations

Federal agents are investigating allegations that a former employee of a Chicago-based firm, DigitalMint, which specializes in cryptocurrency payments and ransomware negotiations, may have profited by collaborating with hackers in extortion cases. Founded in 2014, DigitalMint operates under the name Red Leaf Chicago and is recognized for securing cryptocurrency payments for companies that face ransomware threats. 

About DigitalMint

DigitalMint has taken over 2,000 ransomware cases since 2017, offering services like direct negotiations with hackers and incident response. The clients range from small firms to Fortune 500 companies. 

DigitalMint President Marc Jason told partner firms that the US Department of Justice (DoJ) is investigating the allegations. The employee (identity unknown) was sacked soon after the scam was found. According to Bloomberg, Grens said, “As soon as we were able, we began communicating the facts to affected stakeholders.” 

About the investigation

DigitalMint is currently working with the DoJ, and it clarified that the company is not the target of investigation. Grens did not provide more details as the investigation is ongoing. The DoJ declined to offer any comments. 

The incident has led a few firms to warn clients against dealing with DigitalMint, concerned about the dangers involved in ransomware deals. Ransomware attacks can compromise systems, leak sensitive information, and encrypt data. The ransom demands sometimes go upto millions of dollars, worldwide, the extortion attacks cost billions of dollars every year.

Is ransomware negotiation worth it?

The controversy has also raised questions about conflicts of interest in the ransomware negotiation industry. According to James Talientoo, chief executive of the cyber intelligence services company AFTRDRK, “A negotiator is not incentivized to drive the price down or to inform the victim of all the facts if the company they work for is profiting off the size of the demand paid. Plain and simple.”

Security experts cautioned that paying ransom is a dangerous effort, even when done by expert ransom negotiation firms. A payment helps in furthering the operations of ransomware gangs, and sometimes it can also lead to further attacks.

Read more »
North Korea
AI Artificial Intelligence Cloud Cyber Crime Data Deepfakes Extortion Internet lazarus North Korea

Amid Federal Crackdown, Microsoft Warns Against Rising North Korean Jobs Scams

Amid Federal Crackdown, Microsoft Warns Against Rising North Korean Jobs Scams

North Korean hackers are infiltrating high-profile US-based tech firms through scams. Recently, they have even advanced their tactics, according to the experts. In a recent investigation by Microsoft, the company has requested its peers to enforce stronger pre-employment verification measures and make policies to stop unauthorized IT management tools. 

Further investigation by the US government revealed that these actors were working to steal money for the North Korean government and use the funds to run its government operations and its weapons program.  

US imposes sanctions against North Korea

The US has imposed strict sanctions on North Korea, which restrict US companies from hiring North Korean nationals. It has led to threat actors making fake identities and using all kinds of tricks (such as VPNs) to obscure their real identities and locations. This is being done to avoid getting caught and get easily hired. 

Recently, the threat actors have started using spoof tactics such as voice-changing tools and AI-generated documents to appear credible. In one incident, the scammers somehow used an individual residing in New Jersey, who set up shell companies to fool victims into believing they were paying a legitimate local business. The same individual also helped overseas partners to get recruited. 

DoJ arrests accused

The clever campaign has now come to an end, as the US Department of Justice (DoJ) arrested and charged a US national called Zhenxing “Danny” Wanf with operating a “year-long” scam. The scheme earned over $5 million. The agency also arrested eight more people - six Chinese and two Taiwanese nationals. The arrested individuals are charged with money laundering, identity theft, hacking, sanctions violations, and conspiring to commit wire fraud.

In addition to getting paid in these jobs, which Microsoft says is a hefty payment, these individuals also get access to private organization data. They exploit this access by stealing sensitive information and blackmailing the company.

Lazarus group behind such scams

One of the largest and most infamous hacking gangs worldwide is the North Korean state-sponsored group, Lazarus. According to experts, the gang extorted billions of dollars from the Korean government through similar scams. The entire campaign is popular as “Operation DreamJob”. 

"To disrupt this activity and protect our customers, we’ve suspended 3,000 known Microsoft consumer accounts (Outlook/Hotmail) created by North Korean IT workers," said Microsoft.

Read more »
Ransomware
Artificial Intelligence Cyber Security Digital Landscape insurance Internet Ransomware

How Ransomware Has Impacted Cyber Insurance Assessment Approach

How Ransomware Has Impacted Cyber Insurance Assessment Approach

Cyber insurance and ransomware

The surge in ransomware campaigns has compelled cyber insurers to rethink their security measures. Ransomware attacks have been a threat for many years, but it was only recently that threat actors realized the significant financial benefits they could reap from such attacks. The rise of ransomware-as-a-service (RaaS) and double extortion tactics has changed the threat landscape, as organizations continue to fall victim and suffer data leaks that are accessible to everyone. 

According to a 2024 threat report by Cisco, "Ransomware remains a prevalent threat as it directly monetizes attacks by holding data or systems hostage for ransom. Its high profitability, coupled with the increasing availability of ransomware-as-a-service platforms, allows even less skilled attackers to launch campaigns."

Changing insurance landscape due to ransomware

Cyber insurance is helping businesses to address such threats by offering services such as ransom negotiation, ransom reimbursement, and incident response. Such support, however, comes with a price. The years 2020 and 2021 witnessed a surge in insurance premiums. The Black Hat USA conference, scheduled in Las Vegas, will discuss how ransomware has changed businesses’ partnerships with insurers. Ransomware impacts an organization’s business model.

At the start of the 21st century, insurance firms required companies to buy a security audit to get a 25% policy discount. Insurance back then used to be a hands-on approach. The 2000s were followed by the data breach era; however, breaches were less common and frequent, targeting the hospitality and retail sectors. 

This caused insurers to stop checking for in-depth security audits, and they began using questionnaires to measure risk. In 2019, the ransomware wave happened, and insurers started paying out more claims than they were accepting. It was a sign that the business model was inadequate.

Questionnaires tend to be tricky for businesses to fill out. For instance, multifactor authentication (MFA) can be a complicated question to answer. Besides questionnaires, insurers have started using scans. 

Incentives to promote security measures

Threats have risen, but so have assessments, coverage incentives like vanishing retention mean that if policy users follow security instructions, retention disappears. Safety awareness training and patching vulnerabilities are other measures that can help in cost reductions. Scanning assessment can help in premium pricing, as it is lower currently. 

Read more »
Subscribe to: Posts (Atom)