The burgeoning trend of integrating Artificial Intelligence (AI) into organizational structures, by treating AI as virtual employees, is paradoxically leading to a decline in the performance and accountability of human workers. While companies are increasingly adopting AI agents and even assigning them formal roles within their organizational charts, new research indicates this practice, far from boosting efficiency, is fostering a culture of blame and reducing the critical faculties of human staff. This phenomenon, termed "AI anthropomorphization," is prompting concerns among experts about its long-term impact on productivity and workplace dynamics.
The initial spark for this discussion emerged in the summer of 2024 when the software company Lattice, a tech unicorn founded by Sam Altman’s brother, announced its intention to onboard, train, and manage AI as if they were human employees. This initiative, however, faced significant backlash, particularly following the company’s layoff of 15% of its human workforce. In response to public and employee concerns, Lattice reportedly scaled back some of the "rights" extended to its digital employees. Despite this specific instance, the broader trend of incorporating AI agents into organizational frameworks has not abated; instead, it appears to be gaining momentum.
A comprehensive study conducted by the Boston Consulting Group (BCG) sheds critical light on this growing practice. The research revealed that nearly one-third of managers across the United States, Canada, and the European Union now conceptualize AI as a colleague or employee. More alarmingly, over 20% of these managers have formally listed AI agents on their company’s organizational charts, effectively blurring the lines between human and artificial contributions. This widespread integration, however, comes with a significant caveat: the study warns of the inherent dangers associated with personifying AI tools and treating them with the same considerations as human personnel.
The BCG study, led by Matthew Kropp, a managing director and senior partner at BCG, surveyed over 1,200 human resources and finance professionals to understand their current AI integration strategies. Following the survey, participants were presented with a workplace document containing multiple errors. The document was distributed to three distinct groups: one where the errors were attributed to a human employee, another where they were attributed to a generic AI tool, and a third where the errors were linked to a specifically named AI "employee." The findings were stark. The group that received the document attributed to a named AI employee demonstrated a significantly lower capacity to identify errors. Furthermore, this group reported a diminished sense of personal accountability, tending to blame the AI agent for the mistakes rather than taking ownership. This tendency also led to an increased likelihood of asking a human colleague to review the AI’s work, thereby offloading tasks and increasing the workload for other employees.
Kropp elaborated on these findings in an interview with Fortune, stating, "If you translate that into an organizational context, that means people sort of passing the buck. That’s creating more work for somebody else in the organization, and therefore you’re creating churn and excess overhead in the organization." This "passing the buck" phenomenon directly contradicts the expected gains in efficiency and productivity often touted as primary benefits of AI adoption. Instead, it introduces a new layer of operational friction and potential inefficiency.
The broader context of AI adoption in the workplace is also marked by a degree of skepticism and unfulfilled promises. Despite significant investments, estimated to reach $2.5 trillion, AI has yet to demonstrate consistent, consensus-driven productivity gains across the board. Many organizations are still grappling with how to fundamentally transform their business operations using AI, leading to a growing questioning of the technology’s promised return on investment.
This study by BCG is not an isolated finding; it builds upon previous research that has highlighted the unintended consequences of deploying AI without adequate expectations and robust guardrails. For instance, a study published the previous year indicated that experienced software engineers who relied heavily on AI tools actually took longer to complete tasks than their counterparts who did not. This slowdown was attributed to the additional time spent debugging AI-generated code, a task that requires human oversight and correction. Furthermore, BCG’s earlier research on "AI brain fry" found that when software engineers utilized too many AI tools simultaneously, they experienced a sense of overwhelm and cognitive fog, leading to increased mistakes and a subsequent drain on productivity. These findings collectively suggest that the uncritical integration of AI can, in fact, be counterproductive.
The Perils of AI Anthropomorphism
The core of the problem, according to Kropp, lies in the human tendency to anthropomorphize AI, attributing human-like qualities and intentions to what are essentially sophisticated software programs. This attribution leads employees to feel a diminished sense of responsibility. "AI doesn’t have responsibility," Kropp emphasized. "It isn’t a person. It can’t be hired or fired. There’s no performance reviews for AI. It’s just a piece of software. So there can’t actually be accountability for an AI. Some human person has to be responsible for that process, and so we can’t let people sort of transfer the responsibility mentally." This psychological transfer of accountability is a critical flaw in the current approach to AI integration.
Beyond simply becoming an office scapegoat, these "AI employees" are proving counterproductive to companies’ broader goals of increasing AI adoption. Managers, in an attempt to foster greater acceptance of these new technologies, have resorted to assigning human traits to AI agents or integrating them into organizational charts. Kropp, however, notes that this strategy has backfired. Instead of increasing employees’ willingness to adopt AI, the study found that participants who interacted with a named AI employee reported a 7% higher concern that AI would replace their roles and a 10% lower trust level in how AI would be deployed within the workplace. This suggests that personification, rather than fostering trust, breeds anxiety and suspicion.
Kropp clarified that his research is not an indictment of AI in the workplace itself, but rather a call for a more nuanced and strategic approach to its integration. The goal is to leverage AI’s capabilities without sacrificing productivity and professionalism. He drew parallels to managing "AI brain fry," suggesting that solutions involve not eliminating AI, but encouraging employees to compartmentalize AI-assisted tasks within their workday and to take intentional breaks to maintain cognitive clarity.
Similarly, Kropp argues that AI agents can and should be leveraged effectively, but their roles must be clearly defined. This clarity is essential to ensure that human employees do not feel threatened by AI but are instead empowered to maintain accountability for their own work, including any errors that arise from their use of AI tools. "We’re not saying you should not put AI on your org charter, you should not name your AI systems," he stated. "What we’re saying is, if you do that, you should be aware of the behavior change that creates for the rest of the organization, and you should be actively managed for that."
The implications of this research are far-reaching for businesses navigating the AI revolution. As organizations continue to experiment with integrating AI into their workflows, a conscious effort must be made to avoid the pitfalls of anthropomorphism. This requires a deliberate strategy that clearly delineates the roles and responsibilities of both human and AI agents. Companies must actively manage the behavioral shifts that occur when AI is personified, ensuring that accountability remains firmly with human oversight. Failure to do so risks not only diminishing individual performance but also undermining the very goals of AI adoption, potentially leading to increased operational costs and reduced overall organizational effectiveness. The path forward lies in a balanced approach, one that harnesses the power of AI while safeguarding the critical human elements of accountability, critical thinking, and professional responsibility. The future of work will likely involve a symbiotic relationship between humans and AI, but it is imperative that this relationship is built on a foundation of clear understanding and strategic management, rather than on the illusion of artificial personhood.
