Expert Perspectives on the Application and Future of Data Science

Expert Perspectives on the Application and Future of Data Science

When it comes to considering the role of big data in the world, it’s hard to ignore national efforts to be the top innovators in this field. All of the experts interviewed agreed that the US is currently at the forefront of big data innovation; however, according to Schoeder, the US’s “role is slipping.  This is true for many reasons, but perhaps none so important as the steady loss of a hopeful vision of the future, broadly supported by people throughout the world.  In place of a clear vision of democratic, accountable and inclusive data and technology systems, especially for the future of AI, we see the rise of a counter vision of digital authoritarianism, championed particularly at the national level by China. This vision was to some degree implicit already in the emergence of ‘Surveillance Capitalism’ among the large US tech firms but has been pushed to new extremes and implemented in new ways by Chinese tech firms and government. Unless a new type of hopeful vision can be recovered, either in the US or some combination of other countries, then it is highly likely that digital authoritarianism, especially in its Chinese incarnation, will become the wave of the future for so-called ‘big data.’” Subsequently, dominance in the fields of data science and AI could also be broken down via different countries whereby Yu notes that “the US will continue to lead in theoretical research, China will lead in facial recognition and expert systems, and Japan will lead in AI robotics. 

Although different countries may emerge to lead in specific areas of data science, all of them will have to deal with the ethical ramifications of the application of data science. Among the leading concern is eliminating bias, and according to Ranganathan, “figuring out how to collect and use data in a privacy-preserving manner.” However, it is not just bias that needs to be addressed. In regard to ethical challenges, Hickok also notes that “there is a significant lack of ethical training in college that provides historical and social background to data scientists, engineers etc. Data scientists are coming into a field without properly framed expectations, role responsibilities, ethical guidelines or training to help them ask the right questions.”  

A lack of training on ethics in AI and data science can not only affect those working in those fields, but also the general public. Among the greatest threats to AI and data science, according to Hickok, Ranganathan, and Yu, are a level of trust that can lead to an overestimation of the efficacy of automation and the belief that data and AI can help to solve our prevailing problems. Part of the reasons for these beliefs, according to Schoeder, can be related back to “a version of AI which is not designed with values of democracy, participation, transparency, and inclusion from the ground up. In that case, we risk creating forms of digital governance which enforce social divisions and repression rather than unlocking the creativity of a post-scarcity economy and the remarkable social potential of ubiquitous digital intelligence.” AsDonalek points out, this can also be attributed to a lack of comprehensive education in the field of data science as there is an “influx of people without a proper data science background working on ML models…Companies need to understand that an expert is always needed and data science is not something that can be mastered taking an online class.”   

Knowledge on the part of the general public and experts in the field will be a part of increasing importance as data science becomes more a part of everyday life. Looking at the future of data science, Ranganathan notes that “it’ll be more commoditized, automated, and accessible to all kinds of users. Building on this concept, Schoeder also posits that data science will ultimately be implemented in different fields, as within a “decade or less from now there will no longer be a discipline called ‘data science. Every field will, to one degree or another, absorb the key methods and practices which we now distinguish as ‘data science’ – which will at once make it disappear and become entirely ubiquitous. With its growing applications to various fields across the world, the need for an unbiased approach to data collection in the hands of both a well-educated public and practitioners is of increasing importance.

To learn more about our experts, visit them on their websites:  

Merve Hickok: www.linkedin.com/in/mervehickok/ 

Anand Ranganathan:www.linkedin.com/in/anand-ranganathan/ 

Andrew Schroeder:www.linkedin.com/in/aischroeder/ 

Chong Ho Yu:www.apu.edu/bas/faculty/cyu/ 

Ciro Donalek:www.linkedin.com/in/ciro-donalek-1ba1875/

Leave a Comment

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