New research by Alteryx, Inc., finds that while over three-quarters (79%) of organizations believe that generative AI (genAI) adds business value, various risks are holding deployments back.
The ‘Market Research: Attitudes and Adoption of Generative AI’ report, which surveyed 690 EMEA IT business leaders, highlights trust, ethics, and skills issues as consistent concerns for business leaders.
Trusting the output of genAI is a major concern with 50% of business leaders saying that their organization has already experienced misinformation produced by genAI.
Alongside this, the accuracy of information delivered by genAI was also called into question, with businesses also reporting issues with genAI generating infringements on copyright or intellectual property rights (40%), as well as receiving unexpected or unintended outputs (36%).
The biggest issue impacting trust in genAI from businesses (62%) is ‘AI hallucinations’ – when genAI produces incorrect predictions or nonsensical outputs. For businesses, this is where applying generative AI to the right use cases, with the right technology and guardrails is needed to alleviate concerns.
Alongside concerns about the accuracy and trustworthiness of genAI outputs, business leaders also have strong views on where GenAI should never be used. 41% of business respondents said that genAI should not be used in critical decision-making.
Any AI-driven system is only as good as the data it’s trained on, however, the data highlights worrying shortfalls within organizations. Only 33% of leaders said that their business ensures that data used to train genAI is diverse and not biased. Added to this, only 36% say they have ethical guidelines in place and 52% say they have data privacy and security policies for genAI use.
This lack of focus on data integrity exposes organizations to risks, with 63% of business leaders stating that ethics is their top genAI risk category, followed by data (62%). This shows a need for greater governance to build trust and mitigate risk in how employees use genAI in the workplace.
Business leaders say they’re turning to genAI in the workplace for data analysis (42%), cybersecurity (41%), and customer support (35%) and since the start of 2023, have run an average of three pilot projects. But, while 74% said these pilots have been successful and 53% stated it was easier than expected to get results, 28% found it harder to leverage genAI in the organization than anticipated. Businesses report facing challenges including security concerns (41%), data privacy and security (39%), and quality and reliability of outputs (32%).
So, while genAI is starting to deliver business value, these roadblocks need to be overcome. A contributing factor alongside insufficient governance frameworks, could be a lack of data literacy. A fifth of businesses say they don’t have mandatory genAI training in place (20%) and 28% report that lacking skilled talent is holding them back from scaling genAI across the organization.
