How Do You Use AI?
Use Cases for Venture Funds and Law Firms
In recent years, venture capitalists (VCs) have increasingly integrated artificial intelligence (AI) into their investment strategies to enhance decision-making, from discovering high-potential startups to effectively managing their portfolios. According to Gartner, by 2025, more than 75% of venture capital and early-stage investor executive reviews will be informed using AI and data analytics.
Dozens of VCs now use AI technologies to screen startups – sourcing, evaluating, and selecting those to fund – through their proprietary platforms that automatically track and score startups based on their future return prospects. By employing algorithms that detect quantitative patterns in historical data from previous startups, VCs can extrapolate these patterns to predict the outcomes of new startups, thereby making more informed and strategic investment decisions.
Similarly, law firms are using AI to improve efficiency and offer faster, more informed work product. According to multiple legal news outlets, about 30% of large law firms are now using generative AI (meaning deep-learning models that can generate high-quality text, images, and other content based on training data) for legal matters. In addition, nearly all large law firms expect their investment in generative AI to increase over the next five years, and more than 70% of lawyers plan to use generative AI in their legal work within the next year. However, an overall trend of the role of AI in these areas is that it is best seen as a way to augment or increase the efficiency of a lawyer’s work – not as a tool to replace it.
This article explores venture fund and law firm use cases and compares and contrasts them.
ABOUT THE AUTHORS
Research assistance from Nahom Yiman.
Sophia Zhao
Partner, AI FundSophia brings a wealth of experience in capital advisory, corporate development, and operational optimization, establishing impactful collaborations with CXOs and Founders. With a diverse industry exposure encompassing cloud computing, mining and minerals, consumer goods, and Web3, Sophia has been at the forefront of transformative technologies.
Tracy Rubin
Partner, Cooley LLPTracy Rubin is a partner in Cooley LLP’s Technology Transactions Group. Her practice focuses on counseling innovative technology companies on complex and transformative intellectual property transactions. She has a particular passion for complex asset deals, mergers and acquisitions, and diving deep on the latest and most transformative technologies.
Brooke Fritz
Special Counsel, Cooley LLPBrooke Fritz has over fifteen years of experience drafting and negotiating complex technology transactions and counseling clients as a valued business advisor. She focuses on her clients' legal and business needs in connection with life sciences and technology transactions. Her clients include multinational public companies and start-ups financed by venture capital firms in the IT, telecommunications, services, health care, business process outsourcing, and other technology industries.
USING AI AT A VENTURE FUND
Deal Sourcing and Prioritization
VCs receive an extensive influx of both solicited and unsolicited pitches daily, from a broad spectrum of sources such as VC referrals, founder recommendations, social media initiatives, cold emails, interactions at conferences or events, demo-day engagements, and referrals from their general network. The process of filtering, organizing, and ultimately prioritizing these leads demands significant effort and time.
VCs have begun using AI co-pilots to help solve these challenges by swiftly processing this voluminous data and automatically categorizing information from diverse channels into a structured format ready for analysis. This enables the creation of a customizable dashboard, where criteria can be adjusted and filtered according to the user’s preferences. Such a tool not only streamlines the initial evaluation process but also enhances the efficiency of identifying and prioritizing potential investment opportunities, allowing VCs to focus their efforts on the most promising prospects.
Additionally, some VCs are utilizing, for both internal and external virtual meetings, AI-enabled notetaker “assistants” to automatically transcribe – and also later summarize – highlights and contextualize conversations for faster filtering. However, there are notable risks tied to the reliance on AI for these tasks. These risks include issues with the accuracy of transcription and summarization, where the AI might fail to capture all information or incorrectly interpret the context. This is particularly problematic in meetings featuring complex terminology, multiple speakers, diverse accents, or overlapping dialogue. Another significant concern is the potential for bias in AI algorithms, which could be influenced by the data on which they are trained. This might affect how information is filtered and presented, leading to a skewed or partial representation of the discussions.
