Survivors Unveil the Dark Reality of Cyber Slavery

Introduction

With the ever-growing technology where cyber-crimes are increasing, a new cyber-attack is on the rise, but it’s not in your inbox or your computer- it's targeting your phone, especially your smartphone. Cybercriminals are expanding their reach in India, with a new text-messaging fraud targeting individuals. The Indian Computer Emergency Response Team (CERT-In) has warned against "smishing," or SMS phishing. 

Understanding Smishing

Smishing is a combination of the terms "SMS" and "phishing." It entails sending false text messages that appear to be from reputable sources such as banks, government organizations, or well-known companies. These communications frequently generate a feeling of urgency in their readers, prompting them to click on harmful links, expose personal information, or conduct financial transactions.

When hackers "phish," they send out phony emails in the hopes of tricking the receiver into clicking on a dangerous link. Smishing is just the use of text messaging rather than email. In essence, these hackers are out to steal your personal information to commit fraud or other cybercrimes. This generally entails stealing money – usually your own, but occasionally also the money of your firm.

The cybercriminals typically use these tactics to lure victims and steal the information.

Malware- The cyber crooks send the smishing URL link that might tick you into downloading malicious software on your phone itself. This SMS malware may appear as legitimate software, deceiving you into putting in sensitive information and transmitting it to crooks.

Malicious website- The URL in the smishing message may direct you to a bogus website that seeks sensitive personal information. Cybercriminals employ custom-made rogue sites meant to seem like legitimate ones, making it simpler to steal your information.

Smishing text messages often appear to be from your bank, asking you to share personal sensitive information, ATM numbers, or account details. Mobile device cybercrime is increasing, as is mobile device usage. Aside from the fact that texting is the most prevalent usage of cell phones, a few additional aspects make this an especially pernicious security issue. Let's go over how smishing attacks operate.

Modus Operandi

The cyber crooks commit the fraud via SMS. As attackers assume an identity that might be of someone trusted, Smishing attackers can use social engineering techniques to sway a victim's decision-making. Three things are causing this deception:

• Trust- Cyber crooks target individuals, by posing to someone from a legitimate individual and organization, this naturally lowers a person’s defense against threats. 

• Context- Using a circumstance that might be relevant to targets helps an attacker to create an effective disguise. The message feels personalized, which helps it overcome any assumption that it is spam.
• Emotion- The nature of the SMS is critical; it makes the victim think that is urgent and requires rapid action. Using these tactics, attackers craft communications that compel the receiver to act.
• Typically, attackers want the victim to click on a URL link within the text message, which takes them to a phishing tool that asks them for sensitive information. This phishing tool is frequently in the form of a website or app that also assumes a phony identity.

How does Smishing Spread?

As we have revealed earlier smishing attacks are delivered through both traditional texts. However, SMS phishing attacks primarily appear to be from known sources People are less careful while they are on their phones. Many people believe that their cell phones are more secure than their desktops. However, smartphone security has limits and cannot always guard against smishing directly.

Considering the fact phones are the target While Android smartphones dominate the market and are a perfect target for malware text messages, iOS devices are as vulnerable. Although Apple's iOS mobile technology has a high reputation for security, no mobile operating system can protect you from phishing-style assaults on its own. A false feeling of security, regardless of platform, might leave users especially exposed.

Kinds of smishing attacks

Some common types of smishing attacks that occurred are;

•  COVID-19 Smishing: The Better Business Bureau observed an increase in reports of US government impersonators sending text messages requesting consumers to take an obligatory COVID-19 test via a connected website in April 2020. The concept of these smishing assaults may readily develop, as feeding on pandemic concerns is a successful technique of victimizing the public.

• Gift Smishing: Give away, shopping rewards, or any number of other free offers, this kind of smishing includes free services or products, from a reputable or other company. attackers plan in such a way that the offer is for a limited time or is an exclusive offer and the offers are so lucrative that one gets excited and falls into the trap. 

CERT Guidelines

 CERT-In shared some steps to avoid falling victim to smishing. 

• Never click on any suspicious link in SMS/social media charts or posts.
• Use online resources to validate shortened URLs.
• Always check the link before clicking.
• Use updated antivirus and antimalware tools.
• If you receive any suspicious message pretending to be from a bank or institution, immediately contact the bank or institution.
• Use a separate email account for personal online transactions.
• Enforce multi-factor authentication (MFA) for emails and bank accounts.
• Keep your operating system and software updated with the latest patches.

Conclusion

Smishing uses fraudulent mobile text messages to trick people into downloading malware, sharing sensitive data, or paying cybercriminals money. With the latest technological developments, it has become really important to stay vigilant in the digital era not only protecting your computers but safeguarding the devices that fit in the palm of your hand, CERT warning plays a vital role in this. Awareness and best practices play a pivotal role in safeguarding yourself from evolving threats.

