#FactCheck -AI-Edited Video Falsely Attributes Beauty Parlour Remark to PM Modi in Fuel Conservation Speech
Executive Summary
Amid ongoing tensions in West Asia, a video has been circulating on social media claiming that Prime Minister Narendra Modi urged women not to visit beauty parlours as part of an appeal related to fuel conservation. Users are widely sharing the clip as genuine. However, a fact-check by CyberPeace Research Wing has found the claim to be false. The research revealed that the original video, which discussed saving petrol, diesel and cooking gas, has been digitally edited using AI tools and shared with a misleading narrative.
Claim
An Instagram user shared the viral video on May 15, 2026, claiming that the Prime Minister advised women against visiting beauty parlours.
Fact Check
A keyword-based search on Google did not return any credible media reports supporting the viral claim. To verify the video, keyframes were extracted and run through reverse image search. The original footage was traced to the official YouTube channel of Prime Minister Narendra Modi, uploaded on May 10, 2026.
The video shows a rally held in Hyderabad, where the Prime Minister appealed to the public to save fuel by adopting work-from-home practices, promoting swadeshi goods, and refraining from purchasing gold for one year. However, nowhere in the full video is there any mention of advising women not to visit beauty parlours.
Further analysis also raised suspicion that the viral clip had been digitally altered using AI tools. The video was scanned using the HIVE Moderation AI detection tool, which indicated that it is approximately 88% likely to be AI-generated or manipulated.
Conclusion
The viral video is misleading. The original speech by Prime Minister Narendra Modi about conserving petrol, diesel and cooking gas has been edited using AI tools and circulated with a false claim targeting women visiting beauty parlours.
Related Blogs
Executive Summary
A photograph circulating widely on social media has been falsely linked to the reported hantavirus outbreak on a cruise ship in May 2026. The image is being shared with claims that it shows a cameraman filming evacuation operations without wearing personal protective equipment (PPE), suggesting that the global health scare was staged. However, the claim is misleading. The viral image has no connection to the hantavirus outbreak or any real evacuation related to the incident. research found that the photograph actually originates from a Spanish military exercise conducted a year earlier.
Claim
A post on X dated May 13, 2026, shared the image with the caption, “Don’t be fooled! The #Hantavirus HOAX,” alleging that the evacuations were staged events filmed from a movie set. The image shows several individuals in full personal protective equipment (PPE) carrying a patient on a biocontainment stretcher across a dock in front of a large vessel. A cameraman, dressed in plain clothes, is seen descending a ramp connected to the ship while apparently recording the scene.
Fact Check
The image went viral across X, Threads and Instagram in multiple languages, with users claiming the hantavirus outbreak was a “hoax” or that evacuation operations were staged. These posts emerged as authorities were monitoring passengers aboard the Dutch-flagged MV Hondius and others potentially exposed to the virus following reports of an outbreak during its Atlantic voyage.
Health officials have repeatedly stressed that the overall public health risk from the Andes strain of hantavirus remains low. It is a rare rodent-borne virus and the only variant known to transmit between humans.
Reverse image search traced the photograph to the official website of the Port of Almería in Spain, where it was published in a May 7, 2025 article documenting an international maritime exercise involving the Spanish Navy and the health ministry.
The exercise simulated the arrival of a vessel carrying suspected Ebola cases. Ebola is a rare but highly lethal disease that can cause fever, bleeding and organ failure.
Conclusion
The viral claim is false. The photograph has no connection to the reported hantavirus outbreak. It is from a Spanish military exercise conducted in 2025 and is being falsely shared out of context.
Artificial intelligence is revolutionizing industries such as healthcare to finance to influence the decisions that touch the lives of millions daily. However, there is a hidden danger associated with this power: unfair results of AI systems, reinforcement of social inequalities, and distrust of technology. One of the main causes of this issue is training data bias, which appears when the examples on which an AI model is trained are not representative or skewed. To deal with it successfully, this needs a combination of statistical methods, algorithmic design that is mindful of fairness, and robust governance over the AI lifecycle. This article discusses the origin of bias, the ways to reduce it, and the unique position of fairness-conscious algorithms.
Why Bias in Training Data Matters
The bias in AI occurs when the models mirror and reproduce the trends of inequality in the training data. When a dataset has a biased representation of a demographic group or includes historical biases, the model will be trained to make decisions in ways that will harm the group. This is a fact that has a practical implication: prejudiced AI may cause discrimination during the recruitment of employees, lending, and evaluation of criminal risks, as well as various other spheres of social life, thus compromising justice and equity. These problems are not only technical in nature but also require moral principles and a system of governance (E&ICTA).
