Consumers open to AI, demand stronger safeguards

Only 32% are likely to let AI make decisions independently.

Consumers are open to giving artificial intelligence (AI) greater responsibilities but want stronger safeguards around personal data, transparency, and human oversight, according to a Ping Identity survey.

The findings were based on responses from 1,000 consumers in Singapore.

Only 32% said they were likely to allow AI to make decisions independently on their behalf.

About 45% said medium-stakes recommendations represented the highest level of responsibility they were comfortable giving AI. These included comparing products or travel options and helping users evaluate a decision.

Another 10% were not comfortable handing AI any responsibility.

Concerns were particularly strong when AI agents acted on users’ behalf. Around 81% worried that an AI agent could misrepresent their personality or personal values.

About 62% were more concerned about having to constantly double-check an AI agent’s work than about the AI making a mistake.

Consumers were evenly divided on the importance of disclosure. Some 41% said knowing when AI was being used was most important, whilst another 41% prioritised the fastest, most accurate result regardless of disclosure.

Despite their concerns, 41% said they would be likely to give an AI agent read access to their primary email if doing so saved them five hours a week.

Data protection emerged as a central condition for trust. Half of respondents said personal information used in an AI interaction must remain strictly confidential.

Another 40% said guarantees that their data would not be used to train public AI models would make an AI tool or company more trustworthy.

Meanwhile, 36% said having a clear and easy way to escalate an issue to a person would improve trust.

Government oversight also received broad support, with 84% of Singapore respondents describing regulation protecting identity data in AI applications as important.