Portfolio Monitoring and Relationship Management
A customer relationship management (CRM) platform serves as a comprehensive repository for a firm’s engagements with companies, encompassing past, present, and anticipated interactions. Each relationship is documented within a profile, containing essential information such as contact details, interaction history, recent funding, and call notes, organized primarily by the status within the sales pipeline.
The integration of CRM-native semantic search could significantly expedite the identification of contextually relevant relationships and opportunities, while the inclusion of an assistant facilitates ongoing conversations and follow-ups on specific queries. For instance, an investor planning a trip to Miami might query, “Who is involved in data infrastructure in Miami?” Or, before attending a conference, the investor might ask, “Which founders I met in the last six months are likely attending the next AI conference?” – receiving a response by retrieving mentions of AI/graphical processing unit (GPU) technologies in call notes.
Moreover, as part of portfolio monitoring, investors typically gather new information through direct interactions, backchannel communications, and publicly available data online. Implementing automated agents to continuously monitor the internet for relevant updates (such as developer platform activities, press releases, social media updates, team growth, community engagement, and customer announcements) provides a valuable tool exceeding manual research. These agents alert investors to key developments by identifying milestones that adjust dynamically based on the investor’s preferences and CRM activity. This approach not only broadens an investor’s operational bandwidth but also enables them to sustain relationships that might otherwise be neglected due to other priorities.
Research and Analysis
Research to thoroughly comprehend a company is key aspect of due diligence. During this process, VCs need to collect diverse intelligence on the company, competitors, market, and comparable transactions. This involves a deep dive into the company’s data room, industry reports, news articles, competitor websites, and previous transactions. The process can be immensely time-consuming and occasionally suffers from limited scope, particularly when there’s insufficient time to conduct extensive research.
A key advantage of using AI to consume external information is aggregate analysis. Chat-based large language models (LLMs) that are optimized for academia/research analyze research materials across various sources and formats, leveraging the provided files as the primary knowledge base.
Semantic search is a core enabler for market research providers as it looks for matches with the meaning of words vs. literal matches. However, the volume of data supplied by these platforms makes manual analysis very time consuming. Investors complement semantic search with an internal search and chat-based interface that retrieves and analyzes relevant internal research or prior memos/diligence, providing organizational and domain-specific context.
As VCs and investors explore the investment landscape, there’s an increasing trend toward relying on quantitative analysis to complement intuition, experience, and evaluation of a firm’s potential. AI technologies are pivotal in this shift, aggregating data from diverse platforms and third-party data marketplaces. These tools organize the information in a way that enables investors to make more informed and data-driven decisions.
However, there are risks associated with this approach. Qualitative factors that are difficult to quantify – such as the background of an incoming key executive, company culture, and market dynamics – might be overlooked. Additionally, the widespread use of AI-driven tools could lead to a homogenization of investment strategies. If many investors rely on similar systems, this could diminish diversity in funding and potentially cause unconventional opportunities to be missed.
Two specific core work products of analysis are financial models and investment memos. LLMs natively integrated with spreadsheet tools can automate the rote aspects of financial modeling – such as comparables analysis, discounted cash flow (DCF) construction, and formatting. LLM-based output may be both quantitative, such as retrieving, populating, and modifying a model; as well as qualitative, such as generating long-form memos based on data. Additionally, VCs use native chat to summarize, troubleshoot, identify key drivers, or derive further insights with both model-specific and organization-wide context to accelerate financial analysis.
Content Creation for Education and Fundraising
It is crucial for VCs to produce thought leadership content – not only to educate their investors and community, but also to support fundraising efforts. This content can range from analyses on the “current state of X” and year-end reviews to blogs, webinars, and strategic social media posts on “why to invest in venture capital.” Staying ahead of investment trends and offering unique insights are essential for investors to stand out from their competitors. AI can significantly streamline this process by assisting in ideation and co-writing.
VCs and startup targets alike use LLMs to generate new graphics based on prompts and existing reference designs to make their marketing materials more professional. Some have even created video content leveraging AI-generated visuals and broadened reach via functions like translation.
