#FactCheck - Afghan Cricket Team's Chant Misrepresented in Viral Video

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 

‍

Misinformation is a scourge in the digital world, making the most mundane experiences fraught with risk. The threat is considerably heightened in conflict settings, especially in the modern era, where geographical borders blur and civilians and conflict actors alike can take to the online realm to discuss -and influence- conflict events. Propaganda can complicate the narrative and distract from the humanitarian crises affecting civilians, while also posing a serious threat to security operations and law and order efforts. Sensationalised reports of casualties and manipulated portrayals of military actions contribute to a cycle of violence and suffering. 

A study conducted by MIT found the mere thought of sharing news on social media reduced the ability to judge whether a story was true or false; the urge to share outweighed the consideration of accuracy (2023).  Cross-border misinformation has become a critical issue in today's interconnected world, driven by the rise of digital communication platforms. To effectively combat misinformation, coordinated international policy frameworks and cooperation between governments, platforms, and global institutions are created.

The Global Nature of Misinformation

Cross-border misinformation is false or misleading information that spreads across countries. Out-of-border creators amplify information through social media and digital platforms and are a key source of misinformation. Misinformation can interfere with elections, and create serious misconceptions about health concerns such as those witnessed during the COVID-19 pandemic, or even lead to military conflicts. 

The primary challenge in countering cross-border misinformation is the difference in national policies, legal frameworks and governance policies of social media platforms across various jurisdictions. Examining the existing international frameworks, such as cybersecurity treaties and data-sharing agreements used for financial crimes might be helpful to effectively address cross-border misinformation. Adapting these approaches to the digital information ecosystem, nations could strengthen their collective response to the spread of misinformation across borders. Global institutions like the United Nations or regional bodies like the EU and ASEAN can work together to set a unified response and uniform international standards for regulation dealing with misinformation specifically. 

Current National and Regional Efforts

Many countries have taken action to deal with misinformation within their borders. Some examples include:

• The EU’s Digital Services Act has been instrumental in regulating online intermediaries and platforms including marketplaces, social networks, content-sharing platforms, app stores, etc. The legislation aims to prevent illegal and harmful activities online and the spread of disinformation.
• The primary legislation that governs cyberspace in India is the IT Act of 2000 and its corresponding rules (IT Rules, 2023), which impose strict requirements on social media platforms to counter misinformation content and enable the traceability of the creator responsible for the origin of misinformation. Platforms have to conduct due diligence, failing which they risk losing their safe harbour protection. The recently-enacted DPDP Act of 2023 indirectly addresses personal data misuse that can be used to contribute to the creation and spread of misinformation. Also, the proposed Digital India Act is expected to focus on “user harms” specific to the online world. 
• In the U.S., the Right to Editorial Discretion and Section 230 of the Communications Decency Act place the responsibility for regulating misinformation on private actors like social media platforms and social media regulations. The US government has not created a specific framework addressing misinformation and has rather encouraged voluntary measures by SMPs to have independent policies to regulate misinformation on their platforms.

The common gap area across these policies is the absence of a standardised, global framework for addressing cross-border misinformation which results in uneven enforcement and dependence on national regulations.

Key Challenges in Achieving International Cooperation

Some of the key challenges identified in achieving international cooperation to address cross-border misinformation are as follows:

• Geopolitical tensions can emerge due to the differences in political systems, priorities, and trust issues between countries that hinder attempts to cooperate and create a universal regulation.
• The diversity in approaches to internet governance and freedom of speech across countries complicates the matters further.
• Further complications arise due to technical and legal obstacles around the issues of sovereignty, jurisdiction and enforcement, further complicating matters relating to the monitoring and removal of cross-border misinformation. 

CyberPeace Recommendations 

• The UN Global Principles For Information Integrity Recommendations for Multi-stakeholder Action, unveiled on 24 June 2024, are a welcome step for addressing cross-border misinformation. This can act as the stepping stone for developing a framework for international cooperation on misinformation, drawing inspiration from other successful models like climate change agreements, international criminal law framework etc. 
• Collaborations like public-private partnerships between government, tech companies and civil societies can help enhance transparency, data sharing and accountability in tackling cross-border misinformation. 
• Engaging in capacity building and technology transfers in less developed countries would help to create a global front against misinformation. 

Conclusion

We are in an era where misinformation knows no borders and the need for international cooperation has never been more urgent. Global democracies are exploring solutions, both regulatory and legislative, to limit the spread of misinformation, however, these fragmented efforts fall short of addressing the global scale of the problem. Establishing a standardised, international framework, backed by multilateral bodies like the UN and regional alliances, can foster accountability and facilitate shared resources in this fight. Through collaborative action, transparent regulations, and support for developing nations, the world can create a united front to curb misinformation and protect democratic values, ensuring information integrity across borders.

