#6 Lexis+AI Review + AI for PE Funds
This week, I review Lexis+AI by LexisNexis, discuss AI hallucinations, and provide information about AI and the business side of M&A.
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.
M&Ai Tool Review
As a law student I have free access to the LexisNexis AI tool “Lexis+AI,” so I thought I would share it here. Overall, it is pretty cool to use, but it has limited capabilities, as you can see in the second prompt I give below.
Here, I gave the model a question about cases that are important to consider in crafting a MAE provision in a purchase agreement.
What are the most important Delaware cases to consider when crafting a material adverse effect provision in an M&A purchase agreement?
A material adverse effect (MAE) provision in a merger and acquisition (M&A) purchase agreement is a critical component that addresses the risk of a significant deterioration in the value of the seller's business between signing and closing. It is primarily concerned with a change in valuation, irrespective of any change in how the business is being operated AB Stable VIII LLC v. Maps Hotels & Resorts One LLC, 2020 Del. Ch. LEXIS 353.
In Delaware, the interpretation of MAE provisions has been addressed in several cases. For instance, in the case of Akorn, Inc. v. Fresenius Kabi AG, it was held that the burden of proof with respect to a material adverse effect in a merger agreement rests on the party seeking to excuse its performance under the contract Akorn, Inc. v. Fresenius Kabi AG, 2018 Del. Ch. LEXIS 325.
In the case of 3M Co. v. Neology, Inc., it was clarified that a buyer faces a heavy burden when it attempts to invoke a material adverse effect clause. A material adverse effect is generally an unexpected event or series of events that threatens a business's overall earnings in a manner. It is not a short-term hiccup in earnings, but a material change when viewed through a reasonable acquirer's long-term perspective 3M Co. v. Neology, Inc., 2019 Del. Super. LEXIS 312.
In the case of Bardy Diagnostics, Inc. v. Hill-Rom, Inc., it was held that whether a material adverse effect exists is ultimately a factual issue. The court will look to whether there has been an adverse change in the target's business that is consequential to the company's long-term earnings power over a commercially reasonable period Bardy Diagnostics, Inc. v. Hill-Rom, Inc., 2021 Del. Ch. LEXIS 146.
In the case of Boardwalk Pipeline Partners, LP v. Bandera Master Fund LP, the court noted that the well-known case law in Delaware interpreting 'material adverse effect' is primarily focused on the use of that term in connection with contracts governing mergers and acquisitions Boardwalk Pipeline Partners, LP v. Bandera Master Fund LP, 288 A.3d 1083.
These cases provide important guidance on how to craft a material adverse effect provision in an M&A purchase agreement under Delaware law. However, the specific language and structure of the provision will depend on the unique circumstances of each transaction.
As you can see, the model (I think) did a good job of finding cases related to the MAE provision. I am somewhat familiar with the MAE provision, and I recognized some of the cases the model answered with as important in the MAE world (in particular the Akorns case, which was the first time a Delaware court enforced a MAE provision).
Here is an example of a prompt where I ask the model for advice in negotiating reps and warranties in an M&A deal. As you can see, the answer is less than satisfactory.
I am doing an M&A deal involving one Delaware limited partnership, the buyer, and one Delaware corporation, the target. It is a stock purchase, with the buyer taking a 25% interest in the target. I represent the buyer. What issues do I need to look out for while negotiating the representations and warranties in the stock purchase agreements? Give me a list of the 3 main issues, using bullet points for each issue.
In the context of a stock purchase agreement involving a Delaware limited partnership (buyer) and a Delaware corporation (target), there are several key issues to consider while negotiating the representations and warranties.
Firstly, the representations and warranties act as a benchmark for the buyer's investment in the target. The buyer is essentially stating that they are willing to pay the agreed upon purchase price if the business has the attributes described in the representations and warranties of the target set forth in the merger agreement. This is illustrated in the case of Akorn, Inc. v. Fresenius Kabi AG, where the plaintiff's request for specific performance of a merger agreement was denied because the defendant validly terminated the merger agreement due to inaccuracies in the plaintiff's regulatory compliance representations, which resulted in a material adverse effect (MAE) Akorn, Inc. v. Fresenius Kabi AG, 2018 Del. Ch. LEXIS 325.
Secondly, the Stock Purchase Agreement can be ambiguous with respect to indemnification claims against the sellers for allegedly fraudulent contractual representations of the company. This was the case in EMSI Acquisition, Inc. v. Contrarian Funds, LLC, where the auditor's findings did not constitute an arbitration award that was subject to "confirmation" under the Delaware Arbitration Act EMSI Acquisition, Inc. v. Contrarian Funds, LLC, 2017 Del. Ch. LEXIS 73.
