This week, I review Paxton AI, an exciting AI legal tech tool, and talk about prompt engineering.

AI will change the world, but how will it change M&A? I want to focus on AI’s impact on M&A in this newsletter. I am not an expert on either M&A or AI, but I want to learn about both topics and how they intersect. I thought there may be others in my situation (or people who are experts in one field or the other) who would find information on M&A and AI helpful in their careers, so I created this newsletter to track and share what I learn.

Subscribe now

M&AI Tool Review—Paxton AI

For the past week or so I have been using Paxton AI, a legal tech AI tool that I discovered at the ABA Tech Show.

Paxton AI has several tools, including legal research functionality, document drafting, file analysis, and a redlining tool.

Here is a video that gives an overview of all of Paxton AI’s capabilities.

File Analysis

I have spent the most time using the file analysis function of Paxton.

Basically, the user can upload a file to Paxton and the user can ask Paxton questions about it. Here are some screenshots from a recent interaction where I uploaded a random merger agreement I found on the SEC database:

Paxton did a fair job identifying the obligations of the Parent in the merger agreement. I know I’ve said this a lot on here, but lawyers must be sure to read everything when using an AI tool to make sure that it is correct. I have experienced a couple of instances where relying fully on the AI summary was not 100% accurate, so it is super important to double-check. Thankfully, Paxton makes this very easy by linking to the appropriate parts of the uploaded document (see the right side of the screenshots).

Here, I asked Paxton about a model business code provision. Paxton found the right provision in the code and gave the correct answer to my question.

Impact on M&A

I think Paxton can help M&A practitioners by streamlining document analysis and providing quick answers to legal questions that may come up in day-to-day practice. The file analysis capability will be very helpful because it will allow users to upload a long document and ask the AI questions as I demonstrated above. This tool will be best used to quickly get caught up with long documents, perhaps in anticipation of a client call or negotiation.

Security

I was lucky enough to talk to the founders of Paxton AI at the ABA Tech Show. They told me that they can make Paxton run 100% internally on a firm’s cloud system or internal servers (or both!). This eliminates any potential security risk from using client data in the model.

Conclusion

Paxton AI is a great legal tech tool and promises to streamline a lot of legal work. This tool stands out because of its broad capabilities. I really enjoyed meeting the founders at Tech Show and I think there is a promising future for Paxton AI’s use in law firms. If you are interested in trying Paxton AI, they offer a free 14-day trial on their website.1

AI Basics—Prompting

Prompting is one of the most important aspects of using AI chatbots like Copilot, ChatGPT, and others. It is essential to know and practice good prompting.

“Prompting” refers to the input into an AI chatbot. Good prompters can get the AI model to do what the prompter wants. Bad prompters input a prompt and wonder why the chatbot replies with irrelevant information.

I thought it would be interesting to see what Copilot would say when I asked it to teach me to prompt.

Prompt:

I want to learn how to become a prompt engineer for AI chatbots. I will want you to act as my teacher and teach me how to prompt. I will want multiple lessons on this topic. First, give me 5 bullet points on the basics of prompting. Explain it on a 12th-grade level.

Response:

Sure, I’d be happy to help you understand the basics of prompting for AI chatbots. Here are five key points:

  1. Understanding the AI Model: It’s important to know how the AI model works. Most AI chatbots use machine learning models that are trained on large amounts of text data. They generate responses based on patterns they’ve learned from this data.

  2. Clarity of the Prompt: The prompt should be clear and specific. Vague prompts can lead to ambiguous responses from the AI. It’s also important to consider the user’s perspective when crafting prompts.

  3. Contextual Relevance: The prompt should be relevant to the context of the conversation. This includes considering the user’s previous messages and the overall topic of the conversation.

  4. Safety and Ethics: Prompts should be designed with safety and ethics in mind. They should not encourage harmful behavior or spread misinformation. It’s also important to respect the user’s privacy and not ask for personal information.

  5. Iterative Improvement: Prompt engineering is an iterative process. It involves testing the prompts, analyzing the AI’s responses, and refining the prompts based on the results.

Remember, these are just the basics. There’s a lot more to learn about prompt engineering, including advanced techniques and best practices. But understanding these fundamentals is a great place to start!

