When Artificial Intelligence Drives Decision-Making: The Data Scientist in the Age of Generative Models

For a long time, data scientists were viewed as the company’s advanced analysts, tasked with transforming data into actionable insights. Their role involved collecting, cleaning, and modeling datasets to identify correlations, predict behaviors, or optimize processes. In many sectors—including finance, marketing, healthcare, and manufacturing—data science has established itself as a key driver of decision-making.

But this view of the profession has undergone a profound transformation in recent years. The explosion in data volume, the emergence of deep learning architectures, and the spectacular rise of generative models have changed the scale and nature of analytical work. Organizations are no longer content to simply analyze the past; they now seek to anticipate trends, simulate scenarios, and automate certain decisions.

This transformation is part of a broader trend: the global economy is now structured around data. Companies collect information from multiple sources, including digital transactions, IoT sensors, customer interactions, social media, and logistics platforms. According to the International Data Corporation, the global volume of data has exceeded 175 zettabytes and continues to grow exponentially, driven by the widespread adoption of connected devices, digital services, and artificial intelligence systems¹.

In this context, analytical complexity increases significantly. Organizations must process massive volumes of heterogeneous data—both structured and unstructured—often in real time. Traditional statistical models are no longer sufficient to extract value from these streams of information.

The figures illustrate this shift:

• Global investment in artificial intelligence is expected to exceed $500 billion by 2027.
• More than 80% of international companies report using at least one form of advanced analytics or machine learning in their decision-making processes.
• Generative models capable of producing text, code, images, or analytical simulations are opening up new areas of application for data science.

The profession is thus entering a new era. Data scientists are no longer limited to producing predictive models or analytical dashboards. They are becoming central players in digital strategy, capable of designing intelligent systems that directly inform decision-making. In an environment where algorithms can analyze volumes of data inaccessible to humans, the value of the data scientist lies increasingly in their ability to orchestrate models, ensure data quality, and transform algorithmic intelligence into informed decisions.

 How AI Is Transforming Data Science Practices

Artificial intelligence is not only transforming economic sectors; it is also profoundly changing the way data science is practiced. Historically, the work of a data scientist involved a series of lengthy and often manual steps: data exploration, training dataset preparation, model selection, hyperparameter tuning, and production deployment. With the rise of advanced AI platforms, foundational models, and cloud infrastructures, an increasing portion of these tasks is now automated or augmented by intelligent systems. Data scientists are thus operating in an environment where tools can generate code, propose model architectures, and analyze data at scale, transforming the discipline itself.

This trend is evident at several key stages of the model lifecycle.

• Automated machine learning (AutoML): AutoML platforms now automate model selection, hyperparameter tuning, and performance evaluation. Whereas a data scientist previously had to manually test multiple algorithms, these systems explore hundreds of configurations in parallel. According to Gartner, AutoML technologies could automate a significant portion of modeling tasks in the coming years, allowing experts to focus more on interpretation and analytical strategy².
• Foundational Models and Generative AI: The emergence of massive pre-trained models capable of understanding and generating text, code, or images is profoundly transforming the practice of data science. Data scientists can now rely on these models to accelerate prototype development, generate analytical scripts, or quickly explore complex datasets. This evolution reduces the time required to move from an idea to a functional experiment.
• Intelligent data engineering: Data preparation has historically been the most time-consuming part of analytical work. AI systems can now automatically detect anomalies, identify relevant variables, or suggest data transformations. These capabilities significantly reduce the time spent on data cleaning and structuring.
• MLOps and Model Industrialization: Deploying AI models into production is becoming a major challenge for organizations. MLOps pipelines enable the automation of model deployment, monitoring, and updating. Data scientists now work in environments that incorporate practices similar to those in software development, including version control, performance monitoring, and continuous retraining of algorithms.
• Simulation and Augmented Decision Support: Beyond prediction, AI models now enable the simulation of complex scenarios. Companies use these systems to test different business strategies, anticipate customer behavior, or optimize supply chains. Data science is thus becoming a direct driver of strategic decision-making.

These changes are fundamentally transforming the nature of the profession. Data scientists are no longer limited to building statistical models. They must now coordinate a suite of intelligent tools, understand the architecture of advanced models, and ensure that decisions generated by algorithms remain interpretable and relevant within their business context.