References

• https://economics.mit.edu/sites/default/files/2023-10/A%20Model%20of%20Online%20Misinformation.pdf 
• https://www.indiatoday.in/global/story/in-the-crosshairs-manufacturing-consent-and-the-erosion-of-public-trust-2620734-2024-10-21 
• https://laweconcenter.org/resources/knowledge-and-decisions-in-the-information-age-the-law-economics-of-regulating-misinformation-on-social-media-platforms/ ‍
• https://www.article19.org/resources/un-article-19-global-principles-for-information-integrity/

Introduction

‍
As the calendar pages turn inexorably towards 2024, a question looms large on the horizon of our collective consciousness: Are we cyber-resilient? This is not a rhetorical flourish but a pragmatic inquiry, as the digital landscape we navigate is fraught with cyberattacks and disruptions that threaten to capsize our virtual vessels.

What, then, is Cyber Resilience? It is the capacity to prepare for, respond to, and recover from these cyber squalls. Picture, if you will, a venerable oak amid a howling gale. The roots, those unseen sinews, delve deep into the earth, anchoring the tree – this is preparation. The robust trunk and flexible branches, swaying yet unbroken, embody response. And the new growth that follows the storm's rage is recovery. Cyber resilience is the digital echo of this natural strength and flexibility.

The Need for Resilience

Why, you might ask, is Cyber Resilience of such paramount importance as we approach 2024? The answer lies in the stark reality of our times:

• A staggering half of businesses have been breached by cyberattacks in the past three years.
• The financial haemorrhage from these incursions is projected to exceed a mind-numbing $10 trillion by the end of 2024.
• The relentless march of technology has not only brought innovation but also escalated the arms race against cyber threats.
• Cyber resilience transcends mere cybersecurity; it is a holistic approach that weaves recovery and continuity into the fabric of digital defenses.
• The adaptability of organisations, often through measures such as remote working protocols, is a testament to the evolving strategies of cyber resilience.

• The advent of AI and Machine Learning heralds a new era of automated cyber defense, necessitating an integrated framework that marries security with continuity protocols.
• Societal awareness, particularly of social engineering tactics, and maintaining public relations during crises are now recognised as critical elements of resilience strategies.
• Cyber threats have evolved in sophistication, paralleling the intense competition to develop new AI-driven solutions.
• As we gaze towards the future, cyber resilience is expected to be a prominent trend in both business and consumer technology sectors throughout 2024.

The Virtues 

The benefits of cyber resilience for organisations are manifold, offering a bulwark against the digital onslaught:

• A reduction in the risk of data breaches, safeguarding sensitive information and customer data.
• Business continuity, ensuring operations persist with minimal disruption.
• Protection of reputation, as companies that demonstrate effective cyber resilience engender trust.
• Compliance with data protection and privacy regulations, thus avoiding fines and legal entanglements.
• Financial stability, as the costs associated with breaches can be mitigated or even prevented.
• Enhanced customer trust, as clients feel more secure with companies that take cybersecurity seriously.
• A competitive advantage in a market rife with cyber threats.
• Innovation and agility, as cyber-resilient companies can pivot and adapt without fear of digital disruptions.
• Employee confidence, leading to improved morale and productivity.
• Long-term savings by sidestepping the expenses of frequent or major cyber incidents.

As the year wanes, it is a propitious moment to evaluate your organisation's cyber resilience. In this edition, we will guide you through the labyrinth of cyber investment buy-in, tailored discussions with stakeholders, and the quintessential security tools for your 2024 cybersecurity strategy.

How to be more Resilient 

Cyber resilience is more than a shield; it is the preparedness to withstand and recover from a cyber onslaught. Let us explore the key steps to fortify your digital defenses:

• Know your risks: Map the terrain where you are most vulnerable, identify the treasures that could be plundered, and fortify accordingly.
• Get the technology right: Invest in solutions that not only detect threats with alacrity but also facilitate rapid recovery, all the while staying one step ahead of the cyber brigands.
• Involve your people: Embed cybersecurity awareness into the fabric of every role. Train your crew in the art of recognising and repelling digital dangers.
• Test your strategies: Regularly simulate incidents to stress-test your policies and procedures, honing your ability to contain and neutralise threats.
• Plan for the worst: Develop a playbook so that everyone knows their part in the grand scheme of damage control and communication in the event of a breach.
• Continually review: The digital seas are ever-changing; adjust your sails accordingly. Cyber resilience is not a one-time endeavour but a perpetual commitment.

Conclusion 

As we stand on the precipice of 2024, let us not be daunted by the digital storms that rage on the horizon. Instead, let us embrace the imperative of cyber resilience, for it is our steadfast companion in navigating the treacherous waters of the cyber world. Civil Society Organizations such as ‘CyberPeace Foundation’ playing a crucial role in promoting cyber resilience by bridging the gap between the public and cybersecurity complexities, conducting awareness campaigns, and advocating for robust policies to safeguard collective digital interests. Their active role is imperative in fostering a culture of cyber hygiene and vigilance.

References 

• https://www.loginradius.com/blog/identity/cybersecurity-trends-2024/
• https://ciso.economictimes.indiatimes.com/news/ciso-strategies/cisos-guide-to-2024-top-10-cybersecurity-trends/106293196