- Only 2% of workers say AI-generated content does not need review
- Workers spend hours each week cleaning up ‘AI decay’
- AI training should be mandatory and processes should be standardized
Despite perceived productivity benefits, many companies are spending time and money cleaning up the “AI workplace clutter,” suggesting the technology generates a lot of unnecessary noise, new research claims.
Data analytics and visualizations (55%), research and research (52%), extensive reporting (52%), and marketing writing and content (44-46%) are some of the most common areas where AI tools might not be as effective as businesses once hoped.
Although 92% agree that AI improves their overall productivity, only 2% say AI results do not need revision.
AI is a noisy path to productivity
Three in five (58%) spend more than three hours a week reviewing results, more than a third (35%) spend more than five hours and 11% spend more than 10 hours a week organizing the generated content.
Zapier’s research adds that AI generally lacks precision, context or usefulness despite appearing polished on the surface.
And it’s not just perception that is down: many have experienced rejected work (28%), security or privacy incidents (27%), customer complaints (25%), and legal or compliance issues (24%).
Zapier data indicates two possible solutions: First, AI models must continue to be improved to improve the quality of responses. But in the meantime, workers should be better trained to handle AI in its current form, not what it should be.
“The companies that achieve the best results are not the ones that avoid AI,” explained Emily Mabie, senior AI automation engineer. “They are the ones who have invested in training, context, and orchestration tools that turn AI from a neglected experiment into a managed process.”
Nearly all (94%) of trained workers say AI increases productivity, but only 69% of unskilled workers agree. As a result, only 1% of skilled workers say their productivity has decreased.
Looking ahead, the report calls for AI training to be mandatory for all workers who handle it, prioritizing high-risk equipment and tasks in the first instance. Companies can also help employees by providing quick templates and formalizing review processes.
“The solution is not fewer tools, but better infrastructure,” Mabie concluded.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds. Be sure to click the Follow button!
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp also.