Prompt:

That is great! Now teach me some more advanced techniques that will allow me to obtain the information that I want from an AI chatbot when I prompt it.

Response:

Absolutely, let’s delve into some advanced techniques for prompting AI chatbots:

  1. Prompt Phrasing: The way you phrase your prompt can significantly influence the AI’s response. For instance, open-ended questions can lead to more creative responses, while specific questions can help you get precise information.

  2. Prompt Length: Longer prompts can provide more context, which can be helpful for complex queries. However, shorter prompts are often better for simple queries as they’re more likely to get straight to the point.

  3. User Modeling: This involves crafting prompts that take into account the user’s knowledge, interests, and needs. By understanding the user better, you can create prompts that are more engaging and relevant.

  4. Multi-turn Conversation: Instead of trying to get all the information in one prompt, consider breaking it down into a multi-turn conversation. This can make the interaction feel more natural and can often lead to better results.

  5. Experimentation: Don’t be afraid to experiment with different prompts. AI models can sometimes surprise you with their responses, and experimenting with different prompts is a great way to learn more about how the model works.

  6. Feedback Loop: Incorporate a feedback loop where you analyze the effectiveness of your prompts and make necessary adjustments. This continuous learning and adapting process is crucial for improving the performance of your AI chatbot.

Remember, prompting is both an art and a science. It requires creativity, empathy, and a deep understanding of both the AI model and the user. Happy prompting!

Prompt:

Give me an example of a complex prompt that would extract information about the impact of AI on M&A.

Response:

Sure, here’s an example of a complex prompt that could be used to extract information about the impact of AI on M&A:

“Can you provide a detailed analysis on how AI is transforming the M&A landscape? Specifically, I’m interested in understanding how AI is being used in deal sourcing, due diligence, valuation, and post-merger integration. Additionally, could you discuss the challenges and ethical considerations that come with using AI in this context?”

This prompt is designed to extract comprehensive information about the impact of AI on various stages of the M&A process. It also encourages the AI to consider the potential challenges and ethical implications of using AI in M&A.

Based on the search results12345, AI is indeed transforming the M&A landscape in various ways:

  1. Deal Sourcing: AI can help build a very complete long list and scan the market for potential targets2.

  2. Due Diligence: AI streamlines due diligence by generating data, automating data analysis, spotting inconsistencies, risk assessment, and decision-making25.

  3. Valuation: AI-driven valuation models offer more accurate assessments and predictive insights2.

  4. Post-Merger Integration: Integration planning and execution is enhanced through automated workflows, intelligent systems, and cultural alignment facilitated by AI2.

However, there are challenges and ethical considerations to keep in mind. These include access to data, data privacy, bias, regulatory compliance, and implementation costs2. AI is only as good as the underlying data, and it’s important to be aware of potential biases, errors, or inconsistencies in the data1. Despite these challenges, AI is poised to continue transforming M&A, offering enhanced efficiency, risk mitigation, and new deal opportunities2.

Conclusion

Prompting is an important skill to develop for anyone interested in using AI in daily life. My non-expert advice is to create prompts and see what happens. Prompting is a trial-and-error process and changing even one word can change the AI’s output. Thanks for reading!

About me

My name is Parker Lawter, and I am a law student pursuing a career as an M&A lawyer. I am in my last semester of law school, and with some extra time on my hands, I decided to create this newsletter. I hope it is informative and helpful to anyone who reads it! I am not an expert at either M&A or AI, but I am actively pursuing knowledge in both areas, and this newsletter is a part of that pursuit. I hope you’ll join me!

Follow me on LinkedIn!

All views expressed are my own!

1

Note that I do not think the security features work on the website version of Paxton.

Photo of Parker Lawter Parker Lawter

My name is Parker Lawter, and I am a law student pursuing a career as an M&A lawyer. I am in my last semester of law school, and with some extra time on my hands, I decided to create this blog. I hope…

My name is Parker Lawter, and I am a law student pursuing a career as an M&A lawyer. I am in my last semester of law school, and with some extra time on my hands, I decided to create this blog. I hope it is informative and helpful to anyone who reads it! I am not an expert at either M&A or AI, but I am actively pursuing knowledge in both areas, and this newsletter is a part of that pursuit. I hope you’ll join me!