AI StrategyDigitalFrameworksLawler ModelLLMProduct Management

The Lawler Model for Designing AI Products: A BloombergGPT Case Study

Navigating the Labyrinth: A Deep Dive into The Lawler Model for Designing AI Products with a Focus on BloombergGPT

The world of Artificial Intelligence (AI) is a sprawling labyrinth of possibilities, buzzing with innovations that span various domains and disciplines. Within this sprawling maze, AI product managers find themselves at a crucial intersection where technology, ethics, data science, and user experience converge. However, navigating through this labyrinth to create a product that is not just technically advanced but also ethically sound, user-centric, and commercially viable can be an arduous journey. Here, a systematic roadmap can make a significant difference, and one such comprehensive framework is the Lawler Model for AI Product Development.

This article aims to serve as a primer on the Lawler Model, an intricate approach that enables the holistic development of AI-driven solutions. To bring this to life, we’ll focus on a real-world case study: the development of BloombergGPT, a financial industry-specific Large Language Model (LLM). This model, fine-tuned to the peculiarities and nuances of financial data, represents a monumental leap in the world of Natural Language Processing (NLP) as applied to finance.

Further, we shall also look into alternative approaches to AI product development, because while the Lawler Model is comprehensive, it’s not the one-size-fits-all solution. Each methodology has its unique strengths and weaknesses that may make it more suitable for specific types of projects.

The Lawler Model in a Nutshell

Before diving into the case study, it’s essential to outline the basic tenets of the Lawler Model. This model serves as a comprehensive framework that addresses five critical areas:

  1. Understanding the Problem Space: The first step in any project should be understanding the problem you are solving deeply. This step involves understanding the user needs and market gaps.
  2. Data Consideration: No AI model can function without data. This phase involves selecting the right kind of data, understanding its limitations, and preparing it for use in model training.
  3. Algorithmic Choices: This phase includes the selection of suitable algorithms and the architecture of the AI model.
  4. Ethical Concerns: The ethical implications of an AI model can’t be an afterthought. This phase ensures that the model conforms to ethical and legal norms.
  5. Iterative Design and Feedback Loop: Last but not least, the Lawler Model promotes an iterative approach to design, incorporating feedback from real-world usage to continually refine the product.

With these foundational elements in place, let’s examine how Bloomberg used the Lawler Model to guide the development of their groundbreaking BloombergGPT model.

By the end of this article, product managers and AI enthusiasts should have a clearer understanding of how to traverse the intricate labyrinth of AI product development, using the Lawler Model as a reliable compass and BloombergGPT as a beacon of successful implementation.

In the following sections, we will explore the Lawler Model’s application in the BloombergGPT project, analyze its performance metrics, and understand its strategic importance. We will also delve into alternative methodologies for AI product development to give you a well-rounded view of the landscape. Let’s begin this journey.

Expanding the Understanding of the Lawler Model and Alternative Approaches for AI Product Development

Deep Dive into the Lawler Model

The Lawler Model is often cited for its comprehensive, multi-faceted approach to AI product design, capturing the complexities of the development process. Here, we will delve deeper into its components:

1. Understanding the Problem Space

  • Needs Assessment: Identifying the real-world problem the AI model seeks to solve and assessing the need for an AI-based solution.
  • Stakeholder Mapping: Identifying all individuals or organizations who would be impacted by the product and including them in the consultation and feedback loops.

2. Data Consideration

  • Data Sourcing: Deciding where the data will come from and ensuring it is representative of the target audience.
  • Data Preprocessing: Cleaning and organizing the data for training purposes.

3. Algorithmic Choices

  • Algorithm Selection: Deciding on the algorithmic architecture (e.g., deep learning, decision trees, etc.) most appropriate for solving the problem.
  • Model Training: Actually building and training the AI model.

4. Ethical Concerns

  • Transparency: Creating a system that is understandable and can be interpreted by both technical and non-technical stakeholders.
  • Bias Mitigation: Putting into place mechanisms to detect and mitigate any biases in data or algorithms.

5. Iterative Design

  • Prototyping: Developing a minimum viable product (MVP) for initial testing.
  • User Testing: Gathering user feedback and making necessary adjustments.

6. Deployment and Monitoring

  • Launch Strategy: Deciding how the product will be rolled out, whether as a phased launch or full-scale deployment.
  • Ongoing Evaluation: Constantly monitoring for any issues that could arise post-launch, both from a technical and an ethical standpoint.

