#FactCheck- Old Kerala church raid video falsely shared with exaggerated ₹7000 crore cash seizure claim

As AI language models become more powerful, they are also becoming more prone to errors. One increasingly prominent issue is AI hallucinations, instances where models generate outputs that are factually incorrect, nonsensical, or entirely fabricated, yet present them with complete confidence. Recently, ChatGPT released two new models—o3 and o4-mini, which differ from earlier versions as they focus more on step-by-step reasoning rather than simple text prediction. With the growing reliance on chatbots and generative models for everything from news summaries to legal advice, this phenomenon poses a serious threat to public trust, information accuracy, and decision-making.

What Are AI Hallucinations?

AI hallucinations occur when a model invents facts, misattributes quotes, or cites nonexistent sources. This is not a bug but a side effect of how Large Language Models (LLMs) work, and it is only the probability that can be reduced, not their occurrence altogether. Trained on vast internet data, these models predict what word is likely to come next in a sequence. They have no true understanding of the world or facts, they simulate reasoning based on statistical patterns in text. What is alarming is that the newer and more advanced models are producing more hallucinations, not fewer. seemingly counterintuitive. This has been prevalent reasoning-based models, which generate answers step-by-step in a chain-of-thought style. While this can improve performance on complex tasks, it also opens more room for errors at each step, especially when no factual retrieval or grounding is involved.

As per reports shared on TechCrunch, it mentioned that when users asked AI models for short answers, hallucinations increased by up to 30%. And a study published in eWeek found that ChatGPT hallucinated in 40% of tests involving domain-specific queries, such as medical and legal questions. This was not, however, limited to this particular Large Language Model, but also similar ones like DeepSeek. Even more concerning are hallucinations in multimodal models like those used for deepfakes. Forbes reports that some of these models produce synthetic media that not only look real but are also capable of contributing to fabricated narratives, raising the stakes for the spread of misinformation during elections, crises, and other instances.

It is also notable that AI models are continually improving with each version, focusing on reducing hallucinations and enhancing accuracy. New features, such as providing source links and citations, are being implemented to increase transparency and reliability in responses.

The Misinformation Dilemma

The rise of AI-generated hallucinations exacerbates the already severe problem of online misinformation. Hallucinated content can quickly spread across social platforms, get scraped into training datasets, and re-emerge in new generations of models, creating a dangerous feedback loop. However, it helps that the developers are already aware of such instances and are actively charting out ways in which we can reduce the probability of this error. Some of them are:

1. Retrieval-Augmented Generation (RAG): Instead of relying purely on a model’s internal knowledge, RAG allows the model to “look up” information from external databases or trusted sources during the generation process. This can significantly reduce hallucination rates by anchoring responses in verifiable data.
2. Use of smaller, more specialised language models: Lightweight models fine-tuned on specific domains, such as medical records or legal texts. They tend to hallucinate less because their scope is limited and better curated.

Furthermore, transparency mechanisms such as source citation, model disclaimers, and user feedback loops can help mitigate the impact of hallucinations. For instance, when a model generates a response, linking back to its source allows users to verify the claims made.

Conclusion

AI hallucinations are an intrinsic part of how generative models function today, and such a side-effect would continue to occur until foundational changes are made in how models are trained and deployed. For the time being, developers, companies, and users must approach AI-generated content with caution. LLMs are, fundamentally, word predictors, brilliant but fallible. Recognising their limitations is the first step in navigating the misinformation dilemma they pose.

