West Yorkshire data science specialist Frame has conducted a comprehensive survey analysing firms’ views around AI.
Among key findings, 40% of businesses cited perceived depersonalisation as one of their biggest apprehensions around the technology. Other issues included unpredictability, a lack of accountability and fears around job displacement.
Liam Fulton, CTO of Frame, said: “We know that AI is here to stay, but the key to its success is users understanding it and educating both themselves and their teams about how to maximise it for commercial benefit. One of the most important things is to set realistic expectations and know a technology’s limitations. There are a few simple steps to follow including providing clear training, guidance and updates. If you do that, then you’ll improve trust and results.”
For the survey, Frame approached 400 firms across the UK operating in the banking, finance, IT, retail, e-commerce, logistics and distribution sectors.
Respondents were also asked about what annoys them most about AI, with misunderstandings coming out top. However, 80% reported that they received the correct answer or result between 50% and 75% of the time and nearly two thirds declared they were confident in its abilities.
When questioned about why they like AI, automation and efficiency came out on top with 56%, followed by inspiration at 22% and problem solving on 11%.
The final piece of data showed that 95% of people like using AI and 100% intend to apply it in some capacity next year. The most common types being used are generative image and text AI, speech recognition and data science.
Fulton added: “In general, AI systems are only as good as the people who use them, so it’s essential to know how and why they make decisions as that will help users understand the reasoning behind the outputs. When they can grasp the decision-making process, they are more likely to trust the system.”
“Finally, never forget about user feedback and continuous improvement. That really matters as involving users in the feedback loop helps identify and rectify misunderstandings, improving a system’s accuracy over time. This iterative process fosters trust by demonstrating a commitment to learning and improvement.”
