Some U. K. According to an AI strategist, companies are having trouble getting their AI projects going because the technology is just not useful.

Data management system Qlik recently discovered that 11 % of Americans are. K. At least 50 AI initiatives are still in the planning stages for corporations. Meanwhile, 20 % have had up to 50 jobs progress to planning or beyond — but then had to delay or even cancel them.

“A I has the potential to impact nearly every industry and department, but it ’s not universally applicable, ” James Fisher, Qlik’s chief strategy officer, told TechRepublic.

AI is just not the right tool for the job in some cases, but some projects fail because of equipment and information issues. Companies must be aware of the issue they are attempting to resolve and use AI where it can most effectively. ”

Notice: How to Lower the Loss Charge of Your Digital Transformation Project

This supports studies from Gartner, which found that at least 30 % of relational AI initiatives may be abandoned by the end of 2025, according to research released in September. This idea is hardly novel, as TechRepublic reported on a similar discovering in 2019.

Data management represents a key issue

The biggest reason for AI project losses from the new Qlik studies, cited by 28 % of the 250 U. K. -based C-suite managers and Iot decision manufacturers surveyed, are the issues around data management.

When there is n’t high-quality, structured data available or the goals are very ambiguous, AI initiatives can fail to deliver. ” Fisher said. For instance, automating customer service interactions without the right data required to support it or perform proper screening.

“Without a good information strategy, AI models will often challenge to deliver meaningful insights. ”

Incorrectly implementing a strategy can be “disastrous, ” Fisher said. For instance, AI-generated password has been known to cause interruptions, and security officials are considering banning the tech ’s use in application development.

Additionally, the Qlik study discovered that 41 % of Americans are male. K. senior managers lack faith in AI, which could be related to various high-profile failures of soon, such as Air Canada’s chatbot giving inappropriate suffer policy details, resulting in legal and financial repercussions. New policy, such as the E. U. AI Act, may just increase the prices of such problems.

Notice: Generative AI: A Cause of ‘Costly Mistakes ’ for Enterprise Tech Clients

But, there are business areas where Fisher has seen AI showing useful, such as supply chain efficiency, fraud detection, and tailored advertising.

“These are employ cases where AI models are fed greater volumes of high-quality data, are aligned to evident business results and may produce sharper, more meaningful insight, ” Fisher noted.

Reduce potential financial losses by seeking out “plug-and-play ” AI solutions, experts say

Gartner estimates that building or fine tuning a custom AI model can cost between$ 5 million and$ 20 million, plus$ 8,000 to$ 21,000 per user per year. GenAI requires a higher tolerance for direct, potential financial investment criteria than immediate return on investment, which many CFOs have found uncomfortably, ” according to experts.

Business leaders should ensure that AI will produce a real profit before investing, according to Fisher, who suggests trying to find a “plug-and-play ” option first.

He explained: “ In an environment where CIOs are now reconsidering the cost-effectiveness of relational AI options, a focus on smaller, purpose-driven designs and targeted uses may, in the near-term, likely prove to be a more sustainable solution.

By reducing risk and richness while ensuring companies are reaping the benefits that AI can provide, plug-and-play alternatives provide businesses with a base for their Artificial jobs. ”

SEE: Generative AI Projects Risk Failure Without Business Executive Understanding

He also advised beginning with smaller AI projects to demonstrate proof of concept before scaling and to regularly evaluate the ROI.

The first thing to do is to have a solid data foundation and have the appropriate data governance, quality, and accessibility in place, Fisher said. Make sure you have a clear business issue or issue in mind before using AI to measure success. To build trust in the technology, try to encourage knowledge sharing and upskilling across the business.

“Finally, take a gradual approach to AI adoption; before making larger bets, start with a proof of concept to validate your project. ”