Why AI disclosure does not establish message authenticity

AI disclosure answers one question about a message

A disclosure that AI helped produce a customer message answers a useful question about authorship. Customers may also want to know who sent it and whether they can return to it later.

In an Exclaimer survey, 46% of UK adults said they had questioned whether a message they received was genuine or legitimate. Exclaimer surveyed 2,000 UK and US adults, including 1,000 UK adults that the company describes as nationally representative by age, gender and region.

Exclaimer sells email signature management software. Its findings therefore come from a company that can benefit when businesses invest in clearer email identity and signature practices.

Authenticity has more than one layer

Authorship and provenance answer different questions. Provenance means evidence about where a communication came from. AI disclosure can tell a recipient that AI participated in an interaction, while sender details and the communication channel can help establish its origin.

The Exclaimer findings show that UK respondents pay attention to several such signals. Among UK respondents, 53% said the platform on which a message arrived affected how much they trusted it. Another 47% said the platform affected whether the communication felt authentic.

Within email, respondents cited several credibility signals:

Ed Bodey, general counsel at Exclaimer, argues that consumers notice details such as the correct company domain or a strong signature. He says businesses should continue prioritizing communications trust as regulation changes. Exclaimer has a commercial interest in that recommendation because it sells email signature management software.

Recipients value records they can return to

Channel choice also reflects whether a communication can be preserved. Among UK adults, 54% said they had kept a message to refer to it later. Another 44% said they had deliberately chosen email over another platform to create a permanent record.

Preferences differed by situation:

These figures connect communications design to later use. A formal complaint, employer update, or healthcare information may need to be retrieved after the first interaction.

The results show reported preferences and behaviors. They do not demonstrate that AI caused those choices. Businesses can use them as evidence about channel and record preferences without treating them as evidence of AI-driven distrust.

Age groups report different patterns

Reported AI use varies sharply by age. Reported concern about authenticity also differs by age:

The age bands for AI use and authenticity concern differ, so these figures cannot establish a causal relationship between AI familiarity and suspicion.

For executives, the practical point is narrower. Customers of different ages report different levels of AI use and concern about whether messages are genuine. Communications teams should account for those differences when deciding how clearly a message identifies its sender.

AI can change interpersonal signaling

AI can also affect how people phrase difficult messages. Exclaimer reports that almost a quarter of Gen Z respondents use AI to soften a difficult message, while 16% use it to avoid an awkward conversation. Among baby boomers, Exclaimer reports corresponding figures of 4% and 2%.

Those uses show how AI can mediate tone when a sender’s confidence, discomfort or intent may shape the wording. Recipients may therefore care whether AI participated in producing a message.

Disclosure addresses that participation. Sender identity and channel cues address the communication’s origin. Businesses that use both give recipients different information for different questions.

Design AI disclosure and verification together

Executives can treat AI disclosure and message verification as related design concerns. The survey findings support attention to the platform carrying a message, the company email address, contact details, sender name, and the ability to retain a communication.

That has practical implications for customer and employee communications. Teams can make AI involvement clear where their policies require it, preserve recognizable sender information, and choose channels suited to communications that recipients may need to revisit.

For CTOs, CEOs, CX leaders, and communications teams, the design question is concrete: can the recipient understand AI’s role, identify who sent the message, and keep the communication when a durable record matters?

Main highlights

• AI disclosure addresses only part of trust: Disclosure explains AI’s role, but customers also use sender identity, platform and contact details to assess whether a message is legitimate. Leaders should design AI transparency and verification together.
• Durable communication channels matter: UK respondents often prefer email for formal or important communications and value messages they can retain. Choose channels based on whether recipients may need a permanent record.
• Age affects communication expectations: AI use and concerns about message authenticity vary substantially across age groups. Communications teams should avoid assuming that all customers interpret AI involvement and authenticity signals in the same way.
• AI changes interpersonal signaling: Some consumers use AI to soften difficult messages or avoid awkward conversations, particularly among younger respondents. Clear disclosure can help recipients understand when AI has influenced how a message was phrased.
• Verification should complement AI transparency: Make AI involvement clear where required while preserving recognizable sender names, company domains and contact information. This gives recipients clearer evidence of both how a message was created and where it came from.