Over Half of Consumers Prepared to Take ‘Critical Life Advice’ From AI

Despite it being an emerging technology with proven flaws, artificial intelligence has crossed the ’trust threshold,' a new survey says.

This illustration photograph shows screens displaying the logo of DeepSeek, a Chinese artificial intelligence company that develops open-source large language models, and the logo of OpenAI's artificial intelligence chatbot ChatGPT in Toulouse, southwestern France, on Jan. 29, 2025. Lionel Bonaventure/AFP via Getty Images

Over half of consumers have said they would trust artificial intelligence (AI) to provide them with “critical life advice.”

This is despite the fact that AI continues to be plagued by issues such as bias, discrimination, opaqueness of their decision-making processes and “hallucinations”—where it comes up with an answer that is wrong, nonsensical, or misleading, despite appearing plausible.

The 2025 Smart Communications global survey of over 3,000 people, including Australians, found that around half would be willing to take advice from AI on their finances (46 percent), insurance plan changes (51 percent), or follow its health recommendations (54 percent).

Additionally, most people are confident that AI will give them the correct answer—only 39 percent expressed concerns about potential errors.

A significant shift has occurred in the number of people who believe that the use of AI in customer communications should be disclosed.

In 2024, 77 percent said they'd want to know they were talking or texting with an AI chatbot; this year, that figure fell to 37 percent.

There remain concerns about a lack of human oversight (44 percent cited this as an issue) and data security (50 percent).

People in the Asia-Pacific region have greater confidence that AI will have significantly improved by 2030, with 71 percent in the region agreeing versus 67 percent in the US, 49 percent among German-speaking people, and 46 percent of those in the UK.

As might be expected, the younger the respondent’s age, the more optimistic they were about AI’s future.

Just 49 percent of baby Boomers think it will get better, versus 72 percent of Gen Z and Millennials. Gen Xers ranked between the two, at 58 percent optimistic.

There were also significant gender differences, with 48 percent of men but only 38 percent of women confident that AI has no ethical issues, with 52 percent of men neutral or hesitant about it, versus 62 percent of women.

Consumers’ confidence in AI’s accuracy is at variance with numerous scientific studies, such as a recent one by the Columbia Journalism Review (CJR) which gave eight AI tools an excerpt of an article and asked them to identify the “corresponding article’s headline, original publisher, publication date, and URL.”

Collectively, the study found the chatbots “provided incorrect answers to more than 60 percent of queries.”

“Most of the tools we tested presented inaccurate answers with alarming confidence, rarely using qualifying phrases such as ‘it appears,’ ‘it’s possible,’ or ’might,‘ ... or acknowledging knowledge gaps with statements like ’I couldn’t locate the exact article,'” CJR noted.

When OpenAI released its latest update, ChatGPT-4.5, the company admitted it has a hallucination rate of 37.1 percent for its SimpleQA test, which asks it short, fact-based questions.

The company’s CEO, Sam Altman, has previously said hallucinating is more of a feature of AI than a bug, adding, “a lot of value from these systems is heavily related to the fact that they do hallucinate.”