#FactCheck-Old Accident Involving Delhi Police Head Constable Falsely Linked to CJP Protest
Executive Summary
A social media post is being widely circulated claiming that Delhi Police personnel Amit died during the Cockroach Janata Party (CJP) ‘Sansad Chalo’ protest held on July 20, 2026. The post attempts to link his death to violence that allegedly took place during the demonstration. CyberPeace Research Wing’s research found the claim to be false. The research revealed that the death of Head Constable Amit was not related to the CJP protest held on July 20. He died on July 15 after being hit by a speeding vehicle while on duty at Singhu Border. An old accident has been falsely linked to the recent protest through a misleading claim.
Claim:
A Facebook user shared a post on July 23, 2026, claiming that a police personnel was martyred during the alleged protest carried out by “student goons and their supporters from Congress and Kejriwal’s party” on July 20. The user further claimed that the protest was responsible for the death of a police personnel and questioned criticism over alleged police action against protesters.
The post’s link, archive link and screenshot are provided below:
https://www.facebook.com/photo?fbid=3562630610568369&set=a.100838436747621
https://ghostarchive.org/archive/m8QM8
Fact Check:
To verify the authenticity of the viral claim, the Desk first examined the official social media accounts of Delhi Police. During the research, the Desk found a post shared on Delhi Police’s official Instagram account on July 16, 2026, which contained the same image being circulated in the viral post.
https://www.instagram.com/p/Da19BQxEmQv/?igsh=cnN1bXZhaHg5OWRs
In the next stage of the research, the Desk found an NDTV report published on July 16, 2026. According to the report, Delhi Police had arrested two accused in connection with the death of Head Constable Amit. The arrested individuals were identified as Kuldeep alias Karan and Neeraj. The report stated that police had also seized the vehicle bearing registration number DL11CG2268, which was involved in the accident. An FIR was registered at Alipur Police Station and further research was underway.
Further research led the Desk to a related post on the Facebook page of Delhi Samachar TV. According to the post, the hit-and-run case involving Head Constable Amit, who was hit while performing duty at Singhu Border (NH-44), was solved by Delhi Police within hours. The police team arrested two accused and recovered the vehicle used in the incident.
The post’s link and screenshot are provided below:
https://www.facebook.com/reel/1006635078843860
Conclusion:
The research established that the death of Delhi Police Head Constable Amit was not connected to the CJP ‘Sansad Chalo’ protest held on July 20, 2026. Amit died on July 15 after being hit by a speeding vehicle while on duty at Singhu Border. The old accident is being falsely linked to the recent protest with a misleading claim.
Related Blogs
Executive Summary
A video is being widely shared on social media showing soldiers carrying coffins with full military honours. Users are claiming that the footage shows the bodies of American soldiers who were killed in the war with Iran being brought back to the United States.
However, research by the CyberPeacefound the viral claim to be misleading. Our research revealed that the video has no connection to the recent Iran-Israel conflict. The footage actually dates back to December 2025, when an Islamic State gunman in Syria killed two US soldiers and a US civilian.
Following that incident, the bodies of the victims were transported with military honours, and the ceremony was recorded in the viral video. The clip is now being circulated online with a false claim.
Claim
On March 1, 2026, an Instagram user shared the viral video claiming that it shows the bodies of American soldiers returning to the US after being killed in the war against Iran. The caption of the post reads: “Bodies of American soldiers martyred against Iran are returning to the United States. War always brings destruction, which we are now witnessing.”
The link to the post and its archived version can be seen below.
Fact check
To verify the claim, we extracted key frames from the viral video and performed a reverse image search using Google Lens. During the search, we found the full version of the video in a report published by BBC on December 18, 2025. This confirms that the footage predates the current developments.
According to the BBC report, US President Donald Trump attended a dignified transfer ceremony for two members of the US National Guard and a US civilian who were killed in Syria. The somber ceremony took place at Dover Air Force Base in Delaware, United States. The US Central Command (Centcom) stated that the two soldiers and a civilian interpreter were killed in an ambush carried out by an Islamic State (IS) gunman in Syria. The US Army identified the two soldiers as Sgt. Edgar Brian Torres-Tovar (25) of Des Moines and Sgt. William Nathaniel Howard (29) of Marshalltown. A US civilian interpreter, Ayad Mansoor Sakat, was also killed in the attack. Officials said that three other service members were injured during the attack, and the gunman was engaged and killed. Syria’s state media also reported that two Syrian security personnel were injured in the incident.
