According to Vinchol et al. (2023), the invention of more advanced Artificial Intelligence (AI) tools like ChatGPT and Copilot has led to a massive increase in public engagement and acceptance of AI, making it more accessible, popular, and inclusive not just to workers but to everyone. Although this transformation in the application of AI has brought about a positive increase in work output, it has given rise to anxiety amongst workers who are of the view that there is a possibility of human roles being replaced by AI in the near future. The narrative surrounding the application of AI has also shifted from automation (replacement) to augmentation. For a better comprehension of the paradigm shifts, it is essential to understand the meaning of Automation and Augmentation.
Automation can be understood as a process where AI is allowed to perform a task independently without any form of interference from humans, while Augmentation refers to the collaboration between humans and AI. In essence, augmentation allows AI to assist humans in the workplace. This ideology enhances a shift from the notion of AI replacing humans to a more inclusive perspective that is centered on creating systems where AI acts as a collaborative assistant.
This paradigm shift has not only influenced the narrative in organizations concerning the use of AI but also fosters the idea of ‘Collaborative Intelligence’ proposed by Wilson and Daugherty (2018), where AI-humans collaborate to achieve outcomes that cannot be accomplished by either of them independently.
The Nexus Between Human-AI Collaboration
AI and human collaboration are structured on a form of symbiotic association, being that they both have different capabilities. An instance is that while AI seems to be a powerful cognitive assistant in terms of processing speed and accurate data handling, humans contribute to creative reasoning, emotional intelligence, and critical judgment. This synergy is better interpreted through the five models proposed by Wilson & Daugherty (2018). These models not only outline the comparative strengths but also explain them: (1) Amplification: One of the roles of AI is to identify the patterns that humans may omit through enhanced reasoning. Humans provide accurate interpretations of this insight through their experience. An example is when a radiologist applies an AI tool to detect some errors. (2) Interaction: both human and AI teach each other and learn from each other’s feedback. An instance is when AI learns from the corrections made by a developer during code training. (3) Embodiment: AI modifies humans' physical capacity, allowing people to have a more productive workflow, with precision outputs that they are unable to accomplish alone. (4) Extension: AI expands human roles into new dimensions, such as discovering minor errors in equipment used at the workplace. This allows humans to be more focused and function at an elevated cognitive level. (5) Virtualization: AI assists humans in creating expertise with minimal risk by producing an environment for testing and experimenting. This multidimensional interaction creates a workplace where AI + human intelligence is effectively utilized.
Benefits of Human-AI Collaboration in Work Flows
The economic principle of comparative advantage is when humans and AI focus on what each is best at, so as to have a productive outcome in the workspace. This principle is ideal in workflow when talking about the correlation between AI + humans. The merging of AI + Human skills also reshapes how organizations structure and distribute tasks or roles. For instance, humans focus more on creativity and emotional intelligence while AI prioritizes repetitive processing. From this perspective, it can be deduced that both AI and human perform complementary roles to each other: AI has a superhuman ability which enables it to process large datasets at a faster rate without fatigue and analyse information with consistent recognition of pattern, which are almost impossible for human to do, while humans posses an increased level of cognition and emotional intelligence which enables them to generate nuanced frameworks to process ambiguous information, show empathy and contextual understanding, unlike AI. Through formulating models that combine the strengths of AI and humans, an organization can harness more productivity and innovation in the work environment.
The Importance of AI in the Workplace
In various organizations, research shows that there is a huge output in productivity when AI is merged with human effort. This is because, aside from human input, the use of AI tools has proven to be very effective in enhancing workflow. For instance, users of GitHub Copilot finish given tasks at about 55% faster scale, while 88% of developers report that the use of AI makes them feel more productive due to its affordances (Ziegler et al., 2022). These instances do not displace the role of humans in workflow but rather show that the coexistence of human + AI leads to exceptional outputs in organizations. In the healthcare sector, radiologists have also observed the relevance of human + AI collaboration, as they are gradually moving from a solitary image detection to other roles such as consultants during emergencies, and participating in interdisciplinary research. This has reportedly led to about a 30% reduction in errors encountered by medical practitioners during diagnosis. Research by Mickinney et al. (2020) States that there is a positive increase of about 94% in the accurate detection of breast cancer. In the field of software development, there is a transformation in the role of developers. Being that AI has taken over the tedious role of writing boilerplate code, developers now have more time to concentrate on high-value roles like designing system architecture and solving emergency technical issues. Also, a study by Ziegler et al. (2022) shows that by applying AI tools like GitHub Copilot, developers are able to work 55% faster, with an 88% increase in productivity. Additionally, this transformation is significant in Customer service as the use of AI leads to a qualitative adjustment. Human agents do not need to depend solely on scripts but can now have empathy-driven conversations, show emotional intelligence during conflicts, and build good customer relationships through the support of AI. Research by IBM (2022) has shown that when human customer service agents are assisted by AI, they are able to solve problems at a 14% faster rate because AI is capable of handling about 73% of the challenges faced by customer service immediately. Therefore, the collaboration with AI enables human agents to focus more on the roles that only humans can perform, like showing empathy.
