#FactCheck -Claim That India’s GDP Was “Fake for 10 Years” is Misleading

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

Misinformation in India has emerged as a significant societal challenge, wielding a potent influence on public perception, political discourse, and social dynamics. A potential number of first-time voters across India identified fake news as a real problem in the nation. With the widespread adoption of digital platforms, false narratives, manipulated content, and fake news have found fertile ground to spread unchecked information and news.

In the backdrop of India being the largest market of WhatsApp users, who forward more content on chats than anywhere else, the practice of fact-checking forwarded information continues to remain low. The heavy reliance on print media, television, unreliable news channels and primarily, social media platforms acts as a catalyst since studies reveal that most Indians trust any content forwarded by family and friends. It is noted that out of all risks, misinformation and disinformation ranked the highest in India, coming before infectious diseases, illicit economic activity, inequality and labour shortages. World Economic Forum analysts, in connection with their 2024 Global Risk Report, note that “misinformation and disinformation in electoral processes could seriously destabilise the real and perceived legitimacy of newly elected governments, risking political unrest, violence and terrorism and long-term erosion of democratic processes.”

The Supreme Court of India on Misinformation

The Supreme Court of India, through various judgements, has noted the impact of misinformation on democratic processes within the country, especially during elections and voting. In 1995, while adjudicating a matter pertaining to keeping the broadcasting media under the control of the public, it noted that democracy becomes a farce when the medium of information is monopolized either by partisan central authority or by private individuals or oligarchic organizations. 

In 2003, the Court stated that “Right to participate by casting a vote at the time of election would be meaningless unless the voters are well informed about all sides of the issue in respect of which they are called upon to express their views by casting their votes. Disinformation, misinformation, non-information all equally create an uninformed citizenry which would finally make democracy a mobocracy and a farce.” It noted that elections would be a useless procedure if voters remained unaware of the antecedents of the candidates contesting elections. Thus, a necessary aspect of a voter’s duty to cast intelligent and rational votes is being well-informed. Such information forms one facet of the fundamental right under Article 19 (1)(a) pertaining to freedom of speech and expression. Quoting James Madison, it stated that a citizen’s right to know the true facts about their country’s administration is one of the pillars of a democratic State.

On a similar note, the Supreme Court, while discussing the disclosure of information by an election candidate, gave weightage to the High Court of Bombay‘s opinion on the matter, which opined that non-disclosure of information resulted in misinformation and disinformation, thereby influencing voters to take uninformed decisions. It stated that a voter had the elementary right to know the full particulars of a candidate who is to represent him in Parliament/Assemblies.

While misinformation was discussed primarily in relation to elections, the effects of misinformation in other sectors have also been discussed from time to time. In particular, The court highlighted the World Health Organisation’s observation in 2021 while discussing the spread of COVID-19, noting that the pandemic was not only an epidemic but also an “infodemic” due to the overabundance of information on the internet, which was riddled with misinformation and disinformation. While condemning governments’ direct or indirect threats of prosecution to citizens, it noted that various citizens who relied on the internet to provide help in securing medical facilities and oxygen tanks were being targeted by alleging that the information posted by them was false and was posted to create panic, defame the administration or damage national image. It instructed authorities to cease such threats and prevent clampdown on information sharing.

More recently, in Facebook v. Delhi Legislative Assembly [(2022) 3 SCC 529], the apex court, while upholding the summons issued to Facebook by the Delhi Legislative Assembly in the aftermath of the 2020 Delhi Riots, noted that while social media enables equal and open dialogue between citizens and policymakers, it is also a tool in the where extremist views are peddled into mainstream media, thereby spreading misinformation. It noted Facebook’s role in the Mynmar, where misinformation and posts that Facebook employees missed fueled offline violence. Since Facebook is one of the most popular social media applications, the platform itself acts as a power center by hosting various opinions and voices on its forum. This directly impacts the governance of States, and some form of liability must be attached to the platform. The Supreme Court objected to Facebook taking contrary stands in various jurisdictions; while in the US, it projected itself as a publisher, which enabled it to maintain control over the material disseminated from its platform, while in India, “it has chosen to identify itself purely as a social media platform, despite its similar functions and services in the two countries.”

