Source: Alpha Community
Image Source: Generated by Wujie AI
Generative AI has entered the second stage, with application and commercialization becoming the focus of development. In terms of commercialization, legal AI has progressed rapidly and effectively.
Both EvenUp and Darrow have received at least Series B financing and have performed well in commercialization. Among them, Harvey has stood out by not only collaborating with multiple top law firms but also recently securing a new round of financing of $80 million led by Elad Gil and Kleiner Perkins, with participation from OpenAI Entrepreneurship Fund and Sequoia Capital USA, following its $21 million Series A financing in April 2023, bringing its total financing to over $100 million and a valuation of $7.15 billion.
Founded in 2022, Harvey is one of the first investment projects of the OpenAI Fund. Its AI capabilities are built on OpenAI's GPT-4 large model and it has also collaborated with OpenAI and PwC to develop AI models in vertical fields (tax, law, and human resources).
Harvey can help lawyers analyze and generate contracts, conduct due diligence, and perform legal research, significantly reducing the workload of repetitive text work, allowing them to focus on client relationships and driving targeted strategic work.
The success of legal AI in application and commercialization can be attributed to two reasons:
First, the nature of legal work involves text input and output business, especially a large number of legal documents, which naturally matches the strongest language processing capabilities of large language models.
Second, lawyers' time and energy are very valuable, and legal documents (contracts, patents) also have high value. As long as AI can save lawyers' time and complete high-value legal documents, it will generate direct value for law firms.
If you are interested in the new wave of artificial intelligence, have insights, and entrepreneurial aspirations, please feel free to scan and add "Alpha Assistant" and indicate your "name + position" for in-depth connection with us.
Using AI to Disrupt the $344.1 Billion Legal Market
According to Statista's estimate, the size of the legal services market in the United States in 2023 is $344.1 billion, which is a large market. Solving the efficiency problem of lawyers will further expand the size of this market (lawyers will have more time to spend on cases).
Harvey was co-founded by Gabriel Pereyra (CEO) and Winston Weinberg, both of whom have deep technical and legal backgrounds. Before starting the business, Gabriel Pereyra served as an AI research scientist at GoogleBrain, DeepMind, and Meta, while Winston Weinberg worked at the renowned law firm O'Melveny & Myers LLP, focusing primarily on antitrust litigation.
This is a typical combination of technical experts and industry experts. Gabriel Pereyra's deep understanding of AI has given Harvey better AI capabilities compared to companies founded solely by industry entrepreneurs, enabling them to customize large models with OpenAI.
Winston Weinberg's accumulation in the legal industry also allows Harvey to develop AI products that better meet the needs of legal professionals.
This is also reflected in Harvey's revenue and customer acceptance. Harvey recently announced that its revenue has grown more than tenfold since April 2023. They have not only established deep cooperation with major law firms such as Allen & Overy and accounting giant PwC, but also have potential clients in over ten thousand law firms. In addition, their clients include not only law firms but also corporate legal teams and private equity firms.
For example, Allen & Overy uses Harvey in collaboration with OpenAI to provide legal AI capabilities for its 3,500 lawyers and 43 offices worldwide.
Wim Dejonghe, Senior Partner at Allen & Overy, said, "This marks a new era for A&O and the legal industry. Harvey AI is a game-changing product that will enable us to provide unprecedented value, efficiency, and innovation to our clients."
Harvey enables PwC's professionals in over 100 countries to access leading generative AI technology. This will enhance PwC's ability to provide human-guided and technologically supported legal solutions in areas such as contract analysis, regulatory compliance, claims management, due diligence, and broader legal consulting and advisory services. However, all AI outputs will be supervised and reviewed by PwC professionals.
Gabriel Pereyra, Co-founder and CEO of Harvey, said, "Collaborating with PwC has provided us with a unique opportunity to further enhance the potential of AI to solve the most complex legal problems. PwC's capabilities enable us to develop AI systems that extend beyond Harvey's core legal use cases, providing more comprehensive solutions for PwC and its clients."
