Artificial intelligence has rapidly accelerated many areas of research, aiding in target validation and the visualisation of as yet uncharacterised proteins. Previous studies have used AI to generate mini-protein binders, which have the potential to act as antivirals and can be used improve our understanding of mechanisms underpinning viral infection, however current methodologies significantly limit the number which can be assessed. This project aimed to develop alternative methods of screening, allowing for greater throughput.   Here, we used generative AI protein design to create a panel of mini-protein binders targeting the E1E2 glycoproteins of hepatitis C virus (HCV). The binders were screened, by inclusion in a pseudo-virus (PV) assay; whereby binder-expressing plasmids were co-transfected in HEK293T cells alongside E1E2, HIV gagpol and luciferase reporter. From this screen three hits were identified. To confirm interaction with E1E2, binders were pulled down (via strep tag). Western blotting of co-preciptated E1 and E2 demonstrated that binder activity correlated with binder-target interactions. Further experiments using alternative viral glycoproteins were conducted to determine the specificity of the binders’ action. All three binders were found to be specific to HCV. Moreover, screening against HCV E1E2 variants demonstrated that one binder achieved broad inhibition of genetically distant HCV clones. Notably, the hits identified in the initial screen were corroborated using alternative AI prediction tools (e.g. AlpahFold3) suggesting that computational screening could be used to pre-screen prior to experimental validation.   This approach has the potential to allow researchers to rapidly assess binders targeting a wide range of viral glycoproteins.