Thirdly, the Purchase Agreement may contain 'nonreliance' language, which can impact the ability to prove reasonable reliance to establish misrepresentation claims. For instance, the Purchase Agreement may contain clauses that disclaim any representations or warranties with respect to the businesses, assets, liabilities, obligations, or prospects of any of the Purchased Subsidiaries or their Subsidiaries § 8.144 Liability-Managing Provisions.
In addition, it is important to consider the Material Adverse Change Definitions, Covenants in Acquisition Agreements, and Disclosure Schedules: Preparation and Review. These are key elements in the negotiation of representations and warranties in acquisition agreements DE Chancery Court Issues Rare Decision Enforcing MAE Clause: Client Alert Digest.
Lastly, it is crucial to be aware of the specific regulations and statutes that govern the entities involved in the transaction. For instance, 6 Del. C. § 17-1101 outlines the limitations or elimination of liabilities for breach of contract and breach of duties of a partner or other person to a limited partnership § 17-1101. Construction and application of chapter and partnership agreement..
Due to my lack of M&A knowledge, the model’s answer seems pretty good. It certainly points out some things that (I assume) are important to remember in crafting representations and warranties based on Delaware case law. However, it missed my instruction to give me 3 points using bullet points and did not give an answer that was slanted toward the buyer’s position. I asked a follow-up question to try to elicit advice explicitly in favor of the buyer, but the model spit out an answer similar to the first, except it gave “pro-buyer forms” as a source. It did not give any real advice on the negotiating position of the buyer.1 This may be because the model is slanted towards answering purely legal questions, like the first example.
Overall, this is a cool tool to have access to as a law student, and I think it is the very beginning of what is to come in the legal AI tools space, so there is only room to improve!
AI Hallucinations
Here is a link to a Bloomberg Law article about a study on AI’s accuracy in answering legal questions. For those unfamiliar, a “hallucination” in the AI world is where an AI model makes a mistake, either by giving inaccurate information or by making something up.2 The study found that when asked legal questions, popular models (like ChatGPT) hallucinated 75% of the time. It may not come as a huge surprise, as AI legal mistakes have made it into national news several times over the past few months.3 Still, it is pretty alarming and goes to show that the popular AI models cannot be trusted when asked legal questions.
A plausible explanation for this is that ChatGPT and other popular models are trained on a ton of general information. This means they are good at making a poem, like this one: “Silicon alliances bloom/ AI whispers in boardroom/ Mergers dance with binary plume”4 (okay, maybe not that great at poems). But not so good at giving litigants citations to cases that support their position (just ask Michael Cohen). A model (like Lexis+AI) that is trained on caselaw and other legal resources will be infinitely better than ChatGPT.
AI Business Tools
Here is an interview with famed private equity investor and founder of Carlyle, David Rubenstein. Rubenstein talks about AI’s effect on private equity and investing towards the end of the interview. Rubenstein thinks that, despite the use of AI, investors will still stick to their gut when making investment decisions—they will use AI as a tool to assist in the deal-making process, but the ultimate investment decision remains with the humans. Rubenstein believes that AI will significantly streamline dealmaking by allowing buyers to quickly evaluate potential targets.
I think this is an insightful take by someone who has clearly done his research on AI. It seems like he agrees with my point in an earlier post that AI will be a tool. Business, just like the law, is a relationship business, and until AI can replace human relationships, dealmakers and lawyers will continue to rely on other humans to make big decisions.
Speaking of AI business tools, Josh Kubicki recently included Metal in one of his Brainyacts newsletters. Metal is an AI tool that is designed to help private equity firms evaluate potential targets. Metal’s website is pretty basic right now, but it appears that Metal is trained (or trainable) on the financial data of several companies and can instantaneously compare the targets. Metal claims that its model will be helpful to everyone from junior analysts to the C-suite. The site says the model is in its early stages and the model is available to funds upon request.
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: www.linkedin.com/in/parker-w-lawter-58a6a41b
All views expressed are my own!
There may be some user error here, so if you can think of a better prompt, please let me know and I will try it!
Definition of AI hallucination
Perhaps ChatGPT is not so great at poems involving AI and M&A, but either way. The poem was what ChatGPT gave me in response to the prompt, “Write a 3-line poem about AI and mergers and acquisitions.”