The popularity of generative AI technology has skyrocketed in 2023, and that trend is likely to continue. Fortune Business Insights forecasts global market growth of generative AI technology to increase from USD 43.87 billion in 2023 to USD 667.96 billion by 2030. To gain insights into user experiences with generative AI services, my organization, Applause, surveyed more than 3,000 digital quality testing professionals across the globe. Here's what our survey uncovered.
Generative AI Use Is Growing
The majority of digital quality testing professionals are using generative AI; specifically:
■ 59% said their workplace allows for generative AI use.
■ 79% said they are actively using a generative AI service.
Reasons cited for using generative AI include writing software, doing research, and for more creative pursuits like writing song lyrics and creating electronic music.
Concerns About Bias Are Growing, Too
Despite the increased usage of generative AI, concerns around the technology are growing. When we asked testers about their concerns about bias, these worries became evident:
■ 90% of respondents expressed concern over bias affecting the accuracy, relevance, or tone of AI-generated content, a 24% increase from Applause's AI and Voice Applications Survey from March of this year.
■ 47% shared they've experienced responses or content they considered to be biased.
■ 18% said they had received offensive responses.
Concerns About AI Hallucinations
Hallucinations in AI is another area of concern, with our survey finding:
■ 37% of respondents have seen examples of hallucinations in AI responses.
■ 88% of respondents expressed concern about the technology for software development because of the potential for hallucinations.
Data Quality Concerns
The survey queried digital quality testing professionals about the quality of data being used to train AI algorithms:
■ 91% expressed some level of concern with data quality.
■ 22% expressed extreme concern.
Concerns Around Data Privacy and Copyright Infringement
We also asked if data privacy and copyright infringement were worries related to generative AI. Our survey found:
■ 98% of respondents said data privacy needs to be considered when developing new technologies.
■ 67% said they feel that most generative AI services infringe on data privacy.
When asked about the level of concern that content produced using generative AI that may be in breach of copyright or intellectual property protections, respondents shared:
■ 21% are extremely concerned
■ 41% are concerned
■ 29% are slightly concerned
■ 9% are not concerned
Testing With Real People
As the excitement, adoption, and use cases around leveraging generative AI continue to advance, there is no denying the potential and possibilities around this technology. However, there are still some concerns that need to be addressed and overcome, including bias, hallucinations and harmful content. Utilizing teams of real people to test the technology and gaining their insights into the quality of the content produced — in terms of correctness, subtleties of bias, and relevance — will help improve AI performance over time. Meanwhile, we must also ensure the data these algorithms are being trained on is high quality and does not violate privacy or copyright laws.