Learn More about the AI Fund
We are seeing strong interest in this fund as prior AI Fund vintages were oversubscribed, and we’ve had to establish a waitlist to accommodate interest.
If interested, we recommend securing a spot promptly.
Max Accredited Investor Limit: 249
USING AI AT A LAW FIRM
Legal Research
Similar to the venture fund use cases, AI also can provide valuable tools for legal research, balanced against potential risks. There have been several reports of lawyers using generative AI tools for research purposes, only to find themselves sanctioned by a court or subject to disciplinary proceedings – and/or public ridicule – due to the fact that cases referenced were imagined by the tool.
This scenario highlights an important aspect of how general-purpose generative AI tools operate. They are not a search engine or search database. Rather, based on massive amounts of data training, they supply an answer by predicting the likeliest next word. While sometimes this answer is an objectively correct, other times the tool may hallucinate or produce a false response due to conflating inputs in the training data or making up a response that looks good when no answer exists. For example, our law firm ran an experiment asking a generative AI tool for 10 cases standing for a particular proposition when we knew there was only one. The tool returned the one correct case and nine fabricated cases in order to fulfill the prompt request.
While these risks must be kept in mind and human verification should be employed, there also are great opportunities for AI and enhanced natural language processing to greatly improve legal research. The most popular providers of legal research repositories are now developing or piloting such functionality, including generative AI-powered features for searching, summarizing, drafting, and extracting data. Lawyers reading this article may remember the days of detailed training on how to run searches in these databases. Imagine how much easier it would be to simply request the desired case law the same way you would speak and get short summaries at the same time. This allows lawyers to get more accurate results from their case law research faster than before, saving time and expense. Legal research provides a powerful opportunity for use of generative AI, as long as it is used wisely.
Legal Due Diligence
As an overlap to the venture fund scenarios, there also are an increasing number of tools being marketed to legal professionals designed to make the due diligence process more efficient for mergers, acquisitions, financings, initial public offerings, and similar transactions. This is again an important use case, due to the fact that due diligence can be a significant driver of costs and time in these transactions. AI tools for due diligence can be especially useful, running a mass comparison of documents to identify deviations from a contract template or rapidly identifying and extracting specific provisions, for example.
That said, there remains an important human element to interpreting the nuances of contractual provisions, such as understanding the difference between an inbound and outbound license, or whether the word “exclusive” appears in a material or innocuous manner. For example, a broad exclusive grant to a company’s product or intellectual property (IP) would be a highly material issue to flag in transactional due diligence. A statement that a company retains exclusive ownership of its IP would be consistent with expectations and not noteworthy.
In their current form, generative AI tools focused on legal due diligence are best utilized as an assist — not a replacement — for human due diligence. This distinction also touches on the ethical obligations lawyers have to their clients, such as to perform legal services with competence and reasonable diligence. However, as these tools continue to improve and evolve, they will likely play an increasing role in legal work.
Data Analytics and Synthesis
It is not surprising that various aspects of law involve making sense of large amounts of data, such as eDiscovery (the process of scanning electronic information to obtain nonprivileged information relevant to a case) and document management. AI software has the potential to make the process of sifting through data for relevant information significantly faster and more efficient. For example, in eDiscovery, AI-driven computer assisted review (CAR), conceptual clustering, and predictive coding can reduce the number of documents required for review by lawyers, allowing them to prioritize documents for review. The tools can then present the data visually, such as in pie charts based on conceptual similarity, making it easier to digest. Additionally, AI software can complete some repetitive tasks, such as document redaction, more quickly and consistently, increasing efficiency and reducing costs. In fact, the historic success of using AI tools for eDiscovery is a model to follow for due diligence.
