LSE - Small Logo
LSE - Small Logo

Louis-David Benyayer

Hao (Howard) Zhong

December 12th, 2023

AI-human collaboration can unlock new sources of competitive advantage

0 comments | 17 shares

Estimated reading time: 5 minutes

Louis-David Benyayer

Hao (Howard) Zhong

December 12th, 2023

AI-human collaboration can unlock new sources of competitive advantage

0 comments | 17 shares

Estimated reading time: 5 minutes

While artificial intelligence erodes human-based advantages in the workplace, it offers only a temporary edge. For sustained competitiveness, combining AI’s computational capabilities with human judgment is key. Louis-David Benyayer and Howard Zhong write that organisations must synchronise human and technical resources, creating roles that blend AI and human skills. That requires investment in technology, talent, and a cultural shift toward collaborative, cross-functional approaches.

Editor’s note: this introduction was written with the help of ChatGPT.


Artificial Intelligence (AI) is rapidly transforming the business landscape, and companies are scrambling to adapt to this new paradigm. AI has the potential to deeply alter the sources of competitive advantage by reducing the importance of some traditional factors and introducing new ones.

For example, Generative AI tools such as Chat-GPT and DALL-E may potentially impact the key success factors of many industries that rely on human talents for some of their processes. We examined how AI affects existing human-based sources of advantage and can be used to create new, sustainable ones.

AI constitutes only a temporary advantage

To be a source of competitive advantage, resources must be valuable and rare. With AI, many tasks previously performed by humans such as setting a price, writing a memo, organising work, etc., are automated. Given their superior computational capabilities, algorithms process more information more rapidly than humans.

Consequently, given the wide scope of activities under threat of being automated, companies that built an advantage out of their human talent and processes see this advantage vanishing. Conversely, companies using algorithms to make decisions see their advantage increasing. Such is the case of credit scoring, risk assessment and insurance premium calculation.

However, for a resource to form the basis of competitive advantage, it must also be difficult to imitate or copy. Since the information to train an algorithm for making a specific decision is widely available, as are the necessary talent, software and infrastructure to create and run it, the uniqueness and distinctiveness of such an algorithm are reduced.

All this means that we may expect automated decision-making to become somewhat of a commodity: more available and easier to access at first, then less distinctive and a weaker source of competitive advantage (unless the company has proprietary access to a very specific and valuable dataset to train the model). Technical capabilities constitute imitable and thus outsourceable non-core resources.

More sustainable advantages

To be useful in building a competitive advantage, resources also need to be embedded in the organisation’s processes. This is why technical AI-related resources and capabilities must be combined with human-based ones.

Firms can perform better than their competitors when they develop capabilities that combine technical and social assets. In particular, the ability to interpret data insights and make decisions on the basis of such insights are core internal capabilities that create value. Personnel and management capabilities cannot be easily imitated and are, therefore, an important source of competitive advantage.

What AI systems struggle to imitate:

  • First, AI systems, while highly advanced, may still fall short in the exercise of good judgment or common sense. Unlike machines, humans are capable of bringing their intuition and judgment to the table when making decisions.
  • Second, in certain contexts, AI algorithms may inadvertently introduce biases based on factors such as race, gender, or socioeconomic status. Human experts play a vital role in ensuring that AI systems are designed and trained in an ethical and unbiased manner, reducing the risk of unintended consequences and negative impacts on society.
  • Third, while AI algorithms have made it easier to analyse data and make predictions based on past patterns, they still have limitations when it comes to adapting to new situations. A human expert with experience in the industry might be able to draw on their own knowledge and experience to make decisions that take into account a wider range of factors.
  • Fourth, while AI systems can perform complex tasks and analyse large volumes of data with ease, human experts are able to bring their creative thinking to the table, helping to come up with new approaches and ideas.

The transformation required

Synchronising human and technical resources, then, is a clear path to achieve a sustainable competitive advantage. However, this combination constitutes a significant challenge.

One thing that organisations may want to consider is developing new job roles that require a combination of artificial intelligence and human skills. In the meantime, with the increasing adoption of human-AI collaboration in the workplace, a redefinition of performance metrics is necessary to account for the changing nature of work in an AI-enabled environment. In order to facilitate better collaboration between human and AI workers, they may also choose to develop training programs that help employees augment their existing skills and make them more effective at their jobs. Lastly, organisations need to recognise the paramount importance of data and shift their focus from being merely data-driven to data-centric.

“Organisations must invest in the necessary resources and capabilities to integrate AI and human expertise”.

In the context of human-AI collaboration, another key question is how to choose AI models wisely. When choosing between in-house and off-the-shelf models, organisations should consider factors such as the complexity of the problem addressed, the availability of data, and the required level of customisation. When it comes to infrastructure, organisations must consider many different factors. These include not only cost, scalability, and security, but also the specific needs of the organisation, the level of technical expertise available, and the potential impact of any outages or downtime.

In short, organisations must be willing to invest in the necessary resources and capabilities to successfully integrate AI and human expertise. This may involve investing in new technologies, hiring new talent, and developing new job roles and training programs. It may also require a cultural shift as organisations move from a traditional, hierarchical model to a more collaborative, cross-functional approach.


 

About the author

Louis-David Benyayer

Louis-David Benyayer is a Permanent Affiliate Professor of entrepreneurship, joint Scientific Director of the MSc (Master of Science) in Big Data and Business Analytics at ESCP Business School.

Hao (Howard) Zhong

Hao (Howard) Zhong is an Assistant Professor of information and operations management, joint Scientific Director of the MSc (Master of Science) in Big Data and Business Analytics at ESCP Business School.

Posted In: Economics and Finance | Technology

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.