Trusting the Unknown: Navigating Misinformation in Global Risk Intelligence
In an era of instantaneous information, the hardest question facing security and risk professionals is no longer where to find information — it’s whether to trust it.
Imagine receiving a breaking news alert: reports of an attack in the Central African Republic, where your organization has staff. Or a coup attempt in a country where you have active operations. The news is fast-moving, the sources are few, and the pressure to act is immediate. This is the reality confronting risk professionals every day, and it is getting harder, not easier, to navigate.
The Anatomy of Misinformation
Misinformation doesn’t emerge fully formed from a single bad actor. It follows a deliberate, structured process, and one that is becoming increasingly sophisticated and harder to detect.
State-linked organizations, or well-funded bad actors, create a narrative and push it through their own controlled channels. At this stage, most experienced analysts will dismiss it. A Russian state media outlet or an obviously Iran-linked Telegram channel carries little credibility on its own.
But that’s precisely the point. The goal isn’t to persuade the skeptics; it’s to launder the information until its origin is obscured. The narrative gets tailored to regional audiences, seeded into fringe networks, and gradually absorbed into more mainstream outlets. By the time it appears on a platform that looks credible, its origins are difficult to trace.
A recent example illustrates this vividly. Iranian state media published a story claiming a drone attack on a US aircraft carrier. Within hours, the story had migrated through pro-Iranian Telegram channels, onto misinformation-for-hire accounts on X, and eventually onto the YouTube channel of one of India’s largest newspapers by circulation. Along the way, AI-generated imagery of a ship explosion was added to make the story visually compelling and emotionally convincing. The entire chain, from fabrication to mainstream publication, unfolded in a matter of hours.
Quality Over Quantity in Source Intelligence
One of the most persistent misconceptions in intelligence is that more sources equals better intelligence. It doesn’t. What matters is the quality and structure of those sources.
At Seerist, we evaluate the reliability of sources across four areas: the proportion of factual versus opinion-based reporting; factual accuracy over time and across geographies; the use of emotive language designed to steer interpretation rather than inform; and resilience — how well a source maintains its editorial standards when information volumes spike during a crisis.
That last dimension is frequently overlooked. A newspaper may be an excellent source on any given day, but during election season — when its newsroom is overwhelmed — its standards may slip significantly. Similarly, a news outlet may be highly reliable for local crime reporting but far less so when covering international events. Smaller outlets often rely on wire services of wildly varying credibility, for example mixing content from Reuters and AFP with RT and Sputnik.
The lesson: source reliability is not binary, and it is not static. It is contextual, topic-specific, and subject to change.
The Human Element Is Non-Negotiable
Artificial intelligence has a meaningful role to play in evaluating sources at scale. At Seerist, we use trained models to analyze source content, suggest reliability ratings, and challenge analyst assessments. But the final judgment always rests with a human analyst who must explain and justify their rating.
That accountability matters most when the information environment offers no clean answers. And that cuts both ways. In heavily restricted environments where press freedom is limited and journalists face intimidation, an automated or unstructured approach to sourcing would surface low-quality sources without flagging their limitations. In dense media environments, the opposite problem applies: an undiscriminating approach risks drowning useful signals in low-quality noise. The solution in either case is to be transparent about confidence levels and to clearly segregate lower-quality intelligence from the strongest available sources.
A Difficult Road Ahead
The outlook for the global information environment is sobering. Newsrooms are shrinking. Field reporters are being laid off. Advertising revenues are collapsing. And the tools to produce convincing misinformation, including AI-generated imagery and video, are now accessible to almost anyone, with results that can be produced in minutes.
The organizations that will navigate this environment successfully are those that invest in the unglamorous, painstaking work of source investigation: understanding who owns a media outlet, how it is funded, what its editorial processes are, and where its blind spots lie. In a world where trust is eroding, the competitive advantage belongs to those who have done the work to know what and whom they can actually trust.