And then, what of law firms using AI to process and interpret data for firm analytics, marketing synthesis, and commercial contract benchmarking? With AI, applications can sift through large volumes of data to determine patterns and trends. For example, while marketing departments have historically used CRM tools, databases, and structured query language (SQL) queries to pull targeted lists for cross-selling campaigns, AI can now greatly reduce or eliminate manual data analysis. AI identifies patterns that categorize target groups by areas of interest, allowing marketing teams to promote personalized solutions, products, or programs to highly targeted audience segments. AI applications could similarly support fee estimates, subject to data availability regarding billable hour rates, hours expended, client demographics, matter descriptions, and all of the variables that play a role in legal costs for a transaction.
We have likewise seen discussion of AI analytics to provide benchmarking around market practice with respect to contractual terms (including standard commercial rates and the frequency with respect to which certain terms or exclusions appear). Such analyses are currently undertaken based on experience and/or manual processes from publicly reported or sanitized data. While leveraging AI instead could be massively powerful, doing so within a law firm will require careful analysis of whether the associated client and other data can be used in this manner without violating client confidences and other obligations to clients.
Content Creation
While humans still lead the legal drafting process, leveraging their expertise, creativity, and contextual vision to compose a written work, generative AI enhances efficiency by accelerating tasks like summarizing legal content and contract analysis. AI provides valuable support by generating initial content, suggesting appropriate legal language, or aiding in formatting text based on the provided structure. Examples include generating law firm website content and marketing materials, writing blog posts and articles, beginning briefs or contract provisions, and creating slides for presentations.
In cases involving routine documents, AI may take on a more significant role in drafting, with lawyer reviewers conducting subsequent revisions. However, risks still exist with AI – for example, plagiarism, copyright infringement, and inaccuracy need to be checked. Overall, human involvement is a must for avoiding these risks and preventing robotic-sounding writing – and adding the personal touch, intuition, editing (for flow, tone, organization, etc.), and creativity into any written work.
As in the venture fund space, AI is a helpful tool for marketing, providing tailored content upon entry of a detailed prompt or query. In a recent study from The American Lawyer, law firms attest to using AI to research clients’ industries and draft marketing materials, but cite concerns with bias, accuracy, and the “black box” of the lack of transparency regarding how the AI produces outputs. Some firms have mitigated these risks by restricting early use of generative AI to “legal-adjacent use cases” – i.e., functions that don’t require the input of client-specific information and instead rely on public data. In the last year, large firms have deployed generative AI to create client alerts about regulatory and statutory updates, research laws relevant to a client’s case, and generate first drafts of marketing and legal materials.
LOOKING AHEAD
As is true across all industries, the introduction of a new technology requires institutional buy-in. Successful AI integration in venture funds and law firms requires identifying and addressing common barriers, such as resistance to technological change, data security concerns, lack of AI literacy, and budget constraints. We expect that over time, given the potential for efficiencies in productivity, both venture firms and law firms will be forced to foster an innovative AI culture, prioritize data security around AI, and invest in AI education to implement this powerful tool.
Learn More about the AI Fund
We are seeing strong interest in this fund as prior AI Fund vintages were oversubscribed, and we’ve had to establish a waitlist to accommodate interest.
If interested, we recommend securing a spot promptly.
Max Accredited Investor Limit: 249
This communication is from Alumni Ventures, a for-profit venture capital company that is not affiliated with or endorsed by any school. It is not personalized advice, and AV only provides advice to its client funds. This communication is neither an offer to sell, nor a solicitation of an offer to purchase, any security. Such offers are made only pursuant to the formal offering documents for the fund(s) concerned, and describe significant risks and other material information that should be carefully considered before investing. For additional information, please see here. Venture capital investing involves substantial risk, including risk of loss of all capital invested. This communication includes forward-looking statements, generally consisting of any statement pertaining to any issue other than historical fact, including without limitation predictions, financial projections, the anticipated results of the execution of any plan or strategy, the expectation or belief of the speaker, or other events or circumstances to exist in the future. Forward-looking statements are not representations of actual fact, depend on certain assumptions that may not be realized, and are not guaranteed to occur. Any forward-looking statements included in this communication speak only as of the date of the communication. AV and its affiliates disclaim any obligation to update, amend, or alter such forward-looking statements, whether due to subsequent events, new information, or otherwise.