Human-Written Crisis Statements Win More Trust Than AI, Study Shows

Human-Written Crisis Messages Seen as More Credible Than AI

Artificial intelligence is becoming a common tool in corporate communication. However, new research suggests AI might not be the best fit for handling sensitive crisis situations. A recent study published in Corporate Communications: An International Journal reveals that crisis messages labeled as human-written are perceived as more credible and better for a company’s reputation than identical messages attributed to AI.

How Trust Was Tested in Crisis Messages

The study created an experiment around a fictional company called Chunky Chocolate, which faced backlash after reports linked its products to customer illness. Participants read one of six press releases, all containing the same content but varying in two aspects: whether the message was said to be written by a human or AI, and whether the tone was informational, sympathetic, or apologetic.

The participants were 447 journalism and communication students from a Midwestern university. They rated each message’s credibility, the credibility of its source, and the impact on the company’s reputation.

Human Authorship Boosts Credibility and Reputation

Clear differences emerged between human and AI labels. On a seven-point scale, messages marked as human-written scored higher in every category:

• Source credibility: 4.40 (human) vs. 4.11 (AI)
• Message credibility: 4.82 (human) vs. 4.38 (AI)
• Company reputation: 4.84 (human) vs. 4.49 (AI)

Since the content was identical, these differences were entirely due to the perceived authorship. Labeling a crisis response as AI-generated lowered trust, regardless of the message's wording.

Tone Matters Less Than Who’s Behind the Message

Researchers expected an apologetic or sympathetic tone to influence trust. While participants did notice tonal differences, these did not significantly change credibility or reputation scores. The identity of the communicator had a stronger impact on how the message was received.

Implications for Public Relations Professionals

AI is already widely used in PR for tasks like media monitoring, content targeting, and social media management. Some suggest expanding AI’s role into drafting press releases or crisis responses. This study advises caution: audiences may be less trusting of crisis communications when AI authorship is disclosed.

For PR practitioners, this means carefully considering when and how to use AI-generated content—especially in situations where trust is critical.

Study Limitations to Keep in Mind

The experiment involved a fictional company and a student sample, which might not fully reflect broader public opinion. Participants' familiarity with AI and digital tools could also influence their responses. Additionally, the study explicitly labeled messages as AI-generated, which isn’t always the case in real-world communications.

Despite these limits, the findings highlight the continuing importance of human credibility in crisis communication.

For those interested in enhancing their skills with AI tools while maintaining effective communication strategies, exploring targeted AI courses can be valuable. Visit Complete AI Training for relevant resources.