Alternative Approaches for AI Product Development

The process of developing an AI product is complex, requiring a multidisciplinary approach that combines data science, software engineering, and domain-specific expertise. While the Lawler Model is a comprehensive method for AI development, it’s not the only path. Below are some alternative frameworks, each with its unique advantages and limitations, that can serve as effective roadmaps for AI product development.

1. CRISP-DM (Cross-Industry Standard Process for Data Mining)

  • Overview: A robust, cyclic model that has stood the test of time in guiding data mining projects. Adaptable to various industries and use-cases.
  • Phases: Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment.
  • Best For: Projects that involve heavy data extraction, transformation, and modeling.

2. Agile Development

  • Overview: A flexible, iterative approach, Agile is renowned for its adaptability and responsiveness to changes.
  • Phases: Planning, Development, Testing, Review, and Adaptation.
  • Best For: Fast-paced development environments where product requirements can change rapidly.

3. Design Thinking

  • Overview: Design Thinking is particularly useful when the end-users’ needs are at the core of the product development.
  • Phases: Empathy, Define, Ideate, Prototype, and Test.
  • Best For: Customer-centric products where user experience is a significant concern.

4. The FAIR Model

  • Overview: Designed for open and collaborative data ecosystems, this model emphasizes data management and stewardship.
  • Phases: Data Discovery, Data Annotation, Data Integration, and Data Distribution.
  • Best For: Organizations that aim to share or publish their AI datasets or models.

5. Ethical AI Frameworks

  • Overview: These frameworks integrate ethical considerations into AI development from the ground up.
  • Examples: IEEE’s Ethically Aligned Design, EU’s AI Ethics Guidelines.
  • Best For: Projects where ethical considerations are paramount and must be addressed throughout the development process.
ApproachStrengthsLimitationsBest Use Cases
Lawler ModelComprehensive, End-to-EndMay be complex for smaller projectsFull-scope AI projects
CRISP-DMRobust, CyclicalMay be too rigidData-intensive projects
AgileFlexible, AdaptiveLess structureFast-paced development
Design ThinkingUser-centricNot always data-focusedUX-focused products
FAIR ModelData stewardshipNarrow scopeOpen data ecosystems
Ethical AIEthical focusLimited technical guidanceEthically-sensitive projects

Final Thoughts On AI Product Frameworks

The Lawler Model, known for its comprehensive, end-to-end approach, serves as an excellent base for those who are looking for a holistic methodology for AI product development. While it’s highly effective for full-scope AI projects, it might be considered complex for smaller projects. This underscores the importance of context in selecting a framework for AI product development.

Whether you’re working on a data-intensive project and considering CRISP-DM, or you’re developing an AI product with user experience at the forefront and considering Design Thinking, the key is to align the framework with the specific needs of your project. For projects requiring high ethical standards, Ethical AI frameworks should not be overlooked.

By understanding the strengths and limitations of each framework, organizations can make informed decisions that lead to successful AI deployments, meeting both technical and ethical benchmarks.

Application in AI Product Design: BloombergGPT

Incorporating frameworks like the Lawler Model into the real-world development of AI products requires an understanding of their practical applications. Below is an illustrative case study of BloombergGPT that demonstrate how the Lawler Model and other approaches have been effectively utilized for AI product development.

Bloomberg LP has developed its own AI model, named Bloomberg GPT, leveraging the same foundational technology as OpenAI’s GPT models. This model is designed to serve financial professionals through Bloomberg’s terminal software. Bloomberg GPT specializes in understanding and answering financial queries, assessing the sentiment of headlines, and even generating new headlines. The model utilizes both general data and Bloomberg’s proprietary financial data for more accurate task performance in the financial domain.

Gideon Mann, the head of ML Product and Research at Bloomberg, stated that the rise of GPT-3 had a significant impact on their approach to natural language processing (NLP). Unlike the tech giants that primarily dominate the AI industry, Bloomberg has demonstrated that specialized, industry-specific applications of AI are not only possible but also promising. The model doesn’t rely on OpenAI’s architecture but uses open-source AI technologies and the company’s extensive proprietary data, thus challenging the idea that only big tech firms with generalized data can excel in AI development.

