KEY TAKEAWAYS:
- Information disorder — the mix of misinformation, disinformation and malinformation — has moved from a communications problem to a security risk.
- The World Economic Forum’s Global Risks Report 2025 ranked misinformation and disinformation as a leading global risk for the second year running.
- The authors of a September 2026 SecurityLinkIndia feature argue the core shift for analysts is that the problem is no longer too little information but too much of it, spreading faster than it can be verified. Six drivers amplify it: cognitive biases, echo chambers, media saturation, social polarisation, engagement-driven platforms and generative AI.
- Deepfakes add a second-order risk: fabricated content can be mistaken for real, while real content can be dismissed as fake.
- Their recommended response for open-source intelligence (OSINT) teams combines a repository of known manipulations, tracking of AI capabilities, automated detection that supports rather than replaces human analysts, and corroboration across independent sources.
This post summarises a feature by Prerna Panwar (NTT Data GDC) and Saima Shamim (MitKat Advisory) in the September 2026 issue of SecurityLinkIndia. It restates their argument in brief; the full article, with its references, is the authoritative version.
What is information disorder, and why does it matter for security?
The authors describe information disorder as the spread of misinformation, disinformation and malinformation through an ecosystem shaped by people, platforms, algorithms and new technology. They trace deliberate disinformation back well before the internet, but argue that easy access and weak control over information flows in this century turned it into a systemic problem, one that drew wide attention around 2016. They point to the World Economic Forum’s Global Risks Report 2025, which named misinformation and disinformation a leading global risk for the second consecutive year, citing links to polarisation, weaker state legitimacy and impaired civil protection. Their conclusion is that information operations should be treated as systemic hazards needing coordinated operational responses, not as isolated propaganda episodes.
What drives information disorder?
The article’s central point is that information disorder exploits vulnerabilities that already exist rather than appearing from nowhere. It lists six drivers. Cognitive biases, such as prejudice or confirmation bias, make people readier to accept narratives that fit what they already believe. Echo chambers mean content spreads because it resonates, not because it is accurate. Media saturation keeps people constantly exposed to trending narratives, raising the chance they accept them. Social polarisation gives manipulated narratives existing grievances to feed on. Social platforms, the authors say, are built to reward virality over accuracy. Finally, generative AI makes realistic false content cheap to produce, and the article cites the reported withdrawal of a Google Earth AI tool after warnings that it could present manipulated satellite imagery with an appearance of authenticity.
How does information disorder change security assessment?
The authors argue the most consequential shift is an imbalance between how fast information spreads and how fast it can be checked. Traditional open-source intelligence work was mostly about finding enough information; today the problem is volume, and much of it is synthetic. They highlight three forms that strain analysts: false connection (sensational headlines unrelated to the content), misleading content (real material selectively framed, which is harder to spot than an outright lie) and false context (old material recirculated as evidence of a new event). Amplification tactics such as astroturfing, flooding and fogging add to the confusion and can push analysts toward delay or error.
Why are deepfakes a national security concern?
Deepfakes are AI-generated or altered images, video and audio that make someone appear to say or do something they did not. The authors say the danger goes beyond people believing a fake: deepfakes can erode public trust, affect military morale, distort intelligence assessments and influence decisions before authorities can verify them. They cite the manipulated video of Ukrainian President Zelensky appearing to call on his forces to surrender, and research on the May 2025 India–Pakistan conflict finding that manipulated content was used in the contest between strategic narratives. Two analytical problems follow. A fabricated clip may be treated as genuine before it is disproved, while growing awareness of deepfakes creates a ‘false doubt’ problem in which genuine material is dismissed as fake. Proving something is manipulated also does not show who made it or why, which needs separate investigation. The authors add a caution: deepfakes are not universally effective, and their impact depends on timing, audience, credibility and the wider narrative, so a simple edit that confirms existing beliefs can outperform a sophisticated fake that is quickly debunked.
What can OSINT analysts do about it?
The article recommends four measures. First, build and maintain a repository of identified deepfakes and propaganda, recording source, date, type of manipulation and detection indicators. Second, track developments in generative AI to anticipate how new tools may be misused. Third, make deepfake-detection and forensic tools, and AI models for analysing large volumes of text, available to analysts, while treating automated detection as support for human judgement rather than a replacement. Fourth, strengthen contextual analysis and source verification: examine origin, timing, distribution and what may have been left out, and corroborate with multiple independent sources before content enters an assessment. The authors conclude that OSINT remains fundamentally a discipline of critical thinking, where questioning and verifying matters as much as the source used to collect.
Frequently asked questions
What is the difference between misinformation, disinformation and malinformation?
The article groups all three under information disorder. In common usage, misinformation is false content shared without intent to harm, disinformation is false content spread deliberately, and malinformation is genuine information used to cause harm; the authors treat their combined effect as the security problem.
What is the ‘false doubt’ problem with deepfakes?
As awareness of deepfakes grows, genuine footage or documents can be dismissed as fabricated without sufficient evidence, which can delay or distort the assessment of real security incidents.
Can automated tools detect deepfakes reliably on their own?
The authors say automated detection should support, not replace, human analysis, with findings further assessed and verified by analysts.
Read the full article
‘Navigating Information Disorder in the Security Environment’ appears in the September 2026 issue of SecurityLinkIndia (pages 76–80) and includes the full reference list. Read the full article for the authors’ complete argument and sources.
Filed under: Threat Intelligence Special Reports
