Multimodal Outputs for the Workplace From Generative AI: Effective or Not?

Abstract

Workplace documents are a combination of genre (form) and topic (content), and there are a variety of standards they must meet to be usable. Most such files are also multimodal, consisting of both text and still imagery. When outputting multimodal digital works for the workplace from multimodal generative AI (GAI) tools, their usability depends on various factors. This work explores some elicited multimodal outputs (text + imagery) from a popular multimodal generative AI tool to assess the respective quality based on practical dimensions. This work offers an early assessment for just how useful multimodal generative AIs are for this broad use case. And this work offers a checklist of factors to evaluate for multimodal file output quality for the workplace. Some initial observations are made about the practical usability of the multimodal GAI for outputs for professional workplace usage, based on this light prompt-response-analysis exploration.

Key Terms in this Chapter

Multimodal: Consisting of several digital content modalities (text, still image, video, audio, and others, in various combinations).

Generative AI: Computational tools that can create (or autogenerate) digital contents in various modalities based on text and / or image and other prompts by human users [based on large language models and deep learning from a large dataset of examples].