AI Prompts for Evaluating the Credibility of Your Texts

Learn how to prompt AI tools to strengthen your credibility. Use these prompts to establish your credibility. Ask multiple generative AI tools to evaluate your voice. Does your text invoke a caring, trustworthy tone? Learn to prompt AI to assess your tone and uncover ethical blind spots.

Table of Contents

At its core, credible means being believable. Communication experts emphasize that a credible writer is perceived as highly caring, trustworthy, and competent (McCroskey & Teven, 1999). Caring writers can demonstrate they are sensitive to the audience’s needs. Trustworthy writers seem moral and genuine; they are attuned to ethical concerns. Competent writers are both intelligent and well-trained. Audiences can believe a writer who demonstrates they are caring, trustworthy, and competent.

If this sounds like a high bar for writers to achieve, it’s for good reason. Credibility can be incredibly powerful. Decades of research demonstrate that credible sources are highly persuasive – they can motivate audiences to pay attention, change opinions, and follow instructions (Pornpitakpan, 2004). Communicating credibility unlocks a writer’s ability to get a crowd on their side. A prize like that should probably not be too easily won.

Trying to communicate caring, trustworthiness, and competence in addition to writing clearly and correctly may feel overwhelming. Of course, limited space is also an ever-present concern for writers. Say you have 500 words to express your ideas. How many of them can you afford to use building credibility?

Savvy writers can leverage generative artificial intelligence to enhance their credibility efficiently. Large language models such as Copilot, Gemini, or GPT-5 can help evaluate tone and identify blind spots – as long as they are used wisely.

Table of Contents

Related Concepts

Prompting to Assess Tone

An audience begins determining a writer’s credibility using tone. Tone represents the writer’s attitude as conveyed through choices related to context, medium, audience, and purpose (Alred et al., 2012). When writing conveys respect for the audience and aligns with their needs, it signals caring. While writers should not rely wholly on GenAI to craft their writing’s tone, it can be a useful way to double check that a message is giving the desired impression.

While some tools will assess tone automatically, a more strategic prompt helps GenAI assess tone like a writer. AI experts recommend designing prompts that include a clear description, context, and examples (Cook, 2023). Try this template for prompting GenAI: “Describe the text’s tone. The message is (context) shared as (medium) to (audience) for (purpose).” For example, the prompt for GenAI to assess the tone of this Writing Commons entry might read,

“Describe the text’s tone. The message is a writing encyclopedia entry shared as an open access text to college-level students and faculty to inform about AI best practices.”

Writers should compare GenAI’s description of tone with the intended goal of their writing. If the two align, the message is on track. If not, they can ask GenAI to highlight specific sentences that reveal the tone and focus on revising those sections.

Probing for Blind Spots

A writer can build trustworthiness by demonstrating a keen awareness of ethical concerns. According to professors Bazerman and Tenbrunsel (2011), people are commonly unaware of the limits to their own morality. That is, we may struggle to identify our own biases, preconceptions, or conflicts of interest. They call these blind spots – ethical concerns that we struggle to check. For example, a writer may not realize that a saying like “We need a Hail Mary” – an American football metaphor for a desperate, last-ditch attempt – may be confusing to international audiences. While ethical blind spots may be common, credible writers should self-reflect and consider unintended implications of their communication.

When specifically prompted, GenAI can help illuminate blind spots to avoid confusing or alienating the audience. Thanks to its large language model training, GenAI excellent at thinking from multiple perspectives. When writing for a wide or diverse audience, consider naming the different audience segments in the prompt. For example:

Identify possible ethical blind spots in this message, such as biases, assumptions, or conflicts of interest. Explain how these blind spots might affect different audiences (e.g., faculty and college students) and suggest ways I could revise the message to strengthen its credibility.

Generative AI may have helpful suggestions to revise the message, but the writer is ultimately responsible for the ethical implications of the writing. Ethics is most effective when treated as a guide for thoughtful action, not as a rigid checklist of rules. A trustworthy writer is also genuine – adopting GenAI suggestions without self-reflection limits the writing’s authenticity.

Avoiding Overreliance

While GenAI can offer feedback on tone and reveal ethical blind spots, using it inappropriately can also harm credibility. Most tools label GenAI output with some version of “AI can make mistakes.” For a writer, the issue is not just that GenAI writing may be incorrect – it can also undermine competence. Presenting a smart, well-written message is essential. A writer can seem less credible once readers begin to suspect that AI produced the message instead of the writer’s own skill.

Readers are becoming more savvy about detecting AI writing. AI’s writing style often prefers dashes over commas and uses triplet framing, or repetition in threes (Fink, 2025). A recent study found that AI-written news articles were seen as less credible and received less positive engagement (Lee et al., 2025). Human readers still want human writers. Writers can project competence and avoid being mistaken for AI by starting from their own ideas and using GenAI to offer suggestions for revision.

Staying Believable

As artificial intelligence evolves, credibility remains an important asset in the writer’s toolkit. Audiences seek caring, trustworthiness, and competence, which cannot be outsourced.

Generative AI can help writers refine tone and uncover blind spots to build trust. However, writers should be intentional and strategic in inviting GenAI to collaborate and give feedback. Asking GenAI for suggested revisions can become a slippery slope, making it harder to preserve the writer’s unique voice. Critically considering any revisions before adopting them can help writers stay believable while appropriately engaging generative AI technologies.

Share this post:

References

Alred, G. J., Brusaw, C. T., & Oliu, W. E. (2012). Handbook of technical writing (Tenth edition.). St. Martin’s.

Bazerman, M. H., & Tenbrunsel, A. E. (2011). Blind spots: Why we fail to do what’s right and what to do about it. Princeton University Press. https://doi.org/10.1515/9781400837991

Cook, J. (2023, June 26). How to write effective prompts for ChatGPT: 7 Essential steps for best results. Forbes. https://www.forbes.com/sites/jodiecook/2023/06/26/how-to-write-effective-prompts-for-chatgpt-7-essential-steps-for-best-results/?sh=5a76b51e2a18

Fink, C. (2025, June 12). The seven deadly tells of AI writing. Forbes. https://www.forbes.com/sites/charliefink/2025/06/12/the-seven-tells-of-ai-writing/

Lee, D. C., Jhang, J., & Baek, T. H. (2025). AI-generated news content: The impact of AI writer identity and perceived AI human-likeness. International Journal of Human–Computer Interaction, 41(21), 13862–13874. https://doi.org/10.1080/10447318.2025.2477739

McCroskey, J. C., & Teven, J. J. (1999). Goodwill: A reexamination of the construct and its measurement. Communication Monographs, 66, 90-103. https://doi.org/10.1080/03637759909376464

Pornpitakpan, C. (2004). The persuasiveness of source credibility: A critical review of five decades’ evidence. Journal of Applied Social Psychology, 34: 243-281. https://doi.org/10.1111/j.1559-1816.2004.tb02547.x