References

• https://www.eweek.com/news/ai-hallucinations-increase/
• https://www.resilience.org/stories/2025-05-11/better-ai-has-more-hallucinations/ 
• https://www.ekathimerini.com/nytimes/1269076/ai-is-getting-more-powerful-but-its-hallucinations-are-getting-worse/ 
• https://techcrunch.com/2025/05/08/asking-chatbots-for-short-answers-can-increase-hallucinations-study-finds/ 
• https://en.as.com/latest_news/is-chatgpt-having-robot-dreams-ai-is-hallucinating-and-producing-incorrect-information-and-experts-dont-know-why-n/
• https://www.newscientist.com/article/2479545-ai-hallucinations-are-getting-worse-and-theyre-here-to-stay/
• https://www.forbes.com/sites/conormurray/2025/05/06/why-ai-hallucinations-are-worse-than-ever/ 
• https://towardsdatascience.com/how-i-deal-with-hallucinations-at-an-ai-startup-9fc4121295cc/ 
• https://www.informationweek.com/machine-learning-ai/getting-a-handle-on-ai-hallucinations 

‍

Introduction

‍

The trajectory of India's digital economy is growing at an unprecedented rate, and so is India's cybercrime ecosystem. Parliamentary data tabled before the Rajya Sabha in May 2024 by the MHA suggests an overwhelming 900% growth in cybercrime complaints from 2021 to '25, while annual losses crossed 22,800 crore in 2024. The structural issues like the low victim restitution rate, the lack of forensic infrastructure, issues of jurisdiction related to offshore fraud factories targeting Indian citizens, and the huge disparity in awareness levels amongst India's youngest online citizens continue to exist. This brief brings out the clear trends in cybercrime, the role of institutional mechanisms in its prevention and response, failure points, and recommends appropriate policy interventions from the perspective of CyberPeace.

‍

The Data Imperative

‍

Since its operationalisation in 2019 by the Indian Cyber Crime Coordination Centre (I4C), the NCRP serves as India's most significant institutional apparatus for cybercrime reporting and response. Data placed before the Rajya Sabha by the Ministry of Home Affairs on 30 July 2025 show that, with almost no exception, complaints of cybercrime have increased far more quickly than most traditional indicators of public safety. Between 2021 and June 2025, the NCRP received 6.59 million complaints, evidence of both a sustained and escalating expansion of India's cyber threat profile. Complaints per year more than quadrupled from 4.52 lakh in 2021 to 19.18 lakh in 2024 (324% over the period); by 2025, the NCRP had received 28.15 lakh complaints, a 523 percent rise compared with the 2021 baseline:

Clearly, cyber-enabled crime is no longer an occasional crisis but a systemic governance issue requiring consistent regulation and institution-building.

The financial fallout has also accelerated dramatically. Figures indicate that reported financial losses due to cybercrime jumped from 2,290 crore in 2022 to 22,812 crore in 2024 a 895% leap in two years:

‍

‍

Though response mechanisms such as the Citizen Financial Cyber Fraud Reporting and Management System (CFCFRMS) successfully blocked or recovered close to 8,690 crore as of January 2026, victims appear to get back only about 2.18 percent of the losses they report. 

In most areas, reporting and response have expanded greatly, but both the rate and scale of cyber-enabled financial fraud continue to outstrip India's remediation and law enforcement capacity. 

‍

Threat Typology of India’s Fraud Ecosystem

‍

The nature of cyber crime in India has evolved from an opportunistic volume-based activity to a layered transnational criminal environment. I4C intelligence as tabled in Parliament reveals investment scams as the biggest threat: they accounted for 76% of the financial fraud lost in 2025 (although only 35% of complaints were filed, thus, a very high value per case was lost).

‍

‍

Digital arrest frauds, which tap on citizens' unawareness that "digital arrest" is not permissible under Indian law, rose from 39,925 cases (91 crore) in 2022 to 123,672 cases (1,935crore) in 2024.

‍

‍

The fast rise in the number of incidents as well as in the volume of fraud clearly points out that digital arrest fraud has moved away from the phase of novel scam typology to a formidable cyber-extortion landscape. The main orchestrators of investment, trading, dating, and digital arrest scams targeting Indian citizens were recently identified by the I4C CEO Rajesh Kumar as transnational criminal scam networks in Cambodia, Myanmar, and Laos. Hence, this issue does not only fall within the domain of domestic law enforcement but constitutes a transnational cybercrime requiring parallel financial intelligence, diplomatic initiative, platform responsibility, and international investigative collaboration.