Bias is not uniform. It may be based on the data itself, the algorithm design, or even the lack of diversity among developers. The bias in data occurs when data does not represent the real world. Algorithm bias may arise when design decisions inadvertently put one group at an unfair advantage over another. Both the interpretation of the model and data collection may be affected by human bias. (MDPI)
Statistical Principles for Reducing Training Data Bias
Statistical principles are at the core of bias mitigation and they redefine the data-model interaction. These approaches are focused on data preparation, training process adjustment, and model output corrections in such a way that the notion of fairness becomes a quantifiable goal.
Balancing Data Through Re-Sampling and Re-Weighting
Among the aforementioned methods, a fair representation of all the relevant groups in the dataset is one way. This can be achieved by oversampling underrepresented groups and undersampling overrepresented groups. Oversampling gives greater weight to minority examples, whereas re-weighting gives greater weight to under-represented data points in training. The methods minimize the tendency of models to fit to salient patterns and improve coverage among vulnerable groups. (GeeksforGeeks)
Feature Engineering and Data Transformation
The other statistical technique is to convert data characteristics in such a way that sensitive characteristics have a lesser impact on the results. In one example, fair representation learning adjusts the data representation to discourage bias during the untraining of the model. The disparate impact remover adjust technique performs the adjustment of features of the model in such a way that the impact of sensitive features is reduced during learning. (GeeksforGeeks)
Measuring Fairness With Metrics
Statistical fairness measures are used to measure the effectiveness of a model in groups.
Fairness-Aware Algorithms Explained
Fair algorithms do not simply detect bias. They incorporate fairness goals in model construction and run in three phases including pre-processing, in-processing, and post-processing.
Pre-Processing Techniques
Fairness-aware pre-processing deals with bias prior to the model consuming the information. This involves the following ways:
- Rebalancing training data through sampling and re-weighting training data to address sample imbalances.
- Data augmentation to generate examples of underrepresented groups.
- Feature transformation removes or downplays the impact of sensitive attributes prior to the commencement of training. (IJMRSET)
These methods can be used to guarantee that the model is trained on more balanced data and to reduce the chances of bias transfer between historical data.
In-Processing Techniques
The in-processing techniques alter the learning algorithm. These include:
- Fairness constraints that penalize the model for making biased predictions during training.
- Adversarial debiasing, where a second model is used to ensure that sensitive attributes are not predicted by the learned representations.
- Fair representation learning that modifies internal model representations in favor of
Post-Processing Techniques
Fairness may be enhanced after training by changing the model outputs. These strategies comprise:
- Threshold adjustments to various groups to meet conditions of fairness, like equalized odds.
- Calibration techniques such that the estimated probabilities are fair indicators of the actual probabilities in groups. (GeeksforGeeks)
Challenges
Mitigating bias is complex. The statistical bias minimization may at times come at the cost of the model accuracy, and there is a conflict between predictive performance and fairness. The definition of fairness itself is potentially a difficult task because various applications of fairness require various criteria, and various criteria can be conflicting. (MDPI)
Gaining varied and representative data is also a challenge that is experienced because of privacy issues, incomplete records, and a lack of resources. The auditing and reporting done on a continuous basis are needed so that mitigation processes are up to date, as models are continually updated. (E&ICTA)
Why Fairness-Aware Development Matters
The outcomes of the unfair treatment of some groups by AI systems are far-reaching. Discriminatory software in recruitment may support inequality in the workplace. Subjective credit rating may deprive deserving people of opportunities. Unbiased medical forecasts might result in the flawed allocation of medical resources. In both cases, prejudice contravenes the credibility and clouds the greater prospect of AI. (E&ICTA)
Algorithms that are fair and statistical mitigation plans provide a way to create not only powerful AI but also fair and trustworthy AI. They admit that the results of AI systems are social tools whose effects extend across society. Responsible development will necessitate sustained fairness quantification, model adjustment, and upholding human control.
Conclusion
AI bias is not a technical malfunction. It is a mirror of real-world disparities in data and exaggerated by models. Statistical rigor, wise algorithm design, and readiness to address the trade-offs between fairness and performance are required to reduce training data bias. Fairness-conscious algorithms (which can be implemented in pre-processing, in-processing, or post-processing) are useful in delivering more fair results. As AI is taking part in the most crucial decisions, it is necessary to consider fairness at the beginning to have a system that serves the population in a responsible and fair manner.
