Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Correspondence
  • Published:

Judging the creative prowess of AI

Nature Machine Intelligence volume 5, page 558 (2023)Cite this article

Subjects

Until recently, ‘generative model’ was a specialized term used only by the artificial intelligence (AI) research community, but this changed with the release of content-generating AI programs such as DALL-E1, Make-A-Video2, ChatGPT and MusicLM3. These generative models use prompting techniques4 to compose seemingly realistic content. Prompts, as the name suggests, are initial instructions provided colloquially to nudge these models to produce an output. Many users worldwide are now interacting with these systems daily and are excited about their creative capabilities.

The output of a generative AI model can be formally vetted when the prompt focuses on objective topics. For example, when the system is asked ‘what is Avogadro’s number?’, the accuracy of the output can be measured in a straightforward way. But with the prompt ‘draw an image of a horse riding in space’, how should the correctness of the output by judged? There may be technical aspects associated with a creative work, but such markers contribute only partially to the judgement of creativity. Evaluating a creative element is difficult; this is true for any technical or non-technical work produced by humans, and may be even more so for content composed via generative AI models.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

USD 39.95

Prices may be subject to local taxes which are calculated during checkout

References

  1. Ramesh, A., Dhariwal, P., Nichol, A., Chu, C. & Chen, M. Preprint at https://arxiv.org/abs/2204.06125 (2022).

  2. Singer, U. et al. Preprint at https://arxiv.org/abs/2209.14792 (2022).

  3. Agostinelli, A. et al. Preprint at https://arxiv.org/abs/2301.11325 (2023).

  4. Dang, H., Mecke, L., Lehmann, F., Goller, S. & Buschek, D. Preprint at https://arxiv.org/abs/2209.01390 (2022).

  5. Kaufman, J.C. & Glãveanu, V.P. in Cambridge Handbooks in Psychology 2nd edn. (eds. Kaufman, J. C. & Sternberg, R. J. E.) 27–43 (Cambridge Univ. Press, 2019).

  6. Glãveanu, V. P. Rev. Gen. Psychol. 17, 69–81 (2013).

    Article  Google Scholar 

  7. Feist, G. J. Curr. Opin. Behav. Sci. 27, 31–35 (2019).

    Article  Google Scholar 

  8. Franceschelli, G. & Musolesi, M. Preprint at https://arxiv.org/abs/2104.02726 (2021).

  9. Rohrmeier, M. Transactions of the International Society for Music Information Retrieval Vol. 5, 50–66 (Ubiquity Press, 2022).

Download references

Author information

Authors and Affiliations

  1. Department of Electrical Engineering, Indian Institute of Technology Delhi, New Delhi, India

    Tanmoy Chakraborty

  2. Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi, New Delhi, India

    Tanmoy Chakraborty

  3. Department of Computer Science & Engineering, Indraprastha Institute of Information Technology Delhi, New Delhi, India

    Sarah Masud

Authors
  1. Tanmoy Chakraborty
    View author publications

    Search author on:PubMed Google Scholar

  2. Sarah Masud
    View author publications

    Search author on:PubMed Google Scholar

Contributions

T.C. and S.M. wrote the paper.

Corresponding author

Correspondence to Tanmoy Chakraborty.

Ethics declarations

Competing interests

The authors declare no competing interests.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chakraborty, T., Masud, S. Judging the creative prowess of AI. Nat Mach Intell 5, 558 (2023). https://doi.org/10.1038/s42256-023-00664-y

Download citation

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s42256-023-00664-y

This article is cited by

Search

Advanced search

Quick links

Nature Briefing AI and Robotics

Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing: AI and Robotics