Conclusion

The pervasive issue of misinformation in India is a multifaceted challenge with profound implications for democratic processes, public awareness, and social harmony. The alarming statistics of fake news recognition among first-time voters, coupled with a lack of awareness regarding fact-checking organizations, underscore the urgency of addressing this issue. The Supreme Court of India has consistently recognized the detrimental impact of misinformation, particularly in elections. The judiciary has stressed the pivotal role of an informed citizenry in upholding the essence of democracy. It has emphasized the right to access accurate information as a fundamental aspect of freedom of speech and expression. As India grapples with the challenges of misinformation, the intersection of technology, media literacy and legal frameworks will be crucial in mitigating the adverse effects and fostering a more resilient and informed society.

References

• https://thewire.in/media/survey-finds-false-information-risk-highest-in-india 
• https://www.statista.com/topics/5846/fake-news-in-india/#topicOverview
• https://www.weforum.org/publications/global-risks-report-2024/digest/  
• https://main.sci.gov.in/supremecourt/2020/20428/20428_2020_37_1501_28386_Judgement_08-Jul-2021.pdf
• Secretary, Ministry of Information & Broadcasting, Govt, of India and Others v. Cricket Association of Bengal and Another [(1995) 2 SCC 161]
• People’s Union for Civil Liberties (PUCL) v. Union of India [(2003) 4 SCC 399]
• Kisan Shankar Kathore v. Arun Dattatray Sawant and Others [(2014) 14 SCC 162]
• Distribution of Essential Supplies & Services During Pandemic, In re [(2021) 18 SCC 201]
• Facebook v. Delhi Legislative Assembly [(2022) 3 SCC 529]

‍

Introduction

‍

In today's era, where the threat from the digital world is growing rapidly, good developments in the war against cybercrimes cannot be ignored when they do happen. The state, which was notorious for being one of the most notorious criminal states in the country concerning digital crimes, has brought about remarkable changes in the law and order situation in the country in the last few years. According to the most recent data released by the Union Ministry of Home Affairs, Assam recorded only 408 cybercrimes in the year 2024, while the figure for the previous year was 909, which means a decline of over 55% in one year. But what is even more notable about the feat is the fact that, on the other hand, the country as a whole witnessed a rise of nearly 18% in the cybercrimes recorded. 

‍

 Assam's Cybercrime Journey

‍

To understand where Assam is today can only achieve this by understanding how far it has come. In 2021, it was ranked the 5^th highest state or union territory in India in the realm of cybercrime, with a staggering number of 4846 cases. The state kept the worrisome numbers continuing in the year 2022, as it ranked 9^th with 1,733 cases before sliding down to 13^th place in 2023 with 909 cases. However, the steep fall to a minuscule 408 cases in 2024 is an amazing narrative of how a state managed to completely eradicate the cybercrime infrastructure.

This is not a mere coincidence in statistics. This has proved to be a sustained, systematic operation by the law-enforcing agencies. Police sources say the decline is the result of consistent law enforcement action against cybercrime networks and the positive effect of awareness campaigns. Assam recorded 360 arrests under cybercrime-related offences and charge-sheeted 285 in the year 2024.

‍

The National Picture: A Troubling Contrast
‍

While the case of Assam is inspiring, the numbers on a pan-Indian level look bleak. India saw as many as 101,928 cybercrimes registered on its soil in the year 2024; a sharp rise from 2023, when 86,420 incidents had been reported. And it's not just in the number of cases that have seen an alarming rise; the economic implications are equally devastating. As many as 19.18 lakh complaints were lodged on the National Cyber Crime Reporting Portal in 2024. These complaints were an outcome of the financial losses to the tune of almost Rs 22,811.95 crore, according to a statement by the Home Ministry.