Common Misconceptions and Challenges in AI-Human Collaboration
The growing tension amongst people that AI will take over human roles in the near future is nothing but a Myth. This is because, although almost 15% of jobs currently may be influenced by automation, a recent research by the World Economic Forum (2020) shows that AI will aid in the creation of about 26% of new jobs, leading to an increase and not a decrease in job opportunities. This analysis is backed up by history, as Autor (2015) agrees that technology provides more job opportunities than it eliminates. This is true because recently, we are already seeing these emerging shifts with the introduction of roles like human AI interaction designers, data scientists, AI ethics officers, A1 trainers, and operational specialists. However, the utilization of these tools does not eradicate the possibilities of encountering some challenges in workplaces that humans will have to handle. Some of these challenges are: (1) Bias and Fairness: AI models can inadvertently amplify societal biases found in their programming or training data, leading to unfair or controversial outcomes in workforce recruitments. (2) Over-reliance and Trust: This occurs when humans fail to carry out a form of scrutiny or critical evaluation of AI outputs, thereby leading to ‘automation bias’. Trusting AI tools completely without verification can result in errors that could have been avoided or spotted on time. (3) Security and Data Privacy: Exposing crucial workflows to AI tools gives it access to sensitive data, which is at risk of being exposed or accessed without any formal compliance.
Conclusion
In the coming decade, as AI becomes increasingly intertwined with daily workflow, the competitive advantage will be centered on how well organizations can combine AI and human intelligence to produce what Wilson and Daugherty (2018) call “Collaborative Intelligence.” A term that describes the combined effort higher than what either human or AI can achieve while working independently. The ability of an organization to familiarize itself with this collaboration will enhance its productivity, making the organization outperform others that are still relying on traditional workplace methods or only AI. To achieve this collaborative framework, organizations will have to adjust their perspective from replacement to augmentation, invest in reskilling to prepare the workforce for human-AI collaboration, and begin gradually with innovative programs that show value and transformation. The goal of organizations should not just be to create a future where humans are replaced by AI, but one that allows for a collaborative existence between AI and humans. By architecting this partnership where both work collaboratively, organizations will be at an advantage of gaining from both human and AI input, thereby maximizing significant output in the workplace. This will also ensure an increase in innovation and productivity.
References:
- Accenture. (2022). Reinventing the workforce: AI and human Collaboration. Accenture Research.
- Amershi, S., et. al. (2019). Guidelines for human-AI interaction. Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems.
- Autor, D.H. (2015). Why are there still so many jobs? Journal of Economic Perspectives, 29(3), 3-30.
- Huang, M.H., & Rust, R. T. (2018). Artificial Intelligence in Service. Journal of Service Research, 21(2).
- Manyika, J., et al. (2020). Jobs lost, jobs gained. Workforce transitions in a time of automation. McKinsey Global Institute.
- McKinney, S. M., et al. (2020). International evaluation of an AI system for breast cancer screening. Nature, 577(7788), 89-94.
- Prentice. C., et al. (2020). The influence of identity-driven customer engagement on purchase intention. Journal of Retailing and Customer Services, 47, 339-347.
- Norvig, P. and Russell, S. (2020). Artificial Intelligence: A modern approach (4th ed.). Pearson. Artificial Intelligence, Global Edition | Pearson eLibrary
- Topol, E.J. High-performance medicine: the convergence of human and artificial intelligence. Nat Med 25, 44–56 (2019). https://doi.org/10.1038/s41591-018-0300-7
- Ziegler, A., et Al. (2022). Productivity assessment of neutral code completion. arXiv preprint.
- Vinchon, F., Gironnay, V., Lubart, T., Barlotta, S., Glaveanu, V., Botella, M., Bourgeois-Bourgogne, S., Burkhardt, J.-M., Bonnardel, N., Corazza, G. E., Hanson, M. H., Ivcevic, Z., Karwowski, M., Kaufman, J. C., Okada, T., Reuter-Palmon, Karwowski, M., R., & Gaggioli, A. (2023). “Artificial Intelligence and Creativity: A manifesto for collaboration.” The Journal of Creative Behavior, 57(4), 472-484.
- Wilson, H. J., & Daugherty, F. R. (2018). “Collaborative intelligence: Humans and AI are joining forces.” Harvard Business Review, 96(4).
- World Economic Forum. (2020). The future of jobs depart 2020. WEF