#FactCheck - Old Video of Yogi Adityanath Edited to Push ‘Next PM 2029’ Claim

Introduction

The advent of frontier AI has significantly widened the range of actors who can launch cyberattacks, extending beyond state actors with immense capabilities or organized professional cybercriminal rings. In its most critical advisory, CIAD-2026-0020, titled "Defending against frontier AI-driven cyber risks," which was released on April 26, 2026, the Indian Computer Emergency Response Team (CERT-In) officially stated that AI can now carry out autonomous cyber activities of unprecedented scale and speed. The advisory highlights that these frontier AI models can perform automated reconnaissance, phishing, malware creation, vulnerability identification, and social engineering with minimal human involvement, thus "lowering the barrier to orchestrating complex cyber attacks." The risks that such AI models pose are not restricted to state actors and corporate entities anymore and also extend to MSMEs, public organizations, and individuals.

‍

India’s Escalating Cybercrisis

The Indian digital economy has been developing at a very fast pace, but the same cannot be said about its cybersecurity. Having a base of over 850 million internet users and a digital payment sector that records a massive 22,495 crore in monthly transaction volumes, coupled with the fastest-growing cloud sector in the world, India continues to remain a lucrative prey for cybercriminals. There were over 265 million attempts reported in the last year, 2025, alone, where close to 46% of all incidents detected were in enterprises with fewer than 1,000 employees, a very grave reality for MSMEs. MHA confirmed there were 28.15 lakh reported cybercrime complaints in 2025 as compared to 2024, with a jump of 24%. In this worsening environment the advisory is a breakthrough in Indian cyber governance. Where previously advisories covered only conventional threats like phishing and malware, the new warning names frontier agentic AI systems as autonomous multipliers of threats, capable of conducting operations at scale and speed with significantly reduced human oversight.

‍

What is “Frontier AI” and why does it matter?

CERT-In’s decision to adopt the term "Frontier AI" is deliberate and meaningful. The advisory’s scope is a new category of agentic AI, which moves well beyond traditional chatbot-style AI, having the capacity to reason, plan, perform multiple actions in a single task autonomously, and carry out complicated tasks with minimal or no human guidance. CERT-In highlights that these tools now possess the capabilities that were "previously carried out by a coordinated team of skilled cybersecurity professionals." The advisory clearly flags the risk that these advanced models have the capability to generate malicious code, conduct network scans, probe systems for vulnerabilities, and even orchestrate intricate multi-stage cyberattacks in a single session. Their capacity to analyse a vast number of source code libraries to identify vulnerabilities, even unknown zero-day ones, and then develop proof-of-concept exploits at high speed. This means that the historical lead time to turn a vulnerability discovery into an exploit tool has reduced from weeks to just hours.

‍

Six Core Threat Vectors identified by CERT-In

1. AI-driven Automatic Zero-Day Discovery: AI-based solutions discover zero-day vulnerabilities and automatically create exploits in minutes, reducing the time taken by defenders.
2. AI-driven Autonomous Reconnaissance: AI-driven agents scan cloud infra, APIs, and enterprise networks and outline attack vectors.
3. AI-driven phishing & deepfakes: Multilingual, highly targeted phishing emails, deepfake audio, and deepfake voice/video calls bring sophistication to social engineering.
4. Deepfake Financial Fraud: AI creates deepfake executives for high-value money transfers. For example, reports have indicated crore-level fund loss cases in India.
5. AI-powered Autonomous Attack Chains: Advanced AI models are able to automatically perform multiple malicious stages like privilege escalation, lateral movement, data exfiltration, and data extraction.
6. Cascading failures of interconnected systems: A single AI-supported security breach can have catastrophic domino effects on connected digital systems and critical infrastructures.

‍

Why are MSMEs a target?