Reference

• https://www.ndtv.com/india-news/government-warns-of-smishing-attacks-heres-how-to-stay-safe-4709458
• https://zeenews.india.com/technology/govt-warns-citizens-about-smishing-scam-how-to-protect-against-this-online-threat-2654285.html
• https://www.the420.in/protect-against-smishing-scams-cert-in-advice-online-safety/

Executive Summary:

On 20th May, 2024, Iranian President Ebrahim Raisi and several others died in a helicopter crash that occurred northwest of Iran. The images circulated on social media claiming to show the crash site, are found to be false. CyberPeace Research Team’s investigation revealed that these images show the wreckage of a training plane crash in Iran's Mazandaran province in 2019 or 2020. Reverse image searches and confirmations from Tehran-based Rokna Press and Ten News verified that the viral images originated from an incident involving a police force's two-seater training plane, not the recent helicopter crash.

Claims:

The images circulating on social media claim to show the site of Iranian President Ebrahim Raisi's helicopter crash.

‍

Fact Check:

After receiving the posts, we reverse-searched each of the images and found a link to the 2020 Air Crash incident, except for the blue plane that can be seen in the viral image. We found a website where they uploaded the viral plane crash images on April 22, 2020.

According to the website, a police training plane crashed in the forests of Mazandaran, Swan Motel. We also found the images on another Iran News media outlet named, ‘Ten News’.

‍

The Photos uploaded on to this website were posted in May 2019. The news reads, “A training plane that was flying from Bisheh Kolah to Tehran. The wreckage of the plane was found near Salman Shahr in the area of ​​Qila Kala Abbas Abad.”

Hence, we concluded that the recent viral photos are not of Iranian President Ebrahim Raisi's Chopper Crash, It’s false and Misleading.

Conclusion:

The images being shared on social media as evidence of the helicopter crash involving Iranian President Ebrahim Raisi are incorrectly shown. They actually show the aftermath of a training plane crash that occurred in Mazandaran province in 2019 or 2020 which is uncertain. This has been confirmed through reverse image searches that traced the images back to their original publication by Rokna Press and Ten News. Consequently, the claim that these images are from the site of President Ebrahim Raisi's helicopter crash is false and Misleading.

• Claim:  Viral images of Iranian President Raisi's fatal chopper crash.
• Claimed on: X (Formerly known as Twitter), YouTube, Instagram
• Fact Check: Fake & Misleading

‍

‍

Introduction
‍

Agentic AI systems are autonomous systems that can plan, make decisions, and take actions by interacting with external tools and environments. But they shift the nature of risk by blurring the lines among input, decision, and execution. A conventional model generates an output and stops. An agent takes input, makes plans, invokes tools, updates its state and repeats the cycle. This creates a system where decisions are continuously revised through interaction with external tools and environments, rather than being fixed at the point of input.

This means the attack surface expands in size and becomes more dynamic. Instead of remaining confined to components as in traditional computational systems, they spread in layers and can continue to grow through time. To understand this shift, the system can be analysed through functional layers such as inputs, memory, reasoning, and execution, while recognising that risk does not remain isolated within these layers but emerges through their interaction.

‍

‍

Agentic AI Attack Surface

‍

A layered view of how risks emerge across input, memory, reasoning, execution, and system integration, including feedback loops and cross-system dependencies that amplify vulnerabilities.
‍

Input Layer: Where Untrusted Data Becomes Control
‍

The entry point of an agent is no longer one prompt. The documents, APIs, files, system logs and the outputs of other agents can now be considered input. This diversity is significant due to the fact that every source of input carries its own trust assumptions, and in the majority of cases, they are weak.

The most obvious threat is prompt injection, where inputs are treated as instructions rather than data. Since inputs are treated as instructions, a virus, a malicious webpage, or a document can contain instructions that override system goals without necessarily being detected as something harmful.

Indirect prompt injection extends this risk beyond direct user interaction. Instead of targeting the interface, attackers compromise the retrieval process by embedding malicious instructions within external data sources. When the agent retrieves and processes the data, it treats the embedded content as legitimate input. As a result, the attack is executed through normal reasoning processes, allowing the system to act on untrusted data without recognising the manipulation.

Data poisoning also occurs at runtime. In contrast to classical poisoning (where training data is manipulated), runtime poisoning distorts the agent’s perception of its environment as it runs. This can change decisions without causing apparent failures.

Obfuscation introduces another indirect attacker vector. Encoded instructions or complicated forms may bypass human review but remain readable to the model. This creates asymmetry whereby the system knows more about the attack than those operating it. Once compromised at this layer, the agent implements compromised instructions which affect downstream operations.
‍

Context and Memory: Persistence of Influence
‍

Agentic systems depend on memory to operate efficiently. They often retain context across sessions and frequently store information between sessions.

This introduces a different type of risk: persistence. Through memory poisoning, attackers can insert false or adversarial information into sorted context, which then influences future decisions. Unlike prompt injection, which is often limited to a single interaction, this effect carries forward. Over time, the agent begins to operate on a distorted internal state, shaping decisions in ways that may not be immediately visible.

Another issue is cross-session leakage. Information in a particular context may be replayed in a different context when memory is being shared or there is insufficient memory separation. This is specifically dangerous in those systems that combine retrieval and long-term storage. The context management in itself becomes a weakness. Agents are required to make decisions on what to retain and what to discard. This is susceptible to attackers who can flood the context or manipulate what is still visible and indirectly affect reasoning.

