Wadds Inc. | Professional advisor to agencies & comms teams

View Original

How to use AI to make sense of academic papers

Using large language models like Claude 2 to extract information from academic papers.

Making sense of a research paper and extracting information typically takes half a day. Parsing a paper using the Anthropic Claude 2 large language model and asking the AI bot for a summary, key arguments, and the method takes 15 minutes.

I’ve included an example below from a recent paper published in the Journal of Public Relations Research.

Generative AI gets much attention, but reductive applications such as this could potentially democratise and accelerate the creation of new knowledge. It could also address systemic knowledge gaps.

Once a paper has been uploaded into a large language model, it can be integrated from various stakeholder perspectives or personas.

There’s a long-standing management and public relations issue between theory and practice. Research papers are expensive and inaccessible to practitioners. No longer.

Claude 2 can be used as a powerful bridge between theory and practice. However, I suspect we’ll see publishers move quickly to block these tools as they could potentially have a destructive impact on business models.

Rethinking crisis communication advice when stakeholders experience moral outrage

Reference

W. Timothy Coombs & Elina R. Tachkova (2023) How Emotions Can Enhance Crisis Communication: Theorizing Around Moral Outrage, Journal of Public Relations Research, https://doi.org/10.1080/1062726X.2023.2244615

Summary

Certain "sticky crises" evoke strong moral outrage and require an emotion-driven crisis communication approach; this paper develops propositions to advance theory and practice regarding moral outrage's role in shaping crisis communication during these challenging situations.

Key arguments

Certain crises, termed "sticky crises", are more challenging to manage due to strong emotions like moral outrage evoked in stakeholders.

Stakeholders assess crises through both cognitive and emotional processes. High emotional intensity leads to more emotion-oriented assessments.

  • Moral outrage is a strong negative emotion tied to perceptions of injustice and greed. It fundamentally reshapes crises.

  • Moral outrage-inducing crises merit special attention as a sticky crisis. Moral outrage changes how stakeholders perceive and react to crises.

  • Existing crisis communication advice may not apply well to moral outrage-inducing crises. New guidance is needed to help managers navigate these crises.

  • A new theory-driven approach focused on emotions like moral outrage can enhance crisis communication for moral outrage-inducing crises.

  • Propositions for developing theory and practice regarding communication in moral outrage-inducing crises are outlined.

Method

The methodology is a conceptual analysis and synthesis of prior literature to develop propositions and outline a theory-driven research program on the role of moral outrage in crisis communication during sticky crises. It does not include collection or analysis of any primary data.