#FactCheck: AI-generated image falsely linked to Japan mall blast after 7.1-magnitude earthquake

Introduction

Intricate and winding are the passageways of the modern digital age, a place where the reverberations of truth effortlessly blend, yet hauntingly contrast, with the echoes of falsehood. Within this complex realm, the World Economic Forum (WEF) has illuminated the darkened corners with its powerful spotlight, revealing the festering, insidious network of misinformation and disinformation that snakes through the virtual and physical worlds alike. Gravely identified by the “WEF's Global Risks Report 2024” as the most formidable and immediate threats to our collective well-being, this malignant duo—misinformation and disinformation. 

The report published with the solemn tone suitable for the prelude to such a grand international gathering as the Annual Summit in Davos, the report presents a vivid tableau of our shared global landscape—one that is dominated by the treacherous pitfalls of deceits and unverified claims. These perils, if unrecognised and unchecked by societal checks and balances, possess the force to rip apart the intricate tapestry of our liberal institutions, shaking the pillars of democracies and endangering the vulnerable fabric of social cohesion.

Election Mania 

As we find ourselves perched on the edge of a future, one where the voices of nearly three billion human beings make their mark on the annals of history—within the varied electoral processes of nations such as Bangladesh, India, Indonesia, Mexico, Pakistan, the United Kingdom, and the United States. However, the spectre of misinformation can potentially corrode the integrity of the governing entities that will emerge from these democratic processes. The warning issued by the WEF is unambiguous: we are flirting with the possibility of disorder and turmoil, where the unchecked dispersion of fabrications and lies could kindle flames of unrest, manifesting in violent protests, hate-driven crimes, civil unrest, and the scourge of terrorism.

Derived from the collective wisdom of over 1,400 experts in global risk, esteemed policymakers, and industry leaders, the report crafts a sobering depiction of our world's journey. It paints an ominous future that increasingly endows governments with formidable power—to brandish the weapon of censorship, to unilaterally declare what is deemed 'true' and what ought to be obscured or eliminated in the virtual world of sharing information. This trend signals a looming potential for wider and more comprehensive repression, hindering the freedoms traditionally associated with the Internet, journalism, and unhindered access to a panoply of information sources—vital fora for the exchange of ideas and knowledge in a myriad of countries across the globe.

Prominence of AI 

When the gaze of the report extends further over a decade-long horizon, the prominence of environmental challenges such as the erosion of biodiversity and alarming shifts in the Earth's life-support systems ascend to the pinnacle of concern. Yet, trailing closely, the digital risks continue to pulsate—perpetuated by the distortions of misinformation, the echoing falsities of disinformation, and the unpredictable repercussions stemming from the utilization and, at times, the malevolent deployment of artificial intelligence (AI). These ethereal digital entities, far from being illusory shades, are the precursors of a disintegrating world order, a stage on which regional powers move to assert and maintain their influence, instituting their own unique standards and norms.

The prophecies set forth by the WEF should not be dismissed as mere academic conjecture; they are instead a trumpet's urgent call to mobilize. With a startling 30 percent of surveyed global experts bracing for the prospect of international calamities within the mere span of the coming two years, and an even more significant portion—nearly two-thirds—envisaging such crises within the forthcoming decade, it is unmistakable that the time to confront and tackle these looming risks is now. The clarion is sounding, and the message is clear: inaction is no longer an available luxury.

Maldives and India Row

To pluck precise examples from the boundless field of misinformation, we might observe the Lakshadweep-Malé incident wherein an ordinary boat accident off the coast of Kerala was grotesquely transformed into a vessel for the far-reaching tendrils of fabricated narratives, erroneously implicating Lakshadweep in the spectacle. Similarly, the tension-laden India-Maldives diplomatic exchange becomes a harrowing testament to how strained international relations may become fertile ground for the rampant spread of misleading content. The suspension of Maldivian deputy ministers following offensive remarks, the immediate tumult that followed on social media, and the explosive proliferation of counterfeit news targeting both nations paint a stark and intricate picture of how intertwined are the threads of politics, the digital platforms of social media, and the virulent propagation of falsehoods.