The underlying problem is structural. Memory turns data into a state. Once state is corrupted, the system cannot easily distinguish valid knowledge from adversarial influence.

The issue is structural. Memory converts temporary data into a persistent state. Once this state is weakened, the system cannot reliably separate valid information from adversarial influence, making recovery significantly more difficult.
‍

Reasoning and Planning: Manipulating Intent Without Breaking Logic
‍

The reasoning layer is where agentic AI stands apart from traditional systems. The model no longer reacts to inputs alone. It actively breaks down objectives, analyses alternatives, and ranks actions.

At the reasoning stage, the nature of risk shifts. The concern is no longer limited to injecting instructions, but to influencing how decisions are made. One example is goal manipulation, where the agent subtly reinterprets its objective and produces outcomes that are technically correct but strategically harmful. Reasoning hijacking operates within intermediate steps, altering how constraints are evaluated or how trade-offs are prioritised. The system may remain internally consistent, which makes such deviations difficult to detect.

Tool selection becomes a critical control point. Agents decide which tools to use and when, so influencing these choices can redirect execution without directly accessing the tools themselves. Hallucinations also take on a different role here. In static systems, they remain errors. In agentic systems, they can trigger actions. A perceived need or incorrect judgement can translate into real-world consequences.

This layer introduces probabilistic failure. The system is not fully weakened, but it is nudged towards decisions that appear reasonable yet are incorrect. The risk lies in how those decisions are justified.

‍
‍Tool and Execution: When Decisions Gain Reach
‍

Once an agent begins interacting with tools, its behaviour extends beyond the model into external systems. APIs, databases, and services become part of the execution path.

One key risk is the use of unauthorised tools. When agents operate with broad permissions, any manipulation of the upstream can be converted into real-world actions. This makes access control a central security concern. Command injection also takes a different form here. The agent generates commands based on its reasoning, so if that reasoning is compromised, the resulting actions may still appear valid despite being harmful.

External tool outputs introduce another risk. If these systems return corrupted or misleading data, the agent may accept it without verification and incorporate it into its decisions. It is also becoming increasingly reliant on third-part tools and plugins adds to this exposure. If these components are compromised, they can affect behaviour without directly attacking the core system, creating a supply-side risk.

At this stage, the agent effectively operates as an insider. It holds legitimate credentials and interacts with systems in expected ways, making misuse harder to identify.
‍

Application and Integration: System-Level Exposure
‍

Agentic systems rarely operate in isolation. They are embedded in larger environments, interacting with identity systems, business logic, and operational workflows.

Access control becomes a major vulnerability. Agents tend to operate across multiple systems with various permission models, creating irregularities that can be exploited. Risks also arise from identity and delegation. In case an agent is operating on behalf of a user, then any vulnerabilities in authentication or session management can allow attackers to assume that authority.

Workflow execution amplifies these risks. Agents can initiate multi-step processes such as transactions, updates, or approvals. Manipulating a single step can change the result of the entire workflow. As integrations increase, so do the number of interaction points, making cumulative risk harder to track.

At this layer, failures are not isolated. They propagate into business operations, making consequences harder to contain.
‍

Output and Action: Where Failures Become Visible
‍

The output layer is where failures become visible, though they rarely originate there.

Data leakage has been a key concern. Agents may disclose information they are allowed to access, especially when tasks boundaries are not clearly defined. Misinformation and unsafe outputs are also important, particularly when outputs directly influence actions or decisions.

Generated code and commands introduce execution risk. If outputs are used without validation, errors or manipulations can have system-level effects. The shift towards autonomous action increases this risk, as small upstream deviations can lead to significant consequences without human intervention. This layer reflects symptoms rather than root causes. Addressing it alone does not reduce the underlying risk.
‍

Beyond Layers: The Missing Dimension
‍

A layered view helps, but it does not capture the full picture. Agentic systems are defined by continuous interaction across layers.

The key missing dimension is the runtime loop. Inputs shape reasoning, reasoning drives action, and actions feed back into both reasoning and memory. These cycles create feedback loops, where small manipulations may escalate over time. This also reduces observability. With multiple interacting components, it becomes difficult to trace cause and effect or identify where failures originate.

Supply chain dependencies add another layer of risk. Models, datasets, APIs, and plugins each introduce their own points of failure. A compromise at any of these points can propagate across the system. The attack surface also includes governance. Weak supervision, unclear responsibility, or excessive autonomy increase overall risk. Human control is not external to the system; it is part of its security.
‍

Conclusion: Structuring the Attack Surface
‍

Agentic AI expands the attack surface beyond traditional systems. It is both recursive and stateful. Risk does not just accumulate across layers; it moves and changes as the system operates.

Any useful representation must go beyond a linear stack. It should capture feedback loops, persistent state, and cross-layer dependencies that characterise the way these systems actually behave. The system is not a pipeline but a cycle. That is where both its capability and its risk emerge.

‍