CAPABILITIES OF AI IN TEXT-TO-3D MODEL GENERATORS FOR MECHANICAL COMPONENTS DESIGN

This paper explores how effectively Artificial Intelligence (AI) can translate text descriptions into accurate and functional 3D models of mechanical components, highlighting both the potential and the current shortcomings of this technology. A functional 3D model implies that it is suitable for 3D printing. Three different AI 3D model generators were selected for this analysis. Text prompts involving gear mechanisms were used as input. Each generated model was assessed on key parameters such as the accurate depiction of gear teeth, the correct alignment and meshing of gears, and the overall structural integrity required for functional gear mechanisms. Additionally, the generated models were verified for any deviations from the expected design standards of mechanical gears. The analysis revealed that none of the selected AI 3D model generators could consistently produce fully functional and technically accurate gear models based on simple text prompts. While the AI generators could produce visually correct models that generally resembled gears, significant shortcomings were observed in their functional and technical accuracy. The generated models often failed to meet the precise geometric and structural requirements necessary for practical application in mechanical systems. Automating the 3D modelling process for mechanical components using AI technology would reduce time and cost. However, further advancements in AI and machine learning are required to enhance the accuracy and reliability of AI-generated 3D models for use in real-world mechanical design and manufacturing