Within this article research that was conducted at University of Glaslow to predict what could be the next likely virus to jump to humans from animals using machine learning models using viral genomes. Most of the viruses that are known today such as COVID-19, the advantage of being able to possibly detect newer viruses early can help improve research and surveillance priorities. The machine learning models demonstrated that there are a large amount of zoonotic related viruses can be inferred to a great extent just from genome sequences that can take a toll on humans. Since most viruses are caused by animals there could be a huge advantage in gaining early preparedness for these viruses to come. Other viruses that are discovered but don't use genomic sequencing however are not very good with giving phenotypic data, an approach that would be ideal would be able to quantify relative risk of human inefectivity when having relevant exposure from just sequencing data. In order to get a more accurate machine rather than this one and use viral genomes they assigned a percentage of human infection based on virus taxonomy and/or relatedness to known human-infecting viruses, then used the best-performing model to look for trends in the expected zoonotic potential of other virus genomes from a variety of species. Researchers discovered that viral genomes may include generalizable traits that are independent of virus taxonomic relations, and that these properties may allow viruses to pre-adapt to infect people. Using viral genomes, the team was able to construct machine learning models capable of detecting possible zoonoses. Computer models are merely one stage in the process of detecting zoonotic viruses that might infect people. The effectiveness of the models demonstrates how increasingly common and low-cost genome sequence data may guide early-stage decisions on viral research and surveillance priorities with no additional financial or time commitment.