CERT-In’s warning is specifically targeted toward the weakness of the Indian MSMEs. Contributing almost 30% to India's GDP and employing over 110 million individuals, most MSMEs have failed to adequately prepare themselves against contemporary cyber threats. While a large corporation would have a full-time cybersecurity team, a security operation centre, and frequent vulnerability assessments, the majority of MSMEs lack such infrastructure due to budget constraints, out-of-date software, etc. This lack of security has proved to be quite disadvantageous for smaller businesses, as India was identified as one of the top global targets for cyberattacks, where approximately 46% of the total breaches worldwide targeted organizations having fewer than 1000 employees. The advisory claims that frontier AI systems have significantly increased the threats, for the skills necessary to carry out advanced cyberattacks have dramatically decreased. Ransomware, phishing and data exfiltration can be executed by even unsophisticated attackers. The aftermath could result in critical financial, operational, and compliance impact on these MSMEs.

‍

The Global Context

These developments seem to validate CERT-In's warning about threats posed by frontier AI. In its 2026 State of Cybersecurity Report, ISACA listed AI-related threats as the top concern of cybersecurity professionals; 61% of those surveyed reported generative AI/large language models as the top technology trend impacting cyber risk. Worryingly, in 2026 only 7% were confident in their organizations' defenses against ransomware. Check Point Software's Cyber Security Report 2026 corroborates this; in 2025 the report stated that in a single year, the trend of combined social engineering-based campaigns with automated operational execution has risen considerably. In all phases of the lifecycle of a cyberattack reconnaissance, social engineering, and tactical decision-making AI is being applied. KPMG is warning of deepfake-enabled fraud now "spreading at a faster rate than that experienced at the beginning of the phishing era, which is currently still the leading type of attack in the world."

‍

CERT-In Recommendations

‍

For Large Organisations:

‍

1. The use of security monitoring, threat detection, and log analysis should be increased.
2. DDoS protection systems and multi-factor authentication (MFA) should be implemented on all internet-facing devices and assets.
3. Critical security patches should be installed within 24 hours of release.
4. Old VPN and remote-access infrastructure should be updated or replaced.
5. AI-driven cyber drills and incident response simulations should be regularly performed.

‍For MSMEs:

‍

1. Software and security updates should be automatically enabled on all devices and systems.
2. MFA should be enabled on organisational accounts and sensitive platforms.
3. MSMEs should utilize MSSPs for specialized support and monitoring.
4. Detailed inventories of IT assets and system logs should be kept for fast incident response.
5. Staff should be educated about identifying AI-generated phishing, deepfakes, and scams.

‍

For Individuals:

‍

• Independent communication channels should be used to verify any dubious message or money request.
• Software from unverified sources or unauthorised channels should not be downloaded.
• The use of strong and unique passwords along with MFA wherever possible should be enforced.

‍

From Advisory to Action

The May 2026 cybersecurity road map released by CERT-In signals a departure from identification of threats to enabling operations against frontier AI-led cyber threat landscapes. This initiative builds on their April advice and delineates a clearly articulated three-phase roadmap comprising immediate cyber readiness, AI governance controls, and deep integration of AI-driven defenses. It also provides for the establishment of a focused AI Cyber Defense Center and various multisector governance provisions. A prominent area is the increased threat of impersonation via deepfakes, and companies are encouraged to institute executive verification procedures prior to approving high-value transactions. The framework also emphasizes the establishment of an AI asset register requiring formal accounting and governance of all AI systems utilized in an enterprise. Meanwhile, CERT-In also recognizes the twin-use nature of frontier AI: for every threat, the same technology can bolster security with automated threat detection, phishing, and log analysis in real time.  However, the deployment of state-of-the-art defenses is uneven, especially with MSMEs, where there isn’t the requisite domain expertise and funding for this infrastructure. Accordingly, the road map puts the emphasis on immediate and stronger cyber hygiene, compulsory incident reporting, enhancing AI literacy, and proper implementation of the Digital Personal Data Protection Act for long-term security investment and resilience.

‍

Conclusion

The CERT-In advisory CIAD-2026-0020 signifies a vital acknowledgment of AI's transformational impact on the cybersecurity ecosystem. Capabilities formerly exclusive to elite state actors are being deployed by low-skilled users, leveraging state-of-the-art frontier AI tools. India’s MSMEs, enterprises, and digital citizens are experiencing a rapidly accelerating threat milieu. In this context, the CERT-In advisory and the ensuing blueprint can no longer be dismissed as ordinary government pronouncements but as critical operational imperatives. It is the country’s ability over the next few years to shore up its collective cyber resilience to the ever-increasing scale and sophistication of AI-powered attacks that will prove crucial.

