A brand new KPMG report about synthetic intelligence discovered {that a} staggering 75% of respondents within the US are deeply involved about doable unfavorable outcomes because of AI, whereas 45% consider the dangers outweigh the advantages. Additional, 55% of People report having skilled inaccurate outcomes from AI.
As well as, 52% of US-based respondents expressed fear that AI-generated content material is manipulating elections, and an awesome 85% are calling for legal guidelines and actions to fight AI-generated misinformation.
Shoppers need extra AI regulation
In line with KPMG’s report, “Belief, Attitudes and Use of Synthetic Intelligence: A International Research 2025,” solely 29% of US customers consider present rules are enough for AI security, and 72% say extra regulation is required. If legal guidelines and insurance policies have been in place, 81% of US customers can be extra keen to belief AI techniques, the report famous.
But, 43% of US customers have low confidence within the authorities to develop and use AI, preferring to place their belief in universities, analysis establishments, healthcare suppliers, and large tech firms to develop and use AI within the public’s greatest pursuits.
Enthusiasm for AI within the office – and misuse
Within the office, 69% of organizations are utilizing AI, whereas 67% of staff are enhancing their productiveness with the tech.
Nevertheless, there’s a skinny line between acceptable use and misuse: 44% of staff report utilizing AI in inappropriate methods, akin to importing delicate firm data and mental property to AI platforms (46%), violating insurance policies and placing their group in danger.
Fifty-seven p.c of US staff surveyed admit to creating errors of their work because of AI errors. And, 53% of workers have offered AI-generated content material as their very own.
These points spotlight “a major hole in governance and lift severe issues about transparency, moral habits, and the accuracy of AI-generated content material,” stated Samantha Gloede, trusted enterprise chief at KPMG, in an announcement. “This ought to be a wake-up name for employers to offer complete AI coaching to not solely handle dangers but additionally to take care of belief.”
Want for AI to be extra reliable
The blended response towards AI reveals a rising divide between its advantages and its risks. Whilst AI demonstrates its energy to revolutionize industries, there’s a pervasive concern of overreach, notably as misinformation spreads sooner and sooner.
Implementing clear governance insurance policies, guardrails, and accountability will make the know-how extra reliable, the report stated.
DOWNLOAD: Easy methods to Preserve AI Reliable from TechRepublic Premium
Methodology
The research’s findings are based mostly on a survey of greater than 48,000 folks throughout 47 international locations between November 2024 and January 2025, together with 1,019 folks within the US, KPMG stated. This text is targeted on US-based responses.
The analysis was carried out by KPMG Worldwide along with the College of Melbourne analysis staff.