A new role for the data scientist

The widespread adoption of artificial intelligence within organizations is not only changing the tools available to data scientists; it is also fundamentally redefining their role within the company. Long viewed as a technical specialist responsible for building statistical models, the data scientist is now becoming a strategic player in digital transformation. Their role is no longer limited to producing analyses, but rather to shaping the decision-making capabilities of organizations.

In an environment where companies have access to ever-increasing volumes of data and models capable of generating complex predictions or content, the value of a data scientist lies increasingly in their ability to make sense of the results produced by algorithms. The challenge is no longer simply to build a high-performing model, but to understand its impact on decision-making, its interpretability, and its reliability in real-world contexts.

This trend has led to several major changes in the industry.

• Architect of data-driven decision-making: Data scientists now play a direct role in the company’s strategic decisions. Their analyses inform decisions related to marketing, risk management, operational optimization, and product innovation. They serve as a bridge between technology and decision-makers, capable of translating complex findings into actionable recommendations.
• Artificial Intelligence Model Orchestrator: With the emergence of generative models and deep learning architectures, data scientists must understand and oversee increasingly complex systems. They no longer simply develop a single model; instead, they design analytical ecosystems that integrate multiple algorithms, data pipelines, and continuous learning systems.
• Guardian of data quality and governance: A model’s performance depends above all on the quality of the data used to train it. Data scientists play a key role in structuring data pipelines, detecting biases, and validating the datasets used by algorithms. This responsibility becomes critical in contexts where automated decisions can have significant economic or societal impacts.
• The interface between business and technology: Data scientists are increasingly working in interdisciplinary teams that bring together software engineers, data engineers, business specialists, and strategic decision-makers. Their ability to understand business challenges and communicate effectively with people from different backgrounds is becoming a key factor for success.
• Algorithmic Ethics Lead: The growing use of artificial intelligence systems raises questions regarding bias, transparency, and accountability in automated decision-making. The data scientist helps implement practices designed to ensure model transparency and the robustness of analyses.

According to the World Economic Forum, jobs related to advanced data analysis and artificial intelligence are among the fastest-growing professions in the global economy by 2030³.

Thus, the data scientist of the future will not simply be a technical specialist. They will become a key player in organizational strategy, capable of transforming massive data streams into actionable insights and informed decisions.

What skills does a data scientist need in the age of generative AI?

 The fundamentals of the data scientist’s profession—proficiency in statistics, programming, predictive modeling, and an understanding of databases—remain the indispensable foundation of the discipline. Methodological rigor, the ability to structure an analytical problem, and proficiency in data exploration techniques remain at the heart of the practice. However, the rise of generative models, the industrialization of artificial intelligence, and the proliferation of automated systems are significantly expanding the scope of expected skills.

Data scientists must no longer simply build high-performing models; they must understand the entire technological ecosystem surrounding them, from data collection to the impact of algorithms on decision-making.

This shift is transforming data science education, professional practices, and culture.

Technical and digital skills

• Understanding Advanced Artificial Intelligence Architectures: Foundational models, deep neural networks, and transformer-based architectures now play a central role in the AI ecosystem. Data scientists must understand how they work, their limitations, and their implications for performance and resource consumption.
• Mastering development environments and data infrastructure: Artificial intelligence projects rely on complex infrastructures that combine cloud computing, data pipelines, and deployment platforms. Data scientists must understand the technical architectures that enable models to be trained and deployed at scale.
• Interpreting and explaining model results: The rise of complex models underscores the importance of explainability. Model interpretation techniques, such as SHAP or LIME, are becoming essential for understanding algorithmic decisions and making them understandable to users.
• Understanding MLOps Pipelines: Deploying models into production requires rigorous version control, monitoring, and retraining. MLOps skills are becoming a cornerstone of modern data science.

According to a McKinsey study, companies that have industrialized their artificial intelligence pipelines can double the speed at which they deploy analytical models⁴.

Analytical and decision-making skills

The decision-making environment enhanced by artificial intelligence is profoundly changing the way analyses are interpreted and used.