Yet, these are mere fragments within the extensive and elaborate weave of misinformation that threatens to enmesh our globe. As we venture forth into this dangerous and murky topography, it becomes our collective responsibility to maintain a sense of heightened vigilance, to consistently question and verify the sources and content of the information that assails us from all directions, and to cultivate an enduring culture anchored in critical thinking and discernment. The stakes are colossal—for it is not merely truth itself that we defend, but rather the underlying tenets of our societies and the sanctity of our cherished democratic institutions.

Conclusion 

In this fraught era, marked indelibly by uncertainty and perched precariously on the cusp of numerous pivotal electoral ventures, let us refuse the role of passive bystanders to unraveling our collective reality. We must embrace our role as active participants in the relentless pursuit of truth, fortified with the stark awareness that our entwined futures rest precariously on our willingness and ability to distinguish the veritable from the spurious within the perilous lattice of falsehoods of misinformation. We must continually remind ourselves that, in the quest for a stable and just global order, the unerring discernment of fact from fiction becomes not only an act of intellectual integrity but a deed of civic and moral imperative.

References 

• https://www.businessinsider.in/politics/world/election-fuelled-misinformation-is-serious-global-risk-in-2024-says-wef/articleshow/106727033.cms
• https://www.deccanchronicle.com/nation/current-affairs/100124/misinformation-tops-global-risks-2024.html
• https://www.msn.com/en-in/news/India/fact-check-in-lakshadweep-male-row-kerala-boat-accident-becomes-vessel-for-fake-news/ar-AA1mOJqY
• https://www.boomlive.in/news/india-maldives-muizzu-pm-modi-lakshadweep-fact-check-24085
• https://www.weforum.org/press/2024/01/global-risks-report-2024-press-release/

Executive Summary:

New Linux malware has been discovered by a cybersecurity firm Volexity, and this new strain of malware is being referred to as DISGOMOJI. A Pakistan-based threat actor alias ‘UTA0137’ has been identified as having espionage aims, with its primary focus on Indian government entities.  Like other common forms of backdoors and botnets involved in different types of cyberattacks, DISGOMOJI, the malware  allows the use of commands to capture screenshots, search for files to steal, spread additional payloads, and transfer files. DISGOMOJI  uses Discord (messaging service)  for Command & Control (C2)  and uses emojis  for C2 communication. This malware targets Linux operating systems.

The DISCOMOJI Malware:  

• The DISGOMOJI malware opens a specific channel in a Discord server and every new channel corresponds to a new victim. This means that the attacker can communicate with the victim one at a time.

• This particular malware connects with the attacker-controlled Discord server using Emoji, a form of relay protocol. The attacker provides unique emojis as instructions, and the malware uses emojis as a feedback to the subsequent command status.

• For instance, the ‘camera with flash’ emoji is used to screenshots the device of the victim or to steal, the ‘fox’ emoji cracks all Firefox profiles, and the ‘skull’ emoji kills the malware process.

• This C2 communication is done using emojis to ensure messaging between infected contacts, and it is almost impossible for Discord to shut down the malware as it can always change the account details of Discord it is using once the maliciou server is blocked.

• The malware also has capabilities aside from the emoji-based C2 such as network probing, tunneling, and data theft that are needed to help the UTA0137 threat actor in achieving its espionage goals.

Specific emojis used for different commands by UTA0137:

• Camera with Flash (📸): Captures a picture of the target device’s screen as per the victim’s directions.

• Backhand Index Pointing Down (👇): Extracts files from the targeted device and sends them to the command channel in the form of attachments.

• Backhand Index Pointing Right (👉): This process involves sending a file found on the victim’s device to another web-hosted file storage service known as Oshi or oshi[. ]at.

• Backhand Index Pointing Left (👈): Sends a file from the victim’s device to transfer[. ]sh, which is an online service for sharing files on the Internet.

• Fire (🔥): Finds and transmits all files with certain extensions that exist on the victim’s device, such as *. txt, *. doc, *. xls, *. pdf, *. ppt, *. rtf, *. log, *. cfg, *. dat, *. db, *. mdb, *. odb, *. sql, *. json, *. xml, *. php, *. asp, *. pl, *. sh, *. py, *. ino, *. cpp, *. java,

• Fox (🦊): This works by compressing all Firefox related profiles in the affected device. 

