TaxGPT: Advancing AI and the Tax Profession

Introduction

Welcome to the latest edition of "In the Pages"! In this episode, we will explore the impact of artificial intelligence (AI) on the future of legal research, with a focus on tax law. How will tax professionals utilize AI in their research? Will they rely on AI devices for answers, or continue with traditional methods? These are the questions we aim to address in this discussion.

The featured article, "The Rise of Generative AI in Tax Research," written by Benjamin Ellery, Kim Condon, Susan Massey, and Christopher Yan, examines the role of generative AI in the field of tax research. We are fortunate to have two of the co-authors, Benjamin Ellery and Susan Massey, join us today. Benjamin Ellery is the Professor of Business Law at the University of Toronto, as well as the CEO of Blue Jay Legal. Susan Massey is the Vice President of Legal Research with Blue Jay Legal.

Let's begin with a basic understanding of generative AI and chatbots. Generative AI refers to a type of technology that can create new artifacts, such as images, text, audio, and video. It uses machine learning and neural networks to generate creative outputs based on training data. On the other hand, chatbots are automated systems that engage in text-based conversations with users. Many chatbots now utilize generative AI to provide realistic and interactive dialogue.

Generative AI systems, like chatbots, are designed to be user-friendly, even for individuals without a tech background. Users can simply input their questions or prompts, and the AI system will generate responses based on its training data. While some prompt engineering may be necessary to ensure accurate and relevant answers, the overall process is intuitive and accessible.

The article explores the use of various tools, including established chatbots, in tax research. The authors experimented with these tools by asking them the same research question, and they found differences in the responses received. One tool, Ask Bluejay, stood out in terms of providing more accurate and relevant responses compared to other chatbots. Ask Bluejay is specifically tailored to tax research and employs careful curation of information.

However, it's important to note the limitations and pitfalls of generative AI. One challenge is the potential for inaccuracies and hallucinations in the information provided by AI systems. This means that the AI may generate incorrect or fabricated answers, making it essential for users to verify the information independently. Lack of transparency regarding the sources used by AI systems is another concern, as it restricts the ability to trace and validate the information provided.

In the context of legal research, accurate sourcing is crucial, as legal professionals require precise information about cases, statutes, and regulations. While current chatbots may have limitations in this regard, it is possible to design AI systems that rely only on trusted and authoritative sources. Curating and grounding these systems with accurate and reliable reference materials is key to ensuring their usefulness and reliability.

Looking ahead, AI holds great potential for advancing the tax profession. The next wave of AI in tax research could involve more advanced analysis and language generation capabilities. For example, AI systems could provide scenarios and analysis for complex transactions, applying tax laws and regulations to offer optimal solutions. The role of tax professionals will still be crucial as they provide oversight, expertise, and problem-solving skills that AI cannot replace.

Keyword: generative AI, chatbots, tax research, accuracy, reliability, legal profession, sourcing, AI limitations, advanced analysis, tax professionals.

FAQ:

1. Can tax professionals without a tech background use generative AI?

• Yes, generative AI and chatbot systems are designed to be user-friendly, even for individuals without a tech background. Prompt engineering may be required for accurate results, but overall, the systems are accessible to all users.

2. How reliable are the responses from chatbots and generative AI?

• While chatbots and generative AI can provide useful information, there are limitations and potential inaccuracies. Users must independently verify the information and be cautious of potential hallucinations or fabricated responses from the AI system.

3. Will AI replace tax professionals in the future?

• AI is a powerful tool that can assist tax professionals in their work, but it is unlikely to replace them entirely. Tax professionals bring expertise, problem-solving skills, and critical thinking to the table, which AI cannot replicate. AI will be a valuable ally in enhancing productivity and efficiency in tax research and analysis.

4. How can AI systems ensure accurate and reliable sourcing of information?

• AI systems can be designed to rely on trusted and authoritative sources of information, such as regulations, statutes, and rulings. By curating and grounding the AI system with verified sources, the accuracy and reliability of the generated responses can be enhanced.

5. What is the next wave of AI advancement in tax research?

• The next wave of AI in tax research may involve more advanced analysis and language generation capabilities. AI systems could provide detailed scenarios, analysis, and solutions for complex tax transactions, optimizing tax advantages while considering compliance and risk factors.