A Brief History of Adverse Media Screening

Adverse media screening dates back several decades and has evolved alongside the changing media landscape.

In the early days of journalism, the only sources of adverse media were traditional print newspapers and magazines. Screening for negative information was a time-consuming and labour-intensive process.

Today, adverse media can be found in a wide range of sources, including online news articles, blogs, social media, and other publicly accessible data sets, and artificial intelligence (AI) has revolutionised the process of adverse media screening.

AI algorithms, specifically natural language processing, can be used to analyse large volumes of unstructured data, identify relevant information, and automatically categorise and classify said data based on predefined criteria. This allows for a more efficient and effective process of adverse media checks, covering a wider range of sources and reducing the risk of human error.

The history of negative news screening reflects the evolution of the media landscape and the growing need for more sophisticated and automated methods of identifying negative information.

Due diligence processes in the past

The first wave of adverse media screening tools filtered down media results by looking for adverse keywords like ‘crime’ or ‘prison’ within an article where the searched subject, be it an individual or company, was mentioned. This however created too much noise and would surface a lot of documents for humans to read even if the adverse terms were not connected in any way to the searched subject. For instance, an article about the searched subject ‘John Smith’ would be found even if in the article the offending adverse snippet was about a completely different person: “Jane Doe has been arrested for money laundering”.

To try and fix issues like this, the next wave of adverse media screening software used a proximity search, whereby negative news articles or snippets were only flagged if the name of the searched subject was mentioned within a certain word distance from an adverse term. This in itself didn’t completely fix the problem of too many irrelevant articles and in fact created a wholly new problem of actually missing certain red flags entirely.

If searching ‘John Smith’, the following snippet would be incorrectly flagged as adverse about him even though it relates to someone entirely different, as the system matched his name in proximity to an adverse term: “The Judge John Smith has sentenced Jane Doe to 12 years in prison for money laundering”.

Somewhat conversely, the proximity searches would also miss important adverse information as they did not account for pronouns, anaphoras and cataphoras. So they would miss articles that would include snippets like “He was also arrested for fraud in 1999”, where the searched subject’s name is not even mentioned within the same sentence as the adverse term and is at a word distance beyond the proximities settings.

Adverse media screening today

Just searching for adverse keywords is a thing of the past. For a truly effective and automated adverse media screening solution, the only answer is to use NLP technology where it can decipher, analyse, and emulate human language in a contextually relevant and purposeful manner and solve the issues as raised above.

The new wave of adverse media screening software, like smartKYC, relies on using sophisticated natural language processing techniques to understand and contextualise unstructured data like online media so that the search is thorough, accurate and only pertinent and risk relevant facts are presented for human review, whilst not missing any red flags.

See for yourself – book your demo today.

Discover smartKYC

smartKYC is a multi award winning risk profiling solution that uses artificial intelligence to fully automate Know Your Customer (KYC) screening processes on all required data sources, at every stage of the client lifecycle, be it pre-KYC, onboarding, periodic refresh, or continuous monitoring.

Combining a federated search with multilingual natural language processing, sophisticated name matching and unique identity matching, smartKYC is faster, better and more cost effective than any other solution on the market, book your demo today.