• Skull (💀): Kills the malware process in windows using ‘os. Exit()’

• Man Running (🏃‍♂️):  Execute a command on a victim’s device. This command receives an argument, which is the command to execute.

• Index Pointing up (👆) : Upload a file to the victim's device. The file to upload is attached along with this emoji

Analysis:   

The analysis was carried out for one of the indicator of compromised SHA-256 hash file-  C981aa1f05adf030bacffc0e279cf9dc93cef877f7bce33ee27e9296363cf002.

It is found that most of the vendors have marked the file as trojan in virustotal and the graph explains the malicious nature of the contacted domains and IPs.

‍

Discord & C2 Communication  for UTA0137:

• Stealthiness: Discord is a well-known messaging platform used for different purposes, which means that sending any messages or files on the server should not attract suspicion. Such stealthiness makes it possible for UTA0137 to remain dormant for greater periods before launching an attack.

• Customization: UTA0137 connected to Discord is able to create specific channels for distinct victims on the server. Such a framework allows the attackers to communicate with each of the victims individually to make a process more accurate and efficient.

• Emoji-based protocol: For C2 communication, emojis really complicates the attempt that Discord might make to interfere with the operations of the malware. In case the malicious server gets banned, malware could easily be recovered, especially by using the Discord credentials from the C2 server.

• Persistence: The malware, as stated above, has the ability to perpetually exist to hack the system and withstand rebooting of systems so that the virus can continue to operate without being detected by the owner of the hacked system.

• Advanced capabilities: Other features of DISGOMOJI are the Network Map using Nmap scanner, network tunneling through Chisel and Ligolo and Data Exfiltration by File Sharing services. These capabilities thus help in aiding the espionage goals of UTA0137.

• Social engineering: The virus and the trojan can show the pop-up windows and prompt messages, for example the fake update for firefox and similar applications, where the user can be tricked into inputting the password.

• Dynamic credential fetching: The malware does not write the hardcoded values of the credentials in order to connect it to the discord server. This also inconveniences analysts as they are unable to easily locate the position of the C2 server.

• Bogus informational and error messages: They never show any real information or errors because they do not want one to decipher the malicious behavior easily.

Recommendations to mitigate the risk of UTA0137:

• Regularly Update Software and Firmware: It is essential to regularly update all the application software and firmware of different devices, particularly, routers, to prevent hackers from exploiting the discovered and disclosed flaws. This includes fixing bugs such as CVE-2024-3080 and CVE-2024-3912 on ASUS routers, which basically entails solving a set of problems.

• Implement Multi-Factor Authentication: There are statistics that show how often user accounts are attacked, it is important to incorporate multi-factor authentication to further secure the accounts.

• Deploy Advanced Malware Protection: Provide robust guard that will help the user recognize and prevent the execution of the DISGOMOJI malware and similar threats.

• Enhance Network Segmentation: Utilize stringent network isolation mechanisms that seek to compartmentalize the key systems and data from the rest of the network in order to minimize the attack exposure.

• Monitor Network Activity: Scanning Network hour to hour for identifying and handling the security breach and the tools such as Nmap, Chisel, Ligolo etc can be used.

• Utilize Threat Intelligence: To leverage advanced threats intelligence which will help you acquire knowledge on previous threats and vulnerabilities and take informed actions.

• Secure Communication Channels: Mitigate the problem of the leakage of developers’ credentials and ways of engaging with the discord through loss of contact to prevent abusing attacks or gaining control over Discord as an attack vector.

• Enforce Access Control: Regularly review and update the user authentication processes by adopting stricter access control measures that will allow only the right personnel to access the right systems and information.

• Conduct Regular Security Audits: It is important to engage in security audits periodically in an effort to check some of the weaknesses present within the network or systems.

• Implement Incident Response Plan: Conduct a risk assessment, based on that design and establish an efficient incident response kit that helps in the early identification, isolation, and management of security breaches.