‍

References:

1. https://www.cert-in.org.in/s2cMainServlet?pageid=PUBVLNOTES02&VLCODE=CIAD-2026-0020
2. https://www.zeebiz.com/technology/news-cert-in-flags-high-severity-ai-cyber-risks-amid-claude-mythos-concerns-394448
3. https://www.business-standard.com/technology/tech-news/cert-in-warning-ai-scams-frontier-models-mythos-gpt-5-5-what-it-means-126042800988_1.html
4. https://www.businesswire.com/news/home/20251020612551/en/
5. https://corporate.indiamart.com/2025/07/29/staying-ahead-of-cyber-threats/
6. https://kpmg.com/kpmg-us/content/dam/kpmg/pdf/2025/deepfakes-real-threat.pdf

‍

Introduction

According to a new McAfee survey, 88% of American customers believe that cybercriminals will utilize artificial intelligence to "create compelling online scams" over the festive period. In the meanwhile, 31% believe it will be more difficult to determine whether messages from merchants or delivery services are genuine, while 57% believe phishing emails and texts will be more credible. The study, which was conducted in September 2023 in the United States, Australia, India, the United Kingdom, France, Germany, and Japan, yielded 7,100 responses. Some people may decide to cut back on their online shopping as a result of their worries about AI; among those surveyed, 19% stated they would do so this year. 

In 2024, McAfee predicts a rise in AI-driven scams on social media, with cybercriminals using advanced tools to create convincing fake content, exploiting celebrity and influencer identities. Deepfake technology may worsen cyberbullying, enabling the creation of realistic fake content. Charity fraud is expected to rise, leveraging AI to set up fake charity sites. AI's use by cybercriminals will accelerate the development of advanced malware, phishing, and voice/visual cloning scams targeting mobile devices. The 2024 Olympic Games are seen as a breeding ground for scams, with cybercriminals targeting fans for tickets, travel, and exclusive content.

AI Scams' Increase on Social Media

Cybercriminals plan to use strong artificial intelligence capabilities to control social media by 2024. These applications become networking goldmines because they make it possible to create realistic images, videos, and audio. Anticipate the exploitation of influencers and popular identities by cybercriminals. 

AI-powered Deepfakes and the Rise in Cyberbullying

The negative turn that cyberbullying might take in 2024 with the use of counterfeit technology is one trend to be concerned about. This cutting-edge technique is freely accessible to youngsters, who can use it to produce eerily convincing synthetic content that compromises victims' privacy, identities, and wellness.

In addition to sharing false information, cyberbullies have the ability to alter public photographs and re-share edited, detailed versions, which exacerbates the suffering done to children and their families. The study issues a warning, stating that deepfake technology would probably cause online harassment to take a negative turn. With this sophisticated tool, young adults may now generate frighteningly accurate synthetic content in addition to using it for fun. The increasing severity of these deceptive pictures and phrases can cause serious, long-lasting harm to children and their families, impairing their identity, privacy, and overall happiness.

Evolvement of GenAI Fraud in 2023

We simply cannot get enough of these persistent frauds and fake emails. People in general are now rather adept at [recognizing] those that are used extensively. But if they become more precise, such as by utilizing AI-generated audio to seem like a loved one's distress call or information that is highly personal to the person,  users should be much more cautious about them. The rise in popularity of generative AIs brings with it a new wrinkle, as hackers can utilize these systems to refine their attacks:

• Writing communications more skillfully in order to deceive consumers into sending sensitive information, clicking on a link, or uploading a file.
• Recreate emails and business websites as realistically as possible to prevent arousing concern in the minds of the perpetrators.
• People's faces and voices can be cloned, and deepfakes of sounds or images can be created that are undetectable to the target audience. a problem that has the potential to greatly influence schemes like CEO fraud.
• Because generative AIs can now hold conversations, and respond to victims efficiently.
• Conduct psychological manipulation initiatives more quickly, with less money spent, and with greater complexity and difficulty in detecting them. AI generative already in use in the market can write texts, clone voices, or generate images and program websites.