‍

‍

‍

Geographic Concentration
‍

Maharashtra and UP register the highest volumes in total complaints at 3.03 lakh and 3.01 lakh, owing to them being the financial capital and most populous state, respectively. Karnataka, Gujarat, Delhi, WB, Telangana, TN, Rajasthan, and Haryana register above 1 lakh complaints each. However, the critical information that is being missed is that while complaint rate growth is the fastest in Tier 2 and 3 geographies (Haryana leads per-capita complaint rate with 381/100k people in 2023; Telangana (261); Uttarakhand (243)), this signifies rural digital growth as a risk multiplier.
‍

Institutional Architecture: Mechanisms and Performances
‍

India's institutional response to cybercrime, led by the Ministry of Home Affairs' Indian Cyber Crime Coordination Centre (I4C), is one of the world's largest real-time fraud detection and prevention ecosystems. The backbone of this is the Citizen Financial Cyber Fraud Reporting and Management System (CFCFRMS), which has onboarded over 700 banks, payment service providers, e-commerce portals, digital wallets, and, since the Standard Operating Procedure was issued on 2nd January 2026, virtual asset service providers and crypto exchanges. This interconnected network allows for prompt freezing of funds and timely fraud intervention during the 'golden hour' of a cybercrime report. 

‍

Institutional capacity is robust, with approximately 8,690 crore saved via the CFCFRMS since its inception for over 24.65 lakh complaints. The national cybercrime helpline (1930) receives close to 10,000 calls daily, while the Suspect Registry has enabled the rejection of 9,519 crore via the detection of 23.05 lakh suspect entities and 27.37 lakh mule accounts. In parallel, the CyTrain platform has expanded training by registering 151,081 police and judicial officers and issuing 142,025 certificates. Cyberforensic labs in all 33 States and Union Territories have received central assistance totalling 132.93 crore, and data-driven interstate crime analytics and offender linkages through the Samanvaya and Pratibimb platforms have led to 21,857 arrests.
‍

Ecosystem Gaps
‍

Through I4C, CFCFRMS, CyTrain, and the establishment of forensic infrastructure in states, India’s cybercrime ecosystem has greatly grown. But due to the rapid proliferation of cybercrime, systemic shortcomings are revealed regarding the restoration of victims, investigation, forensic capacity, cross-border enforcement, awareness, and stakeholder coordination:

• Victim Restitution Deficit: Although the total of ₹ 8,690 crore frozen has increased, the refund for victim compensation is limited to only ₹ 167 crore (2.18%) due to lengthy restoration processes relying on court orders.
• Forensic Capacity Limitations: 2 national, state-level, unevenly equipped cyber forensic labs can’t match the needs of over 10 million cybercrime complaints per year.
• Low conviction rate: The investigations of cybercrimes suffer from evidence collection and criminal proceedings, leading to limited conviction rates.
• Cross-border enforcement challenges: Many of the investment and digital arrest scams, in fact, are originating from Cambodia, Myanmar, and Laos, rendering the cybercrime response mechanisms of India helpless.
• Lack of Awareness: First-time digital users are quite prone to online scams and fraud, and many of the victims continue not reporting due to social stigma and lack of confidence.
• Partial Stakeholder Integration: Banks and small financial institutions, small companies, and emerging virtual asset providers not yet on board allow the money to slip through without being tracked.

CyberPeace Insights: Strategic Way Forward
‍

India has already built a relatively mature response structure for cybercrime with I4C, CFCFRMS, and CyTrain and is coordinating the financial sector on it. The way ahead lies in outcome-oriented improvements and not just in the ability to report and intercept more. Here are the priority interventions that address the most important institutional shortcomings identified in the current ecosystem:

