Artificial intelligence has promised us productivity gains, and these benefits are certainly on their way, but issues with AI output quality continue to dilute the technology’s benefits. Almost 40% of the productivity offered by artificial intelligence is now being absorbed by rework, with low-quality results having to be redone in order to reach the caliber of output that businesses want to put out. The excitement around this technology has many holding their breath for an explosion in results and business capabilities, and this isn’t an unrealistic expectation long-term.
However, with the technology still hovering in an extremely developmental phase, maintaining high-quality AI output takes active effort. As the gap in workplace AI use continues to shrink, businesses may have to take a step back to assess whether the transformation they’re hoping to see still fits into their plans for the immediate future of their business.
AI-driven productivity gains are redefining operations, but AI output quality issues result in almost 40% of AI time savings being lost to rework. (Image: Freepik)
AI Output Quality Issues Place Additional Burdens on Workers Using the Technology
Workday recently released new research on workplace AI usage in its Beyond Productivity: Measuring the Real Value of AI report, and it helped place some of the current reliance on the technology into perspective. Artificial intelligence is indeed revolutionizing how we operate, transforming roles and responsibilities, and simplifying tasks, all while delivering better results without complications. Unfortunately, AI expectations vs outcomes remain misaligned.
While 85% of employees reportedly save between one and seven hours per week with the help of AI, nearly 40% of the time saved is lost to rework. This includes the task of correcting any errors, rewriting content, and verifying outputs to ensure accuracy. Workday found that only 14% of employees see consistent and clear positive net outcomes from AI.
This tells us that despite the AI workplace adoption trends paying off to a large degree, there is still a need for us to adjust our expectations of just how much can be achieved by relying on this technology.
As AI Reduces Output Quality, Workers Take on Additional Burdens
The data on the AI output quality issues doesn’t stop there. It stands to reason that those who frequently use the technology learn how to weed out the problem areas with greater ease and reduce the frequency of issues faced, but the data suggests otherwise. Employees who use AI daily are extremely optimistic about its capabilities, with 90% certain of its ability to generate success. However, 77% of them have an even bigger burden of reviewing AI-generated results with as much scrutiny as human-produced results, if not more.
Workday’s AI research also showed that younger workers aged between 25 and 34 make up 46% of those dealing with reworking AI results. Before we leap to conclude that they are not adept at using the tech despite being labelled tech-savvy, it’s also safe to assume that they make up the majority of regular AI users. The quality of AI-generated material depends on the capabilities of the user and their prompts to a large degree, but the technology is also an independent entity that cannot always be ordered to furnish perfect results.
The Workplace Is Evolving to Cater to AI, but Not in a Way That Benefits Workers
New reports on increasing AI usage and productivity gains emerge every day, and they appear to be trending towards good news for employers. For employees as well, the technology offers a way to simplify some tasks and expand on others, but the benefits are uneven. Despite the fanfare around AI, many employers seek employees who come with AI qualifications preloaded in their resumes.
About 66% of leaders state that skill training is a priority. But only 37% of employees facing the highest amount of rework responsibilities are beneficiaries of the necessary training. Roles and responsibilities are evolving, but at 89% of organizations, fewer than half of the roles have been updated to reflect AI capabilities.
Pre-AI job structures continue to serve organizations despite their post-AI demands, and this leaves many employees caught up in outdated processes that don’t reflect the time, energy, and resources to confront issues with AI output quality comfortably. Previous research has shown us that while AI is altering tasks and offering productivity support, workers are left with bigger workloads than before, risking burnout with every additional task on their desk.
Companies Are Leaving AI Gains on the Table with Their Existing Strategies
Upward-trending AI workplace adoption trends clearly highlight that businesses are doubling down on squeezing productivity out of these new systems, and this has been beneficial to their operations. At the same time, greater attention needs to be paid to updating these systems in order to fully embrace them. Maintaining high-quality AI outputs is an essential consideration for businesses that want to continue using the technology and maintain their standards of operation, but this can only be achieved when resources are channeled back into its efficient use.
The Workday AI research showed that companies are more likely to direct AI savings into the technology rather than employee development. This may leave employees with heavy workloads, working to master the technology on their own. Doing so can leave hassled workers with more to do in a day than desirable, and could have them allowing the AI output quality issues to show in the final results, compromising accuracy for timelines and other responsibilities.
Just as with any other change to operations, pausing to reflect on current performance and investing in support systems is essential while pursuing AI gains. Learning how to assess AI output, establishing procedures to correct AI flaws, and adjusting timelines and workloads are basic ways for organizations to take change. Ensuring that the investments into the technology result in high-quality, uniform, duplicatable results is the need of the hour.
Have you seen similar quality issues with AI outputs and results? Share your experience with managing them in the comments. Subscribe to The HR Digest for more insights on workplace trends, layoffs, and what to expect with the advent of AI.
