4 factors to consider before adopting AI in your business

Consideration n°1: AI will not replace certain basic human skills for decision making

Will algorithmic decision-making makes us less autonomous? While AI presents promising opportunities, it is to be assessed whether overreliance on AI will lead to a decline in human skills, including critical thinking and analysis. In 2007, German neuroscientist Manfred Spitzer coined the term ‘digital amnesia’ to indicate how the excessive dependence on digital technology may lead to forgetting information that is stored in a digital device and can be easily accessed. Striking a balance is crucial, where AI complements human abilities rather than replacing them entirely.

At the same time, the AI revolution may be the occasion to focus even more on leveraging human characteristics that can be differentiating factors for companies, such as truly creative tasks, critical thinking, interpretation skills and complex decision making. This is particularly true when AI models fail. An example of this was the famous case of an AI model trained to detect skin cancers in pictures, that actually learned to detect the rulers in the medical image instead of the cancer growth itself. In healthcare, the ability to evaluate and choose the best AI technology for a specific use, interpret and deliver the AI-created outcome are skills on their own. Upskilling programs, continuous learning initiatives and critical thinking will be vital in empowering individuals to work effectively in conjunction with AI.

Learn more about how our team can support you in your projects related to AI in healthcare >

Consideration n°2: AI alone cannot ensure ethical use of intellectual property

As AI technologies become more sophisticated, issues related to copyright and knowledge property arise. Generative AI models have the ability to replicate existing works of creatives and artists, posing a threat to their source of income. It is imperative to establish proper mechanisms to protect intellectual property rights and acknowledge the value of datasets on which AI models are built. This includes ensuring that AI technologies do not infringe upon copyrights or plagiarize existing works. For the moment, legislations worldwide are not ready to include this: in June 2023, the Japanese ministry of education declared that no copyright will be enforced onto training generative artificial intelligence programs, hence creating a precedent for AI copyright.

For the moment, legal frameworks establishing AI technologies’ right to use online published information are often blurred. However this could change as soon as these platforms/sources start restricting access to their content, hence preventing the information feeding model. In April 2023, Reddit announced its intention to initiate a fee structure for granting companies access to its application programming interface, the mean allowing external entities to retrieve extensive data. Finding the good balance between innovation and protecting intellectual rights will be crucial to ensure ethical use of intellectual property by AI technologies.

Consideration n°3: in some sectors, maintaining a human touch will be critical

There are sectors such as healthcare, education, and social services, where the human connection has a particularly important role in the quality of the experience. While AI can assist in diagnosis and decision-making processes, it may not be able to provide the same level of care, empathy, and emotional intelligence as humans. In May 2023, due to an overwhelming increase of calls, the National Eating Disorders Association adopted chatbot to answer to people calling the helpline. The chatbot was conceived as a tool for advice rather than for a space for listening, raising many concerns on the chatbot’s capacity to support callers.  In this context, empathy is a crucial characteristic because it is intrinsically linked to the perception of trust. AI’s incapacity to adopt our subjective perspective – empathy deficit – casts doubt on AI’s assessments of humans actions, and more importantly, reasons. Maintaining the human connection in the aforementioned sectors is essential to ensure the delivery of personalized and compassionate care.

An additional issue arises with the increasing integration of these technologies into healthcare environments, leading to a growing prevalence of AI paternalism. Paternalism in the medical field has long been a concern, but now there is a risk that doctors may rely on AI to the detriment of a patient’s personal experience and their own professional judgment. Human professionals should retain responsibility for communicating diagnostic results and addressing patients’ concerns. By utilizing AI as a supportive tool rather than a replacement, these critical sectors can leverage its benefits while preserving the unique qualities that only humans can provide, for instance communication of uncertainty.

Consideration n°4: guarantee AI’s sustainability and global accessibility

In the quest for improved performance, AI’s technologies have increasingly relied on larger and more complex models. However, this approach poses a significant challenge from a sustainability perspective. In fact, these models require a significant computational power: it has been estimated that models like ChatGPT could consume about 1 gigawatt-hour per day, the equivalent of the daily electricity consumption of more than 33’000 households in the USA¹https://www.washington.edu/news/2023/07/27/how-much-energy-does-chatgpt-use/. If used extensively on a global scale, the environmental impact could be unsustainable, unless the tool is optimized to reduce energy consumption.

Moreover, as conveyed by the World Economic Forum, access to AI technologies is not equitable worldwide. Access to these tools is determined by several factors, among which internet connectivity and IT infrastructure. The more we depend on AI technologies without addressing these disparities, the greater the risk of deepening inequalities between the richest and poorest parts of the world. A fair and sustainable approach to AI will need to consider energy efficiency, optimization and efforts to bridge the digital divide. By doing so, we can ensure that the benefits of AI are available to everyone, regardless of socio-economic status, and minimize the risk of exacerbating global disparities.

The integration of AI-powered technologies into our work is both exciting and challenging. As we navigate this transformative era, it is imperative to address key considerations.

By fostering human skills, promoting ethical use of intellectual property, maintaining a human touch in critical sectors, and ensuring AI’s sustainability and global accessibility, we can achieve a smooth integration of AI-powered technologies into our professional realms, harnessing their full potential while safeguarding our values and societal well-being. Alcimed can support to capitalise on integrating AI-powered tools into their activities. Don’t hesitate to contact our team!

About the authors,

Clarisse, Consultant in the Alcimed’s Data team in France

Lucas, Consultant in the Alcimed’s Data team in Switzerland

Lucia, Consultant in the Alcimed’s Healthcare team in Switzerland

Matthieu, Consultant in the Alcimed’s Data team in USA