Bloomberg plans to integrate its GPT model into features and services accessed via its Terminal product. Potential applications range from translating human queries into database-specific language, to cleaning data and performing other backend tasks. The initiative aims to help financial professionals manage the overwhelming amount of news and data they encounter daily.

BloombergGPT—The Financial Sector’s Next Big Leap

The world of finance is a complex landscape, driven by data and dictated by numbers. It’s an industry that thrives on real-time decisions, intricate calculations, and endless paperwork. Enter Artificial Intelligence (AI), a disruptive force that has transformed multiple sectors but has always had a somewhat uneasy relationship with finance. The reason? The unique language, jargon, and rapid pace of the financial world have long been considered too specialized for AI to handle effectively. That is, until now. Bloomberg’s new brainchild, BloombergGPT, seems poised to redefine this narrative. Designed to be more than just a language model, BloombergGPT aims to be the financial sector’s most reliable AI companion.

A Leap Forward in Financial NLP

Bloomberg has been a pioneering force in applying AI, machine learning, and natural language processing to the world of finance. With the release of BloombergGPT, the company has made significant strides in domain-specific language modeling, aimed to revolutionize various NLP tasks in finance.

Leveraging the Lawler Model in BloombergGPT Development

1. Understanding the Problem Space Bloomberg identified the unique challenges that come with processing financial information, such as sentiment analysis, named entity recognition, and news classification. The complexity of financial terminology and the sheer volume of financial data necessitated a domain-specific language model.

2. Data Consideration For training BloombergGPT, the company amalgamated a colossal 363 billion token dataset consisting of English financial documents, spanning over 40 years. This was combined with a 345 billion token public dataset, resulting in a massive training corpus of over 700 billion tokens.

3. Algorithmic Choices The team trained a 50-billion parameter decoder-only causal language model. The focus was on constructing a model that could excel at finance-specific tasks while remaining competitive in general-purpose NLP tasks.

4. Ethical Concerns Financial markets are rife with regulations and ethical pitfalls. BloombergGPT was designed with compliance and ethical usage in mind, ensuring that it would not lead to market distortions or enable unfair trading practices.

5. Iterative Design Bloomberg’s model underwent rigorous validation on finance-specific benchmarks as well as a suite of internal Bloomberg benchmarks. This iterative design and validation process allowed the model to adapt and optimize continually.

6. Deployment and Monitoring BloombergGPT has been integrated into the Bloomberg Terminal to enhance various financial NLP tasks. The company has a robust monitoring and updating protocol to ensure that the model remains ethical, efficient, and effective.

7. Performance Highlights Notably, BloombergGPT has been found to significantly outperform similarly-sized open models on financial NLP tasks, without any compromise on general-purpose benchmarks. This achievement marks a milestone, proving that domain-specific models can be designed without sacrificing broader utility.

8. Strategic Importance and Future Outlook Bloomberg’s CTO, Shawn Edwards, and the Head of ML Product and Research, Gideon Mann, both emphasized the immense value that BloombergGPT brings in terms of performance, scalability, and time-to-market advantages.

The BloombergGPT Revolution

Key Features

Multifaceted Applications

BloombergGPT is not a one-trick pony. Its capabilities extend far beyond basic data analysis.

  • Sentiment Analysis: Understanding market sentiments can be the difference between a profitable trade and a disastrous one. BloombergGPT excels in sifting through mountains of text data to gauge the mood of the market.
  • Named Entity Recognition: In finance, the devil is in the details. Knowing which stock is trending or which index is faltering can offer invaluable insights. BloombergGPT can identify these critical financial entities with ease.
  • Question and Answering: Time is money, and in finance, answers are needed in real-time. BloombergGPT’s robust question-answering capabilities can serve as a real-time consultant for data-driven decisions.


Generic language models are often ill-suited to the financial world’s unique terminology. BloombergGPT, however, speaks the language of finance fluently.


The financial landscape is ever-changing, and a rigid model would quickly become obsolete. BloombergGPT is designed to adapt and scale according to the industry’s needs.

Challenges and Solutions

  • Data Sensitivity: Security is paramount in finance. Recognizing this, BloombergGPT has been built with stringent data privacy measures.
  • Real-Time Adaptability: Finance never sleeps, and neither does BloombergGPT. It continuously updates its knowledge base to provide the most current and reliable insights.