"Harvey's business is about empowering people. First lawyers, then professional services, and ultimately all knowledge work. Sequoia Capital is excited to work with Gabe, Winston, and the entire Harvey team. We look forward to helping them unleash the power of large language models (LLMs) in their work and usher in a new era of superhuman productivity," said the U.S.-based Sequoia Capital, which has invested in Harvey for two consecutive rounds.
Building Legal AI Chatbot Based on GPT-4 Fine-Tuned Large Models
Harvey's AI capabilities come from OpenAI's GPT-4, but based on their team's professional capabilities, they have fine-tuned the base model with a large amount of legal professional data and built an AI chatbot on the fine-tuned model.
The capabilities of this chatbot include:
Legal writing, helping legal professionals write lengthy, formatted legal documents, including contracts, client memoranda, and more.
Answering complex legal questions based on fine-tuned data, such as "Has a clause in a lease contract violated the law of a certain state? Please rewrite it to make it legal."
Understanding and processing contracts and documents to reduce the repetitive work of legal professionals.
In addition, for large clients like PwC, Harvey not only provides ready-made AI products but also collaborates with clients to build vertical AI models, which are not limited to legal matters and can also include tax and human resources.
Based on Harvey's AI capabilities, industry background, and customer resources, it also has the potential to expand its business to industries such as accounting, finance, and consulting in the future.
In terms of security, as a legal AI company, Harvey has placed sufficient emphasis on security. They have a dedicated security advisory team, with team members coming from top companies such as JP Morgan, Deutsche Bank, Goldman Sachs, and Microsoft.
In terms of compliance, their products also comply with industry standards such as SOC2, ISO27001, CCPA, GDPR, and others.
Why has the commercialization of legal AI been so smooth?
Legal software is not a new phenomenon. There have been companies specializing in case management platforms, document resource management platforms, electronic signatures, and others, with representative companies including DocuSign, LegalZoom, and Ironclad.
The new generation of legal AI companies, such as Harvey, EvenUp, and Darrow, have made several improvements compared to traditional legal software companies.
First, legal AI can significantly improve the work experience for legal professionals by taking over time-consuming, tedious, and low-value document work, allowing lawyers to focus on document review and correction.
Second, it can increase the efficiency of legal workflows by 10 times, 100 times, rather than the traditional software's 1.5 times.
Third, the improved efficiency and time freed up for legal professionals can help them earn more income, handle higher case loads, and expedite case processing for their clients.
The logic behind the successful commercialization of legal AI is as follows:
First, the nature of legal work involves text input and output business, especially a large number of legal documents, which naturally matches the strongest language processing capabilities of large language models.
Second, lawyers' time and energy are very valuable, and legal documents (contracts, patents) also have high value. As long as AI can save time for legal professionals and complete high-value legal documents, it will generate direct value for law firms.
Legal AI currently includes general AI assistants such as Harvey, AI tools focused on specific niche areas such as EvenUp, and tools to help law firms find suitable cases such as Darrow.
The capabilities of Harvey have been previously introduced and will not be reiterated. EvenUp focuses on personal injury claims and can generate a complete claim letter in minutes instead of 20 hours. This not only helps clients obtain higher compensation but also saves time for law firms.
Darrow has built an automated case discovery system using AI. Its AI sifts through large amounts of data, including public records, administrative documents, social media, SEC cases, and other information sources, to discover serious legal violations and unrecognized high-value cases, matching them with public interest law firms. Currently, Darrow has discovered over $10 billion in litigation work for more than 50 leading law firms.
Although the legal service industry market in China is smaller than that of the United States (according to data from the All China Lawyers Association, the market size of China's legal service industry reached 117.8 billion RMB in 2021), the logic of value creation for legal AI can be extended to other industries such as auditing, consulting, as long as the individual value of practitioners in the industry is high and their work aligns with the capabilities of large models.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
