#FactCheck-Viral Video Falsely Claims ‘Civil War-Like Situation’ in Lucknow; Actually Shows Slum Fire

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.

‍

Executive Summary:

‍
A video is being shared on social media with the claim that a man cut off the electricity supply to his girlfriend’s village after getting angry with her. A research by CyberPeace’s research wing found that the video is actually from Baksa district in Assam, where a lineman named Anwar was repairing an electric line.

‍

Claim:

‍

The viral video is 10 seconds long and shows a man climbing an electricity pole and cutting an electric wire. The video is being shared on social media with the claim that the man cut off the electricity supply to his girlfriend’s entire village after her phone line was busy.

https://www.instagram.com/reels/Dcamn0aGBJ-/ 

‍

‍

Fact Check:

‍

To investigate the viral claim that a man cut off the electricity supply to his girlfriend’s village after getting angry with her, we conducted a reverse image search using keyframes from the video. During the search, we found the original video uploaded on the YouTube account named ‘Technical Work’ on July 17, 2025. According to the video description, the video is related to electrical work.

https://www.youtube.com/shorts/iHtG5xwHtKI 

‍

‍

Additionally, we found several videos related to electrical work uploaded on the same YouTube account. We also found a Facebook page associated with this YouTube account. 

‍

‍

Conclusion

‍

The evidence gathered during our research makes it clear that the viral claim that a man cut off the electricity supply to his girlfriend’s village after getting angry with her is completely false. In reality, the video shows electricity workers repairing a power line in Assam.

‍

Artificial Intelligence (AI) provides a varied range of services and continues to catch intrigue and experimentation. It has altered how we create and consume content. Specific prompts can now be used to create desired images enhancing experiences of storytelling and even education. However, as this content can influence public perception, its potential to cause misinformation must be noted as well. The realistic nature of the images can make it hard to discern as artificially generated by the untrained eye. As AI operates by analysing the data it was trained on previously to deliver, the lack of contextual knowledge and human biases (while framing prompts) also come into play. The stakes are higher whilst dabbling with subjects such as history, as there is a fine line between the creation of content with the intent of mere entertainment and the spread of misinformation owing to biases and lack of veracity left unchecked. AI-generated images enhance storytelling but can also spread misinformation, especially in historical contexts. For instance, an AI-generated image of London during the Black Death might include inaccurate details, misleading viewers about the past.

The Rise of AI-Generated Historical Images as Entertainment

Recently, generated images and videos of various historical instances along with the point of view of the people present have been floating all over the internet. Some of them include the streets of London during the Black Death in the 1300s in England, the eruption of Mount Vesuvius at Pompeii etc. Hogne and Dan, two creators who operate accounts named POV Lab and Time Traveller POV on TikTok state that they create such videos as they feel that seeing the past through a first-person perspective is an interesting way to bring history back to life while highlighting the cool parts, helping the audience learn something new. Mostly sensationalised for visual impact and storytelling, such content has been called out by historians for inconsistencies with respect to details particular of the time. Presently, artists admit to their creations being inaccurate, reasoning them to be more of an artistic interpretation than fact-checked documentaries. 

It is important to note that AI models may inaccurately depict objects (issues with lateral inversion), people(anatomical implausibilities), or scenes due to "present-ist" bias. As noted by Lauren Tilton, an associate professor of digital humanities at the University of Richmond, many AI models primarily rely on data from the last 15 years, making them prone to modern-day distortions especially when analysing and creating historical content. The idea is to spark interest rather than replace genuine historical facts while it is assumed that engagement with these images and videos is partly a product of the fascination with upcoming AI tools. Apart from this, there are also chatbots like Hello History and Charater.ai which enable simulations of interacting with historical figures that have piqued curiosity.

Although it makes for an interesting perspective, one cannot ignore that our inherent biases play a role in how we perceive the information presented. Dangerous consequences include feeding into conspiracy theories and the erasure of facts as information is geared particularly toward garnering attention and providing entertainment. Furthermore, exposure of such content to an impressionable audience with a lesser attention span increases the gravity of the matter. In such cases, information regarding the sources used for creation becomes an important factor.

Acknowledging the risks posed by AI-generated images and their susceptibility to create misinformation, the Government of Spain has taken a step in regulating the AI content created. It has passed a bill (for regulating AI-Generated content) that mandates the labelling of AI-generated images and failure to do so would warrant massive fines (up to $38 million or 7% of turnover on companies). The idea is to ensure that content creators label their content which would help to spot images that are artificially created from those that are not.

The Way Forward: Navigating AI and Misinformation

While AI-generated images make for exciting possibilities for storytelling and enabling intrigue, their potential to spread misinformation should not be overlooked. To address these challenges, certain measures should be encouraged.

1. Media Literacy and Awareness – In this day and age critical thinking and media literacy among consumers of content is imperative. Awareness, understanding, and access to tools that aid in detecting AI-generated content can prove to be helpful.
2. AI Transparency and Labeling – Implementing regulations similar to Spain’s bill on labelling content could be a guiding crutch for people who have yet to learn to tell apart AI-generated content from others.
3. Ethical AI Development – AI developers must prioritize ethical considerations in training using diverse and historically accurate datasets and sources which would minimise biases.

As AI continues to evolve, balancing innovation with responsibility is essential. By taking proactive measures in the early stages, we can harness AI's potential while safeguarding the integrity and trust of the sources while generating images.

References:

• https://www.npr.org/2023/06/07/1180768459/how-to-identify-ai-generated-deepfake-images
• https://www.nbcnews.com/tech/tech-news/ai-image-misinformation-surged-google-research-finds-rcna154333
• https://www.bbc.com/news/articles/cy87076pdw3o
• https://newskarnataka.com/technology/government-releases-guide-to-help-citizens-identify-ai-generated-images/21052024/
• https://www.technologyreview.com/2023/04/11/1071104/ai-helping-historians-analyze-past/
• https://www.psypost.org/ai-models-struggle-with-expert-level-global-history-knowledge/
• https://www.youtube.com/watch?v=M65IYIWlqes&t=2597s
• https://www.vice.com/en/article/people-are-creating-records-of-fake-historical-events-using-ai/?utm_source=chatgpt.com
• https://www.reuters.com/technology/artificial-intelligence/spain-impose-massive-fines-not-labelling-ai-generated-content-2025-03-11/?utm_source=chatgpt.com
• https://www.theguardian.com/film/2024/sep/13/documentary-ai-guidelines?utm_source=chatgpt.com

‍