References
- Understanding Bias in Artificial Intelligence: Challenges, Impacts, and Mitigation Strategies: E&ICTA, IITK
- Bias and Fairness in Artificial Intelligence: Methods and Mitigation Strategies: JRPS Shodh Sagar
- Fairness and Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, and Mitigation Strategies: MDPI
- Ensuring Fairness in Machine Learning Algorithms: GeeksforGeeks
Bias and Fairness in Machine Learning Models: A Critical Examination of Ethical Implications: IJMRSET - Bias in AI Models: Origins, Impact, and Mitigation Strategies: Preprints
- Bias in Artificial Intelligence and Mitigation Strategies: TCS
- Survey on Machine Learning Biases and Mitigation Techniques: MDPI
Introduction
On May 21st, 2025, the Department of Telecommunications (DoT) launched the Financial Risk Indicator (FRI) feature, marking an important step towards safeguarding mobile phone users from the risks of financial fraud. This was developed as a part of the Digital Intelligence Platform (DIP), which facilitates coordination between stakeholders to curb the misuse of telecom services for conducting cyber crimes.
What is the Financial Risk Indicator (FRI)?
The FRI is a risk-based metric feature that categorises phone numbers into risk, medium risk, and high risk based on their association with financial fraud in the past. The data pool enabling this intelligence sharing includes the Digital Intelligence Unit (DIU) of the DoT, which engages and sends a list of Mobile Numbers that were disconnected (Mobile Number Revocation List - MNRL) to the following stakeholders, creating a network of checks and balances. They are:
- Intelligence from Non-Banking Finance Companies, and UPI (Unified Payment Interface) gateways.
- The Chakshu facility- a feature on the Sanchar Saathi portal that enables users to report suspected fraudulent communication (Calls, SMS, WhatsApp messages), which has also been roped in.
- Complaints from the National Cybercrime Reporting Portal (NCRP) through the I4C (Indian Cyber Coordination Center).
Some other initiatives taken up concerning securing against digital financial fraud are the Citizen Financial Cyber Fraud Reporting and Management System, the International Incoming Spoofed Calls Prevention System, among others.
A United Stance
The ease of payment and increasing digitisation might have enabled the increasing usage of UPI platforms. However, post-adoption, the responsibility of securing the digital payments infrastructure becomes essential. As per a report by CNBC TV18, UPI fraud cases surged by 85% in FY24. The number of incidents have increased from 7.25 lakh in FY23 to 13.42 lakh in FY24. These cases involved a total value of ₹1,087 crore, compared to ₹573 crore in the previous year, and the number continues to increase.
Nevertheless, UPI platforms are taking their own initiative to combat such crimes. PhonePe, one of the most used digital payment interface as of January 2025 (Statista) has already incorporated the FRI into its PhonePe Protect feature; this blocks transactions with high-risk numbers and issues a warning prior to engaging with numbers that are categorised to be of medium risk.
CyberPeace Insights
The launch of a feature addressing the growing threat of financial fraud is crucial for creating a network of stakeholders to coordinate with law enforcement to better track and prevent crimes. Publicity of these measures will raise public awareness and keep end-users informed. A secure infrastructure for digital payments is necessary in this age, with a robust base mechanism that can adapt to both current and future threats.
References
- https://www.thehawk.in/news/economy-and-business/centre-launches-financial-fraud-risk-indicator-to-safeguard-mobile-users
- https://telanganatoday.com/government-launches-financial-fraud-risk-indicator-to-safeguard-mobile-users
- https://www.pib.gov.in/PressReleasePage.aspx?PRID=2130249#:~:text=What%20is%20the%20%E2%80%9CFinancial%20Fraud,High%20risk%20of%20financial%20fraud
- https://www.business-standard.com/industry/news/dot-launches-financial-fraud-risk-indicator-to-aid-cybercrime-detection-125052101912_1.html
- https://www.cnbctv18.com/business/finance/upi-fraud-cases-rise-85-pc-in-fy24-increase-parliament-reply-data-19514295.htm
- https://www.statista.com/statistics/1034443/india-upi-usage-by-platform/#:~:text=In%20January%202025%2C%20PhonePe%20held%20the%20highest,key%20drivers%20of%20UPI%20adoption%20in%20India
- https://telecom.economictimes.indiatimes.com/amp/news/policy/centre-notifies-draft-rules-for-delicensing-lower-6-ghz-band/121260887?nt