• Fast-track victim restoration: Introduce time-bound victim restoration mechanisms for low-value incidents through simplified processes and mandate national-level roll-out of successful Lok Adalat-based settlement mechanisms.
• District-level cyber forensics: Establish cyber forensic support units at the district level and enhance access to mobile, cloud, and blockchain forensic capabilities.
• AI-powered fraud prevention: Mandate deep-fake and voice-clone detection mechanisms across all financial institutions and telecom networks; embed predictive risk analytics into transaction screening frameworks.
• Cyber Suraksha Gram initiative: Increase digital fraud awareness across all common service centres, Jan Dhan enrollment schemes, and rural banking channels, and tackle the awareness asymmetry.
• Regional cybercrime coordination: Establish real-time, operational intelligence-sharing mechanisms with Southeast Asian economies, which have become home to large scam networks preying on Indian citizens.
• Specialised cyber prosecution ecosystem: Develop exclusive cyber courts, standardise digital evidence procedures, and broaden the scope of CyTrain to include the development of specialised cadres of investigators and prosecutors capable of handling increasingly complex cybercrime cases.

Conclusion
‍

The 22,812 crore lost due to cybercrime in 2024 was more than a mere figure; it signifies a serious concern regarding citizen trust, economic security, and digital inclusion. Though India's institutional response to cybercrime is one of the largest, with an operational I4C and a CFCFRMS functioning in real time, the victim compensation and prosecution mechanism falls short. It's time for implementation: faster recovery of resources, increased enforcement, a larger scale of awareness, and finally, translating the institutional innovations into concrete justice for victims nationwide.

References
‍

1. https://sansad.in/getFile/annex/270/AU1341_tmaxdx.pdf?source=pqars
2. https://www.mha.gov.in/MHA1/Par2017/pdfs/par2025-pdfs/LS02122025/452.pdf
3. https://www.pib.gov.in/PressReleasePage.aspx?PRID=2244504&reg=3&lang=2
4. https://static.pib.gov.in/WriteReadData/specificdocs/documents/2025/oct/doc2025107659501.pdf
5. https://www.medianama.com/2025/08/223-india-cybercrime-500-percent-increase-2021-2024/
6. https://theprint.in/india/cybercrime-saw-24-spike-in-2025-indians-lost-rs-22495-crore-mainly-in-investment-scams/2859930/

‍

‍

Introduction

The most significant change seen in the Indian cyber laws this year was the passing of the Digital Personal Data Protection Act, 2023, in the parliament. DPDP Act is the first concrete form of legislation focusing on the protection of Digital Personal Data of Indian netizens in all aspects; the act is analogous to what GDPR is for Europe. The act lays down heavy compliance mandates for the intermediaries and data fiduciaries, this has made it difficult for the tech companies a lot of policy, legal and technical changes have to be made in order to implement the act to its complete efficiency. Recently, the big techs have addressed a letter to the Minister and Minister of State of Meity to extend the implementation timeline of the act. In other news, the union cabinet has given the green light for the much-awaited MoC with Japan focused on establishing a long-term Semiconductor Supply Chain Partnership. 

Letter to Meity

The lobby of the big techs represented by a Trade Body named the Big Tech Asia Internet Coalition (AIC) this week wrote to the Ministry of Electronics and Information Technology (Meity), addressing it to the Minister Ashwini Vaishnav and Minister of State (MoS) Rajeev Chandershekhra recommending a 12-18 month extension on the implementation of the Digital Personal Data Protection Act. This request comes at a time when the government has been voicing its urgency to implement the act in order to safeguard Indian data at the earliest. The trade body represented big names, including Meta, Google, Microsoft, Apple and many more. These big techs essentially comprise the segment recognised under the DPDP as the Significant Data Fiduciaries due to the sheer volume of data processed, hosted, stored, etc. In the protective sense, the act has been designed to focus on preventing the exploitation of personal data of Indian netizens by the big techs, hence, they form an integral part of the Indian Data Ecosystem.  The following reasons/complications concerning the implementation of the act were highlighted in the letter:

• Unrealistic Timelines: The AIC expressed that the current timeline for the implementation of the act seems unrealistic for the big techs to establish technological, policy and legal mechanisms to be in compliance with section 5 of the act, which talks about the Obligations of a Data Fiduciary and the particular notice to be shared with the data principles in accordance with the act. 
• Technical Requirements: Members of AIC expressed that the duration for the implementation of the act is much less in comparison to the time required by the tech companies to set up/deploy relevant technical critical infrastructure, SoPs and capacity building for the same. This will cause a major hindrance in establishing the efficiency of the act.
• Data Rights:  Right to Erasure, Correction, Deletion, Nominate, etc., are guaranteed under the DPDP, but the big techs are not sure about the efficient implementation of these rights and hence will need fundamental changes in the technology architecture of their platform, thus expressing concern of the early implementation of the act. 
• Equivalency to GDPR: The DPDP is taken to be congruent to the European GDPR, but the DPDP focuses on a few more aspects, such as cross-border data flow and compliance mandates for the right to erasure, hence a lot of GDPR-compliant big techs also need to establish more robust mechanisms to maintain compliance to Indian DPDP.     

Indo-Japan MoC

A Memorandum of Cooperation (MoC) on the Japan-India Semiconductor Supply Chain Partnership was signed in July 2023 between the Ministry of Electronics and Information Technology (MeitY) of India and the Ministry of Economy, Trade and Industry (METI) of Japan. This information was shared with the Union Cabinet, which is led by Prime Minister Narendra Modi. The Ministry of Commerce (MoC) aims to expand collaboration between Japan and India in order to improve the semiconductor supply chain. This is because semiconductors are critical to the development of industries and digital technologies. The Parties agree that the MoC will take effect on the date of signature and be in effect for five years. Bilateral cooperation on business-to-business and G2G levels on ways to develop a robust semiconductor supply chain and make use of complementary skills. The cooperation is aimed at harnessing indigenous talent and creating opportunities for higher employment avenues. 

MeitY's purpose also includes promoting international cooperation within bilateral and regional frameworks in the frontier and emerging fields of information technology. MeitY has engaged in Memorandums of Understanding (MoUs), Memorandums of Covenants (MoCs), and Agreements with counterpart organisations/agencies of other nations with the aim of fostering bilateral collaboration and information sharing. Additionally, MeitY aims to establish supply chain resilience, which would enable India to become a reliable partner. An additional step towards mutually advantageous semiconductor-related commercial prospects and collaborations between India & Japan is the strengthening of mutual collaboration between Japanese and Indian enterprises through this Memorandum of Understanding. The “India-Japan Digital Partnership” (IJDP), which was introduced during PM Modi's October 2018 visit to Japan, was created in light of the two countries' complementary and synergistic efforts. Its goal is to advance both current areas of cooperation and new initiatives within the scope of S&T/ICT cooperation, with a particular emphasis on “Digital ICT Technologies."

Conclusion 

As we move ahead into the digital age, it is pertinent to be aware and educated about the latest technological advancements, new forms of cybercrimes and threats and legal aspects of digital rights and responsibilities, whether it is the recommendation to extend the implementation of DPDP or the Indo-Japan MoC, both of these instances impact the Indian netizen and his/her interests. Hence, the indigenous netizen needs to develop a keen interest in the protection of the Indian cyber-ecosystem to create a safer future. In our war against technology, our best weapon is technology and awareness, thus implementing the same in our daily digital lifestyles and routines is a must.  

References

• https://www.eetindia.co.in/cabinet-approves-moc-on-japan-india-semiconductor-supply-chain-partnership/
• https://www.moneycontrol.com/news/business/startup/trade-body-representing-big-tech-urges-govt-to-extend-dpdp-act-implementation-by-1-5-years-11605431.html
• https://www.google.com/url?rct=j&sa=t&url=https://www.eetindia.co.in/cabinet-approves-moc-on-japan-india-semiconductor-supply-chain-partnership/&ct=ga&cd=CAEYACoTOTI3Mzg4NzEyODgwMjI2ODk0MDIaOTBiYzUxNmI5YTRjYTE1NTpjb206ZW46VVM&usg=AOvVaw2lEO7-cIBZ_ox1xV39LGLs

‍

‍