Mass Adoption of AI Leading to Knowledge Degradation in Companies

In the modern business world, the mass implementation of generative AI tools is causing serious systemic problems instead of the expected efficiency. A study by Harvard Business Review shows that companies, in an attempt to reduce dependence on human labor, are facing unexpected obstacles such as the decline of knowledge and the slowing of work processes. This is reported by Ixbt.com news reports.

Analyzing this situation, experts are using the newly emerged term "workslop" in the business lexicon. This concept refers to a flow of low-quality, unverified content created by AI that disrupts work processes. As companies apply AI to all levels to avoid falling behind in the technological race, the overall quality of corporate data is declining, and employees are beginning to lose their skills.

Knowledge Degradation and the Crisis of Trust

According to Harvard Business Review, the problem is rooted in a chain reaction: employees use models like ChatGPT to prepare work materials, but this content is often full of errors or "hallucinations." As a result, colleagues are forced to spend additional time verifying those materials. This leads to a decrease in trust in internal data and the devaluation of corporate experience accumulated over years.

Currently, separate staff units are appearing in many organizations dedicated solely to correcting errors made by AI. This negates the economic benefit expected through automation. Employees have begun to doubt not only the technology but the entire workflow through which data passes, which negatively affects the team environment.

The Labor Market and Talent Selection

The impact of AI is not bypassing the labor market either. Communication between candidates and employers is becoming artificial due to automated stages, which distorts the expectations of both parties. Consequently, the process of selecting qualified specialists is becoming more complex, and the probability of errors is increasing.

Researchers suggest the following recommendations to overcome this crisis:

• Apply generative AI not at all stages, but only in specific tasks that add real value;
• Use specialized systems trained on the company's internal data rather than mass and open models;
• Implement a mandatory verification system for every piece of data generated by AI;
• Support traditional methods of knowledge sharing to preserve employees' critical thinking skills.

In conclusion, the initial views on infinitely increasing efficiency through AI have today collided with reality. By implementing AI without control, companies, instead of reducing the workload, find a need for additional manual labor to correct errors. This demands a more cautious and systemic approach to the technology.