AI Hastens the Development of Malware and Scams

Even while artificial intelligence (AI) has many uses, cybercriminals are becoming more and more dangerous with it. Artificial intelligence facilitates the rapid creation of sophisticated malware, illicit web pages, and plausible phishing and smishing emails. As these risks become more accessible, mobile devices will be attacked more frequently, with a particular emphasis on audio and visual impersonation schemes.

Olympic Games: A Haven for Scammers

The 2024 Olympic Games are seen as a breeding ground for scams, with cybercriminals targeting fans for tickets, travel, and exclusive content. Cybercriminals are skilled at profiting from big occasions, and the buzz that will surround the 2024 Olympic Games around the world will make it an ideal time for scams. Con artists will take advantage of customers' excitement by focusing on followers who are ready to purchase tickets, arrange travel, obtain special content, and take part in giveaways. During this prominent event, vigilance is essential to avoid an invasion of one's personal records and financial data. 

Development of McAfee’s own bot to assist users in screening potential scammers and authenticators for messages they receive

Precisely such kind of technology is under the process of development by McAfee. It's critical to emphasize that solving the issue is a continuous process. AI is being manipulated by bad actors and thus, one of the tricksters can pull off is to exploit the fact that consumers fall for various ruses as parameters to train advanced algorithms. Thus, the con artists may make use of the gadgets, test them on big user bases, and improve with time.

Conclusion

According to the McAfee report, 88% of American customers are consistently concerned about AI-driven internet frauds that target them around the holidays. Social networking poses a growing threat to users' privacy. By 2024, hackers hope to take advantage of AI skills and use deepfake technology to exacerbate harassment. By mimicking voices and faces for intricate schemes, generative AI advances complex fraud. The surge in charitable fraud affects both social and financial aspects, and the 2024 Olympic Games could serve as a haven for scammers. The creation of McAfee's screening bot highlights the ongoing struggle against developing AI threats and highlights the need for continuous modification and increased user comprehension in order to combat increasingly complex cyber deception.

References

• https://www.fonearena.com/blog/412579/deepfake-surge-ai-scams-2024.html
• https://cxotoday.com/press-release/mcafee-reveals-2024-cybersecurity-predictions-advancement-of-ai-shapes-the-future-of-online-scams/#:~:text=McAfee%20Corp.%2C%20a%20global%20leader,and%20increasingly%20sophisticated%20cyber%20scams.
• https://timesofindia.indiatimes.com/gadgets-news/deep-fakes-ai-scams-and-other-tools-cybercriminals-could-use-to-steal-your-money-and-personal-details-in-2024/articleshow/106126288.cms
• https://digiday.com/media-buying/mcafees-cto-on-ai-and-the-cat-and-mouse-game-with-holiday-scams/

Executive Summary

‍

An image of a banknote featuring Iran’s Supreme Leader Ayatollah Khamenei is going viral on social media, with claims that Iran’s central bank has issued a newly designed 5 million rial note bearing his portrait. However, a fact-check by the CyberPeace Research Wing  has found the claim to be misleading.

‍

Claim
‍

The image was shared by a verified user, “Sprinter Press Agency,” on X (formerly Twitter), claiming that the Central Bank had introduced a new banknote design featuring the leader of the Islamic Revolution.

• https://x.com/ImHuMeer786H/status/2048042719231291661

‍

‍

Fact Check

‍

To verify the claim, relevant keywords were searched across multiple credible sources. No reports were found from any reputable international media outlet, Iranian government platform, or the Central Bank of Iran confirming the release of such a banknote. A technical analysis of the viral image was also conducted. According to the AI detection tool Zhuque AI Detection Assistant, there is a 63.8% probability that the image is AI-generated, raising further doubts about its authenticity.

‍

‍

Conclusion:
‍

The claim that Iran’s central bank has issued a new 5 million rial banknote featuring Ayatollah Khamenei is misleading. There is no official confirmation of such a release, and available evidence suggests that the viral image is either edited or AI-generated.

‍