• Educate Users: Educate users on cybersecurity hygiene, opportunities to strengthen affinity with the University, and conduct retraining on threats like phishing and social engineering.

Conclusion:

The new threat actor named UTA0137 from Pakistan who was utilizing DISGOMOJI malware to attack Indian government institutions using embedded emojis with a command line through the Discord app was discovered by Volexity. It has the capability to exfiltrate and aims to steal the data of government entities. The UTA0137 was continuously improved over time to permanently communicate with victims. It underlines the necessity of having strong protection from viruses and hacker attacks, using secure passwords and unique codes every time, updating the software more often and having high-level anti-malware tools. Organizations can minimize advanced threats, the likes of DISGOMOJI and protect sensitive data by improving network segmentation, continuous monitoring of activities, and users’ awareness.

References:

https://otx.alienvault.com/pulse/66712446e23b1d14e4f293eb

https://thehackernews.com/2024/06/pakistani-hackers-use-disgomoji-malware.html?m=1

https://cybernews.com/news/hackers-using-emojis-to-command-malware/

https://www.blackhatethicalhacking.com/news/disgomoji-new-linux-malware-uses-emojis-for-command-execution/

https://www.volexity.com/blog/2024/06/13/disgomoji-malware-used-to-target-indian-government/

‍

Executive Summary
‍

A video is being widely shared on social media with the claim that it shows the recent Assam floods, where nurses are rescuing newborn babies after water entered a hospital. The caption accompanying the viral video claims that the babies are not drowning; instead, healthcare workers and rescue teams are carrying them to safety amid floodwaters inside a hospital. It further states that the video shows humanity and courage during a time of crisis. CyberPeace Research Wing’s research found the claim to be false. The research revealed that the viral video showing healthcare workers rescuing newborn babies is not real but was created using Artificial Intelligence (AI).

‍

Claim:

‍

An Instagram user named kamlakritika shared the video and claimed that it showed the Assam floods. The post stated that the babies were not drowning but were being rescued by healthcare workers and rescue teams from a flooded hospital. In the viral video, several nurses can be seen running through waterlogged areas while trying to protect incubators — containers used to keep newborn babies — from being submerged or swept away. According to the caption, a devastating flood had turned the entire area into a disaster zone. It claimed that incubators containing premature newborn babies were floating in floodwaters, while nurses were running behind them amid strong water currents to save the infants. The caption further claimed that several incubators were overturning and sinking with babies inside, putting hundreds of fragile lives at risk.

https://www.instagram.com/reel/DbXvJwgz72X/?utm_source=ig_web_copy_link&igsh=NTc4MTIwNjQ2YQ==

https://archive.ph/wip/niuGj

‍

‍

Factcheck

‍

To verify the claim, CyberPeace Research Wing conducted a Google search using relevant keywords. During the search, the video was found shared by several other Instagram accounts. During the research, we found a report published by News18 on August 1, 2026. Although the report was related to the Assam floods, it mentioned a different incident. According to the report, a viral video showed floodwaters surrounding houses, with water spread across the area. In the video, a rescue team member can be seen carefully carrying a newborn baby in his arms through the flooded area and handing the child over to another rescue worker near a boat so that the baby could be taken to safety.

https://hindi.news18.com/news/ajab-gajab/viral-crowd-200-people-gather-buy-scrap-bike-sold-5200-police-station-madhya-pradesh-local18-10707653.html

‍

‍

However, during our research, we could not find any information from credible news organisations or official sources confirming the viral video showing newborn babies being rescued in incubators. As part of further research, we closely examined the viral video and noticed several indications suggesting that it could be AI-generated. The video was then analysed using the AI detection tool FauxLens, which indicated that the video was approximately 75% AI-generated.

‍

‍

‍

To further verify the authenticity of the video, we also analysed it using another AI detection tool, Human Meter. The tool showed an 85% probability of the video being generated with the help of AI.

‍

‍

Conclusion:

‍

CyberPeace Research Wing’s research found that the viral video claiming to show nurses rescuing newborn babies during Assam floods is false. The video is not an actual incident but an AI-generated creation.

‍