• Maintain a critical mindset when evaluating algorithmic recommendations: Predictive models can produce misleading correlations if the training data is biased or incomplete. Data scientists must maintain a constant capacity for critical analysis.
• Interpreting complex and sometimes contradictory signals: Analytical systems can produce differing results depending on the assumptions or datasets used. The data scientist’s role is to reconcile these signals and identify the conclusions that are truly actionable.
• Developing a probabilistic approach to decision-making: Data science introduces a framework based on probabilities and uncertainties. Data scientists must be able to explain these concepts to decision-makers and integrate this perspective into the company’s strategy.

Ethical and Regulatory Competencies

The widespread adoption of artificial intelligence in organizations raises significant issues regarding accountability.

• Identifying and correcting algorithmic biases: A model trained on biased data can reproduce or amplify existing inequalities. Data scientists must analyze datasets and implement corrective measures.
• Understanding AI Regulatory Frameworks: With the emergence of regulations such as the European AI Act, certain applications of artificial intelligence will be classified as high-risk systems. Professionals will need to ensure the transparency and traceability of the models used.
• Ensuring transparency in algorithmic systems: Organizations must be able to explain the decisions made by automated systems. Data scientists help implement mechanisms for auditing and documenting models.

Interdisciplinary and strategic skills

Data science is no longer an isolated discipline.

• Collaborate with software engineers and data engineers: Designing high-performance AI systems relies on multidisciplinary teams that combine technical expertise with domain knowledge.
• Understanding the business implications of the analyses produced: Data scientists must be able to link their models to specific business objectives, such as optimizing sales, reducing risk, improving the customer experience, or driving product innovation.
• Supporting organizational transformation: Fostering a data-driven culture within the company requires an educational effort. Data scientists help train teams and integrate data analysis into decision-making processes.

According to the World Economic Forum, skills related to data science, artificial intelligence, and advanced data analysis are expected to be among the most in-demand skills in the global economy by²⁰³⁰.

Thus, the data scientist of the future will not merely be a technical expert. They will become a key player in organizational strategy, capable of bridging the gap between the power of algorithms and the human understanding of economic, social, and technological phenomena.

Can artificial intelligence improve the quality of decisions?

One of the strongest arguments in favor of artificial intelligence in organizations is its ability to improve the quality and speed of decision-making. By analyzing volumes of data that humans could not process on their own, machine learning models can identify hidden correlations, detect weak signals, and generate more accurate predictions.

In many industries, these capabilities have already transformed the way companies make strategic decisions.

Specific examples:

• Optimizing Marketing Decisions: Predictive models enable companies to anticipate purchasing behavior and personalize customer recommendations. Businesses that use advanced data analytics systems can significantly improve the performance of their marketing campaigns.
• Financial risk management: In banking institutions, algorithms analyze millions of transactions to detect fraudulent activity or assess customer creditworthiness. These systems can identify anomalies in a matter of seconds.
• Demand forecasting and operations optimization: In manufacturing and logistics, predictive models analyze consumption trends, sales history, and economic variables to forecast demand and adjust supply chains.
• Strategic analysis and scenario simulation: Companies are increasingly using artificial intelligence to simulate various economic or operational scenarios. These simulations make it possible to anticipate the impact of strategic decisions before they are implemented.

The results are already evident. According to a McKinsey study, companies that use artificial intelligence in their decision-making processes can significantly improve their productivity and operational performance¹.

However, these advances also bring new challenges.

• The risk of algorithmic bias: A model trained on biased data can produce unfair or erroneous decisions.
• The opacity of complex models: Some deep learning systems operate as “black boxes,” making it difficult to explain the decisions they produce.
• Data dependency: The quality of decisions generated by artificial intelligence depends directly on the quality and representativeness of the data used.

Thus, artificial intelligence can significantly improve decision-making, but it does not replace human judgment. The most successful organizations are those that combine the analytical power of algorithms with the expertise and intuition of human decision-makers.

What will the role of a data scientist look like in the future?

The data scientist of tomorrow will work in an environment where artificial intelligence systems are ubiquitous and deeply integrated into organizations’ digital infrastructures. Analytics platforms will become more automated, generative models more powerful, and data volumes even more massive. In this context, the role of the data scientist will not disappear; rather, it will evolve into a role focused on overseeing, designing, and orchestrating intelligent systems.

Several significant changes are already evident.

