Is AI Creating More 'Workslop' Than Real Productivity

Organizations are calling for increasing AI adoption, yet many are struggling to reap returns on the promised AI productivity gains. Not only are initiatives falling flat, but what has become increasingly prevalent is AI-generated “workslop.” 

The term “workslop” refers to low-effort, AI-driven work that, on the surface, appears substantive and meaningful but fails to drive productive outcomes. 

This raises concerns about the increasing adoption of AI despite its current limitations, which lead to counterproductive outcomes in the workplace. It calls upon leaders to be more responsible and intentional when introducing AI and automation technologies into their companies. 

This article examines the increasing prevalence of AI-induced “workslop,” concerns around AI productivity, and the approach leaders need to take towards AI adoption.

What is an AI-Driven “Workslop”? 

Companies actively promote the use of AI at work without sufficient oversight or guardrails. Employees independently develop AI use cases and indiscriminately leverage AI in every aspect of work. This leads to AI-generated “workslop”: a deluge of subpar outcomes that appear to be meaningful but are actually devoid of substance. 

“Workslop” is partly a technology problem. Current AI models are fraught with errors. AI pilots note that GenAI models, especially Large Language Models (LLMs), appear to be unconcerned with the “correctness” or real-world accuracy of their outputs. They are known to generate unfounded, often nonsensical responses and even manufacture information in a phenomenon known as “AI hallucinations.” As a result, AI-driven “workslop” may be increasing. 

Yet, despite the limitations of AI models, the real problem lies in how leaders introduce new technologies in the workplace, i.e., without clearly defining use cases, ensuring adequate ethical safeguards, or providing effective training to users. This lack of intentionality and attention can lead to workplace inefficiencies when utilizing AI and automation technologies that are still in their early stages of development.

Impact of AI-generated “Workslop” on Workplaces

AI has significantly impacted workplace efficiency more than any other technology introduced in recent years. AI undeniably holds promise, but it has flaws, and AI-driven work inefficiencies ("workslop") create challenges:

• “Workslop” creates more work for managers and co-workers. They often spend more time fixing low-quality, AI-generated “workslop” than if a human had completed the job. This costs organizations significantly in lost productivity. Interestingly, the increasing prevalence of “workslop” has given rise to a niche economy, where freelancers are being hired at high rates to elevate low-quality, AI-generated work.

• Frequently and diplomatically handling subpar work can be mentally draining. It leads to feelings of confusion, frustration, and annoyance among workers.

• “Workslop” give rise to interpersonal tensions and distrust in teams. Recipients of AI-generated work may perceive their co-workers as less able, creative, and reliable. 

• If AI-generated work is routinely accepted or tolerated, it sets a precedent that lowers quality standards for individuals, teams, and the organization as a whole. It can also inadvertently add more volume to the poor-quality work that workers must manage on a day-to-day basis.

AI's tendency to “hallucinate” leads to workplace inefficiencies rather than AI productivity. This refers to situations where an AI system generates information that appears convincing but is actually false, misleading, or fabricated. Furthermore, if employees inadvertently volunteer sensitive company data to AI tools, it can lead to confidentiality breaches.

What Employers Need to Do

AI-generated “workslop” typically occurs when companies fail to invest in the people using it. Therefore, reducing the instances of low-effort work and mitigating the negative impact associated with it is the responsibility of the leaders who integrate AI into company processes. 

Essential steps employers can take to prevent AI-generated “workslop” include the following:

1. Organizations need to commit to ensuring AI and automation technologies are used responsibly and ethically. Establishing adequate safeguards and emphasizing human oversight are necessary to extract beneficial outcomes.

2. Training programs should include formal courses and conversations to educate and familiarize employees with the strengths and weaknesses of AI. Including real-world examples that, for instance, highlight the hidden costs of indiscriminate AI use can create an impact. 

3. Managers need to invest in training their employees on the effective use of AI, as well as formally communicate who can and cannot use AI. They must be ultimately responsible for the quality of work their teams produce.

4. Many workplaces discourage the use of AI due to valid concerns about confidentiality and productivity. Still, an emerging phenomenon, "shadow AI," suggests that employees leverage AI in their work but conceal it from management, fearing they may be judged or replaced. What can reduce this ambiguity around AI is having open conversations about its use. Teams can share ethical and productive applications of AI and be forthcoming with coworkers when they do use AI. Leaders should encourage a “pilot mindset,” which emphasizes using AI purposefully and with agency.

5. A comprehensive AI policy needs to be established that formalizes the areas where AI can be leveraged and those that remain off-limits. This focus on communicating the clear use cases of AI is necessary to prevent “workslop.”

6. Organizations must establish concrete metrics to evaluate AI’s impact on productivity, as relying on vague assumptions of effectiveness undermines progress and leads to counterproductive outcomes

Conclusion

The launch of generative AI set off a tidal wave of changes in workplaces. GenAI has come to impact nearly every aspect of both employees’ and employers’ work. Yet, one of the most ubiquitous concerns among employers today is the lack of returns on the promised AI productivity gains.

It should be noted, though, that at its current stage, AI remains largely experimental. This isn't inherently a drawback, as it reflects the exploratory nature of evolution. However, from a practical perspective, this experimental stage makes the real-world value of AI uncertain, despite the technology showing remarkable potential. Practical and effective use of AI still depends on the degree of human intervention involved. 

More intentional efforts are needed to reap the benefits of AI productivity and facilitate broader adoption of AI. Clear guidelines around when and how to use AI are necessary to ensure its applications remain purposeful, ethical, and productive.