The Collaborative Endeavor

Bloomberg didn’t go it alone in developing BloombergGPT. The project is the result of a collaborative effort involving experts from Bloomberg, Johns Hopkins University, and other institutions. This interdisciplinary approach has resulted in a tool that is not just technologically advanced but also deeply attuned to the specific needs of the financial sector.

Future Implications

  • Automated Trading: BloombergGPT could potentially take over routine trading tasks, allowing human traders to focus on strategy.
  • Fraud Detection: With its pattern recognition capabilities, BloombergGPT is well-positioned to identify and flag fraudulent activities.
  • Customer Service: Personalized and efficient customer service is no longer a luxury but a necessity. BloombergGPT could redefine customer interaction in the financial sector.

The CDO TIMES Bottom Line

The advent of BloombergGPT is more than just a technological marvel—it’s a shift in how we understand the application of AI in finance. It represents the culmination of years of research, collaboration, and fine-tuning to create a tool that doesn’t just understand finance but thrives in it. As CDOs and tech leaders evaluate tools to bring their operations into the future, BloombergGPT stands as a compelling benchmark of what is achievable. It’s not just about automating tasks; it’s about enhancing decision-making, improving security, and ultimately, driving profitability. In a nutshell, BloombergGPT is not the future of financial technology; it is the technology that will build that future.

The Lawler Model presents a balanced approach to AI product design by integrating technical, ethical, and social considerations into a holistic framework. As AI continues to permeate every facet of our lives, frameworks like the Lawler Model are invaluable for developing products that are not just technically sound but also ethically responsible and socially beneficial.

Whether you’re a product manager, a developer, or an executive deciding on your company’s next AI venture, understanding and applying the Lawler Model could make the difference between launching a successful, ethical product and one that falls short.

Love this article? Embrace the full potential and become an esteemed full access member, experiencing the exhilaration of unlimited access to captivating articles, exclusive non-public content, empowering hands-on guides, and transformative training material. Unleash your true potential today!

In this context, the expertise of CDO TIMES becomes indispensable for organizations striving to stay ahead in the digital transformation journey. Here are some compelling reasons to engage their experts:

  1. Deep Expertise: CDO TIMES has a team of experts with deep expertise in the field of Digital, Data and AI and its integration into business processes. This knowledge ensures that your organization can leverage digital and AI in the most optimal and innovative ways.
  2. Strategic Insight: Not only can the CDO TIMES team help develop a Digital & AI strategy, but they can also provide insights into how this strategy fits into your overall business model and objectives. They understand that every business is unique, and so should be its Digital & AI strategy.
  3. Future-Proofing: With CDO TIMES, organizations can ensure they are future-proofed against rapid technological changes. Their experts stay abreast of the latest AI advancements and can guide your organization to adapt and evolve as the technology does.
  4. Risk Management: Implementing a Digital & AI strategy is not without its risks. The CDO TIMES can help identify potential pitfalls and develop mitigation strategies, helping you avoid costly mistakes and ensuring a smooth transition.
  5. Competitive Advantage: Finally, by hiring CDO TIMES experts, you are investing in a competitive advantage. Their expertise can help you speed up your innovation processes, bring products to market faster, and stay ahead of your competitors.

By employing the expertise of CDO TIMES, organizations can navigate the complexities of digital innovation with greater confidence and foresight, setting themselves up for success in the rapidly evolving digital economy. The future is digital, and with CDO TIMES, you’ll be well-equipped to lead in this new frontier.

Do you need help with your digital transformation initiatives? We provide fractional CAIO, CDO, CISO and CIO services and have hand-selected partners and solutions to get you started!

We can help. Talk to us at The CDO TIMES!

Subscribe now for free and never miss out on digital insights delivered right to your inbox!

Don't miss out!
Subscribe To Newsletter
Receive top education news, lesson ideas, teaching tips and more!
Invalid email address
Give it a try. You can unsubscribe at any time.

Carsten Krause

As the CDO of The CDO TIMES I am dedicated delivering actionable insights to our readers, explore current and future trends that are relevant to leaders and organizations undertaking digital transformation efforts. Besides writing about these topics we also help organizations make sense of all of the puzzle pieces and deliver actionable roadmaps and capabilities to stay future proof leveraging technology. Contact us at: to get in touch.

Leave a Reply


Discover more from The CDO TIMES

Subscribe now to keep reading and get access to the full archive.

Continue Reading