Further research led us to a report published on the News Nation YouTube channel on December 18, 2025, which also featured the same footage related to the incident.
Conclusion
Our research found that the viral video is not related to the recent Iran-Israel conflict. The footage dates back to December 2025, when two US soldiers and a US civilian were killed in an Islamic State attack in Syria. The video shows the dignified transfer of their remains and is now being shared on social media with a misleading claim.
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.
What is Juice Jacking?
We all use different devices during the day, but they converge to a common point when the battery runs out, the cables and adaptors we use to charge the devices are daily necessities for everyone. These cables and adaptors have access to the only port in the phones and hence are used for juice-jacking attacks. Juice jacking is when someone installs malware or spyware software in your device using an unknown charging port or cable.
How does juice jacking work?
We all use phones and gadgets, like I-phones, smartphones, Android devices: and smartwatches, to simplify our lives. But one thing common in it is the charging cables or USB ports, as the data and power supply pass through the same port/cable.
This is potentially a problem with devastating consequences. When your phone connects to another device, it pairs with it (ports/cables) and establishes a trusted relationship. That means the devices can exchange data. During the charging process, the USB cord opens a path into your device that a cybercriminal can exploit.
There is a default setting in the phones where data transfer is disabled, and the connections which provide the power are visible at the end. For example, in the latest models, when you plug your device into a new port or a computer, a question is pooped asking whether the device is trusted. The device owner cannot see what the USB port connects to in case of juice jacking. So, if you plug in your phone and someone checks on the other end, they may be able to transfer data between your device and theirs, thus leading to a data breach.
A leading airline was recently hacked into, which caused delayed flights across the country. When investigated, it was found that malware was planted in the system by using a USB port, which allowed the hackers access to critical data to launch their malware attack.
FBI’s Advisory
Federal Bureau of Investigation and other Interpol agencies have been very critical of cybercriminals. Inter-agency cooperation has improved the pace of investigation and chances of apprehending criminals. In a tweet by the FBI, the issue of Juice Jakcking was addressed, and public places like airports, railways stations, shopping malls etc., are pinpointed places where such attacks have been seen and reported. These places offer easy access to charging points for various devices, which are the main targets for bad actors. The FBI advises people not to use the charging points and cables at airports, railways stations and hotels and also lays emphasis upon the importance of carrying your own cable and charger.
Tips to protect yourself from juice jacking
There are a few simple and effective tips to keep your smart devices smart, such as –
- Avoid using public charging stations: The best way to protect yourself and your devices is to avoid public charging stations it’s always a good habit to charge your phones in your car, at home, and in offices when not in use.
- Using a wall outlet is a safer option: If it’s too urgent for you to use a public station, try to use wall outlets rather than poles because data can’t get easily transferred.
- Use other methods/modes of charging: If you are travelling, carrying a power bank is always safe, as it is easy to carry.
- Software security: – It’s always advised to update your phone’s software regularly. Once connected to the charging station, lock your device. This will prevent it from syncing or transferring data.
- Enable Airplane mode while charging: If you need to charge your phone from an unknown source in a public area, it is advisable to put the phone on airplane mode or switch it off to prevent anyone from gaining access to your device through any open network.
However, many mobile phones (including iPhones) turn on automatically when connected to power. As a result, your mileage may vary. This is an effective safeguard if your phone does not turn on automatically when connected to power.
Conclusion
As of present, juice-jacking attacks are less frequent. While not the most common type of attack today, the number of occurrences is expected to rise as smartphone gadget usage and penetration are rising across the globe. Our cyber safety and security are in our hands, and hence protecting them is our paramount digital duty. Always remember we see no harm in charging ports, but that doesn’t mean that the possibility of a threat can be ruled out completely. With the increased use of ports for charging, earphones, and data transfer, such crimes will continue and evolve with time. Thus, it is essential to counter these attacks by sharing knowledge and awareness of such crimes and reporting them to competent authorities to eradicate the menace of cybercriminals from our digital ecosystem.
