Social media platforms amplify personal interpretations and half-truths, often overshadowing facts with viral narratives. Against this backdrop, a group of Portuguese university students from the Instituto Superior Técnico (IST) has developed an Artificial Intelligence system designed to accelerate fact-checking and help restore trust in information.

The new system replicates the full fact-checking process normally conducted by professional journalists, condensing hours of work into a much shorter timeframe. Users need only provide the statement to be checked, and the system automatically collects relevant evidence, identifies supporting and contradicting information, and produces a verdict accompanied by an explanatory text. By doing so, it offers a vital tool for addressing the scale of today’s disinformation problem, where falsehoods often spread faster than corrections can be published.

Disinformation thrives at speed. Viral posts can reach millions within hours, while traditional fact-checking can take days. By the time a correction is issued, the false narrative has often solidified into perceived truth. Automated verification aims to bridge this gap, providing journalists with the means to respond quickly and decisively before misinformation takes root. The project has already been assessed with the input of hundreds of journalists, ensuring that its results are practical and relevant for real-world media environments.

At the core of this initiative are large language models, which generate datasets for fact checking tasks and simulate the reasoning steps behind verification. The approach does not eliminate the role of human judgment, on the contrary, it supports it. Journalists remain essential to interpreting results, applying critical thinking, and ensuring that credibility is preserved. What the system does is reduce the time burden, enabling human professionals to act at the speed that the digital ecosystem demands.

The project is part of a wider effort to explore new forms of explanation and personalization in how verified information is delivered. Studies show that younger generations are particularly vulnerable to false narratives, relying heavily on platforms such as Instagram and TikTok for news. Traditional fact-checking methods often fail to capture their attention. For this audience, adapting language, format, and delivery is crucial. Personalization of fact checks may prove key to reaching them before misinformation hardens into belief.

At the same time, the initiative highlights the delicate balance between technology and education. While AI can streamline fact-checking, there are growing concerns about its impact on human cognition. Overreliance on automated tools can erode critical thinking skills, leading to what researchers describe as cognitive debt. For this reason, AI must be seen as a support mechanism rather than a replacement, while media literacy and critical education remain vital pillars in the fight against falsehoods.

Ultimately, this innovation underscores how urgent the challenge of disinformation has become. In a world where personal truths compete for attention, often disconnected from reality, the ability to verify facts quickly and transparently is no longer a luxury; it is a democratic necessity. The Portuguese students’ work demonstrates that technology, when responsibly designed and applied, can empower journalists, protect citizens, and bring society one step closer to a shared foundation of truth.

Author

Paulo Lopes is a multi-talent Portuguese citizen who made his Master of Economics in Switzerland and studied law at Lusófona in Lisbon - CEO of Casaiberia in Lisbon and Algarve.

Paulo Lopes