• The Rise of Augmented Data Science: Analytics platforms incorporating artificial intelligence are increasingly automating technical tasks such as data exploration, model selection, and hyperparameter optimization. Data scientists will devote more time to problem formulation, interpreting results, and developing analytical strategies.
• The emergence of new hybrid roles: The line between data scientists, machine learning engineers, and AI engineers is becoming increasingly blurred. New roles are emerging, such as AI systems architect, algorithmic ethics specialist, and model governance manager.
• The industrialization of artificial intelligence models: Organizations are seeking to deploy these models at scale. MLOps pipelines, cloud infrastructure, and distributed architectures are becoming essential components of the analytics ecosystem.
• The widespread adoption of foundational models: Generative and multimodal models could become the building blocks of many analytical systems. Data scientists will increasingly work with pre-trained models that can be adapted to specific contexts.
• Greater collaboration between humans and machines: Analytical tools will become capable of automatically generating hypotheses or data visualizations. The data scientist’s role will then be to validate, contextualize, and guide these analyses.

According to the World Economic Forum, jobs related to advanced data analysis and artificial intelligence are expected to remain among the most in-demand professions in the coming years, due to their strategic role in the digital transformation of organizations⁴.

In this environment, data scientists will no longer be merely technical experts. They will become data orchestrators, capable of linking the capabilities of algorithms to companies’ strategic priorities.

Toward an augmented industry that remains human

Artificial intelligence is profoundly transforming the way organizations use their data, but it does not change the purpose of that data. It accelerates the analysis of massive volumes of information, automates certain analytical tasks, and enables the detection of correlations invisible to the human eye. It is shifting the priorities of data science: less manual data preparation, more orchestration of complex analytical systems; less ad-hoc modeling, more oversight of algorithmic ecosystems capable of learning and evolving.

Yet, amid all this change, one thing remains constant: data-driven decision-making is still a deeply human process.

Augmented data science does not mean the complete automation of decision-making. It relies on the synergy between algorithmic intelligence and human judgment. Artificial intelligence models can analyze millions of data points, identify emerging trends, or generate predictive simulations. But it is the data scientist who interprets these results, understands their limitations, and places them within an economic, social, or strategic context.

This distinction is crucial. A data-driven decision is not limited to the mathematical optimization of a model. It involves organizational choices, economic impacts, and sometimes significant societal implications. It requires a thorough understanding of the context, the company’s objectives, and the risks associated with automation.

From this perspective, the role of the data scientist increasingly involves shaping the responsible use of artificial intelligence.

This includes, in particular:

• Rigorous validation of models prior to their operational deployment, to ensure their robustness in real-world environments.
• Continuous monitoring of algorithmic performance to detect deviations caused by changes in data or user behavior.
• Transparent governance of data and models, enabling the decisions made by analytical systems to be explained.
• Constant human oversight of high-stakes decisions, particularly in sensitive areas such as finance, healthcare, and public policy.

The rise of augmented data science also opens up significant opportunities. Advanced models can enhance our understanding of complex phenomena, optimize resource allocation, and inform strategic decisions in uncertain environments. They can help improve organizational efficiency, mitigate certain risks, and better anticipate economic shifts.

But this transformation goes far beyond the technological realm. It raises questions about the role of humans in a decision-making system where data is becoming ubiquitous. It requires a redefinition of the data scientist’s skill set—no longer limited to mastery of algorithms or programming languages, but encompassing the ability to manage complex analytical systems with discernment, responsibility, and strategic vision.

In a world where machines can generate analyses at high speed, the true value of a data scientist will not be measured by their ability to compete with algorithms, but by their ability to make sense of them.

The machine can generate predictions. The data scientist, however, must continue to ask the right questions.

What if, in the end, the artificial intelligence revolution in data science isn’t about replacing experts, but about revealing what lies at the heart of the profession: the ability to transform data into knowledge, and knowledge into informed decisions.

Learn more 

To broaden your perspective and understand how AI is reshaping other professions—from human resources to finance, and from healthcare to communications—we invite you to explore our dedicated section “AI & Professions”, which analyzes the concrete impact of intelligent technologies on skills, practices, and the organization of work.

References

1. International Data Corporation. (2021). The Global Datasphere Forecast.
https://www.idc.com

2. Gartner. (2022). AutoML and the Future of Data Science.
https://www.gartner.com

3. World Economic Forum. (2023). The Future of Jobs Report.
https://www.weforum.org

4. McKinsey & Company. (2022). The State of AI in 2022.
https://www.mckinsey.com