Artificial IntelligenceData ScienceGenAIHigher Education

TRAINING THE TRANSLATORS: WHY BUSINESS SCHOOLS NEED TO UP THEIR AI GAME


By Steven Keith Platt, Director of Analytics and Lecturer of Applied AI and Statistics, Quinlan School of Business, Loyola University Chicago

The outlook for the AI industry is outstanding and organizations continue to ramp up their investments. Yet many encounter issues in reaching their AI aspirations. This can be attributable to factors such as a failure to adequately address AI risks, lagging digitization efforts, and HR concerns. However, the most challenging are those related to organizational issues. To address this, business schools need to focus on educating their students to become skilled business translators who can help navigate the adoption and integration of AI technologies. As discussed here, this need affords business schools a unique opportunity to align with industry by teaching students to fill the role of business translator through the creation of an AI concentration.

Broadly defined, the role of a business translator is to work closely with both the business and data science teams. They are unique individuals who possess both knowledge of the business requirements and a fundamental understanding of AI. On the business side, these professionals work with different functional groups to ensure a well-defined scope of work, interface with stakeholders, oversee project management, and address organizational adoption, deployment, and oversight. The business translator should also know core data science skills and best practices, including an understanding of the technology and dev—process, as well as putting models into production and oversight.

Due to the growth of AI in the workplace, organizations are increasingly seeking employees who possess both a strong understanding of operations and AI principles and techniques.

Market Opportunity and AI Adoption

Estimates of the size and growth prospects for the AI industry vary greatly. Markets and Markets, for example, in its June 2023 AI Market report,[i] projected that the global AI market will reach $1.35 billion by 2030 (36.8% CAGR). In contrast, Bloomberg Intelligence, also in June 2023, predicted that the GenAI market alone will reach $1.3 trillion by 2032 (42% CAGR; see Chart 1).[ii] MarketDigits has advanced that the GenAI market will reach $56.5 billion by 2028.[iii]

Chart 1. GenAI Revenue Forecast
Source: Bloomberg Intelligence

On the AI implementation side, estimates of adoption rates also vary. One survey found one-third of respondents are using AI regularly in at least one business function.[iv] Yet another study found that as recently as November 2023, only 3.9% of businesses in the U.S. used AI. This reaches 9.1% in the professional, technical, and scientific services sectors and 13.8% in the information sector.[v]

While the exact size of the AI market and adoption rates are difficult to predict, we do believe that AI has the potential to be a major disruptive technology that affords companies unprecedented opportunity and is growing rapidly. In support of this, we note that nationally, 6.5% of businesses plan to use AI in the next six months (15.2 % in the professional, technical, and scientific services sectors and 21.8% in the information sector).[vi] And interestingly, during the second quarter of 2023, earnings calls made by S&P 500 firms, the number of mentions of AI reached an “all-time high of 7,358, an increase of 366% compared to the first quarter of 2023.”[vii]

Barriers to Successful AI Implementation

There are reasons why firms may not reach their AI potential. These include:

  1. A failure to adequately address AI risks, including privacy concerns, fairness, cybersecurity, inaccurate outputs, regulatory compliance, and model interpretability, among others.
  2. Lagging digitization efforts, ranging from a lack of requisite data to disparate data silos and/or, in general, an ineffectual data architecture.
  3. HR concerns, including a lack of technical skills, resistance to organizational change, and the fear of job displacement.

However, the most pressing of issues is a general failure to address functional organizational integration. This is where business schools can help train the next generation of business translators, as it has been noted that “many companies struggle less with the solution and more with organizational issues.”[viii]

The Role of Business Schools

Due to the growth of AI in the workplace, organizations are increasingly seeking employees who possess both a strong understanding of operations and AI principles and techniques. These professionals will be essential in managing environments where AI plays a role. Therefore, business schools must contribute to equipping students with the skills and knowledge necessary to successfully manage in this environment.

Yet surprisingly, most business schools have failed to recognize this. For example, in its 2023-member survey, the Graduate Business Curriculum Roundtable found few business schools “offering courses, concentrations or programs dedicated to Generative AI,” and that it is incredibly important for business schools to integrate “generative AI as subject matters and as an area of faculty research.”[ix]

To address this shortcoming, we suggest that business schools offer their students a three-course AI business concentration. Courses should include:

  1. Introduction to Applied AI– an introduction to AI with a focus on applied machine learning and GenAI. Key concepts and practical business applications should be covered to ensure that students have an in-depth understanding of the technology. Mathematical models, algorithms and statistical tools needed to perform core tasks in machine learning should be presented, as well as the standard and most popular machine learning algorithms. The goal is for students to develop an understanding of the principles of AI, machine learning, and generative AI, as well as how to derive practical solutions using predictive analytics.
  2. Core Engineering for AI– a course that exposes students to the fundamental aspects of data engineering, software engineering, and the operation skills needed to build, deploy, and maintain AI solutions. Coverage to include data aggregation, management, and platforms; programming including SQL and NoSQL, as well as comprehensive coverage of Python and related libraries; and best practices, i.e., version control, code documentation, testing frameworks, etc. A working understanding of large language model training, fine-tuning, and related architectures, such as transformers and retrieval augmented generation, should also be explored.
  3. AI Capstone– an immersive, hands-on course focused on solving real-world business problems by working with large data sets and applying AI solutions. Students work in consultative teams to develop key performance indicators to understand the business’s challenges and develop solutions. This includes the analysis of proprietary data, as well as syndicated research and other data. The course teaches students how to analyze and frame business issues, manage and fuse large data sets, prepare them for exploratory analyses, build machine learning and AI models and implementations, and document executable AI models. Finally, students should submit a detailed project description and undertake business presentations of their work.

Conclusion

The impact of AI technology on business operations will be profound. For companies to successfully implement these technologies, they must complement the work of their data scientists with skilled business translators who can manage and execute these strategies. Training the business translators represents a major opportunity for business schools.