In 2024, the states that stood out at the top were Telangana (27,230), Karnataka (21,993), Uttar Pradesh (11,073), Maharashtra (9,922), and Bihar (6,380), among others. This clearly goes on to prove how the occurrences of cybercrime are not uniform across all states in the country due to various local and state-specific factors like the enforcement provided by state police forces, literacy levels of the general public, and the range of awareness campaigns carried out.

Even within Assam, the trend is not uniform, as almost every other Northeastern state has recorded a rise in cybercrimes this year, the figures being Arunachal Pradesh 24 to 78, Mizoram 31 to 50, Meghalaya 64 to 97, and Nagaland 2 to 14. It is only Tripura that saw a dip in reported cybercrimes, from 36 to 33. Considering these statistics, it becomes even more crucial to study and emulate Assam's success.
‍

The Drivers and Disrupters of Cybercrime
‍

It is in understanding how cybercrimes in Assam are committed that one might derive how these should be combatted. Of the 408 cases registered so far in 2024, 253 were registered for transmitting, or publishing, electronically, any obscene or sexually explicit material, and 115 cases were under computer-related offences. As for motive, they vary widely; 121 cases were registered out of revenge, 55 for fraud, 43 for extortion, and 42 for sexual exploitation. Sadly, out of 408 crimes reported so far in 2024, 196 victims are women, and 20 are children; in essence, the real impact is on society's most vulnerable.

This is a useful categorisation for the policymakers. It would not be beneficial if it were an all-encompassing strategy against cybercrimes when the motives and mechanisms behind them differ so widely. Customised campaigns educating women on cyberbullying, educating children on online security, and cautioning the public against online fraudulent schemes would be much more effective than general advice.

 On the national front, significant investments have been made by the central government for developing cybercrime-fighting infrastructure. Since the I4C was established in 2018, the launch of NCRP in 2019 provides a reporting and coordination framework against cybercrimes, and it is reported that over 5,489 crore have been saved by freezing illegal transactions, stemming from over 17.88 lakh complaints, through these platforms. Over 9.42 lakh SIM cards and 263,248 International Mobile Equipment Identities (IMEI) numbers have been blocked due to involvement in cybercrime.
‍

The Role of Awareness and Enforcement
‍

The biggest, and perhaps most transferable, lesson learned from Assam is the importance of both enforcement and awareness. Alone, neither proves useful: an enforcement operation without public knowledge leaves the public at risk for the next offence, while a purely informational approach gives criminals the license to proceed. Assam seems to have a more pragmatic approach; at least the statistics support this notion.

As the most persistent weakness, cyber hygiene is still a critical issue for India's cybersecurity. The core problem is the limited public knowledge on the importance of safer online practices, and it has been one of the primary hurdles to reducing crimes online; in instances where crimes were committed and reported, insufficient processes and infrastructure remained challenges in their investigation. Therefore, institutional investment in resources such as local police cyber cells and national coordinating agencies is an integral component to overcoming these challenges.z

‍

Conclusion
‍

By decreasing the rate of cybercrime in Assam by 55%, the successful combination of vigorous prosecution, constant pressure to uphold the law, and thorough public awareness campaigns has demonstrated a viable solution to ever-increasing online threats throughout India. Assam presents an attainable blueprint to diminish cybercrime, although the criminals of this evolving landscape cannot be constrained by individual state borders. To successfully achieve an e-economy that thrives on security and trust, India must adapt and expand the same law enforcement and awareness campaign strategies.

‍

References
‍

1. https://assamtribune.com/assam/assam-records-over-55-decline-in-cyber-crime-cases-in-2024-mha-1611895
   
   
2. https://ddnews.gov.in/en/cybercrime-complaints-cross-19-lakh-in-2024-97-drop-in-spoofed-calls-post-new-measures/
   
   
3. https://www.medianama.com/2025/08/223-india-cybercrime-500-percent-increase-2021-2024/
   
   
4. https://statista.com/topics/5054/cyber-crime-in-india‍
5. https://i4c.mha.gov.in