Important points
- Common AI models are often inappropriate for legal data applications due to the complexity of legal workflows.
- Fine-tuning general models for legal applications is usually not effective, so customized solutions are required.
- Building specific applications on top of AI models is critical to leveraging AI models in legal environments.
- The legal market is rapidly embracing AI technology, changing competitive dynamics.
- Law firms are deploying AI to differentiate their services in a traditionally undifferentiated market.
- The legal sector has historically lacked software solutions, creating opportunities for AI-powered innovation.
- To gain acceptance from tech-savvy lawyers, legal AI products must go beyond the basic model.
- AI software companies are structurally different from traditional software companies because of their evolving model capabilities.
- Rapid advances in AI models can quickly make certain capabilities obsolete.
- Success in the competitive legal tech market requires investment in product and engineering.
- Focusing on product preparation can delay sales and ensure quality and reliability.
- The adoption of AI in law firms is driven by the need to provide better services at competitive prices.
- There is a pent-up demand for AI solutions as the legal sector is underserved in the software sector.
- AI companies need a deep understanding of model functionality to deliver differentiated products.
- The fast-paced nature of AI development impacts product strategy and feature relevance.
Guest introduction
Max Junestrand is CEO and co-founder of Legora, an AI platform that transforms the way lawyers work across 800 clients in more than 50 markets. The 23-year-old with no legal background co-founded the company in Stockholm and grew it from 40 to 400 team members around the world. Legora recently raised a $550 million Series D round at a valuation of $5.55 billion to accelerate its expansion in the United States.
Limitations of common AI models in legal applications
- Generic models are inadequate for legal data applications and customized solutions are required.
-
I think part of the paradigm was that you have to train your own model. Just as no general model is better and fine-tuning becomes very important, there are two reasons why… Fine-tuning doesn’t really seem to be working, at least not at the scale we were operating at.
— Max Junestrand
- Tweaking common models is often ineffective in the legal field.
- Legal workflows are complex and require specific AI applications on top of the model.
-
Too many applications needed to be built on top of the model before it could be used within the environment.
— Max Junestrand
- Customized AI solutions are essential to address data legal challenges.
- Understanding the limitations of common AI models is essential to delivering effective legal tech solutions.
- The need for customized AI applications highlights a unique demand for the legal industry.
Rapid adoption of AI in the legal market
- AI technology is being rapidly adopted in the legal market, surprising many observers.
-
Yes, it felt vivid, but secondly, and perhaps more importantly, the law firm market is a very interesting one. Because it’s like a perfect balance, although frankly the differentiation is quite low.
— Max Junestrand
- Law firms are being encouraged to embrace AI to differentiate their services.
-
If one of them starts leveraging Legora to offer better services at a better price point.
— Max Junestrand
- AI adoption is driven by the need to stand out in a low-differentiation market.
- The competitive landscape for law firms is evolving thanks to AI technology.
- Law firms are leveraging AI to deliver better services at competitive prices.
- The introduction of AI is changing the dynamics of legal service delivery.
Gap in legitimate software solutions
- The demand for AI solutions was created because the legal sector was underserved by software.
-
However, I think the legal field has been under-served by good software for a long time, so there are a lot of problems that can be easily solved using LLMS, but are much harder to solve than they were before LLMS.
— Max Junestrand
- Large-scale language models (LLMs) address long-standing problems in the legal field.
- The historical lack of software solutions in law has created opportunities for AI.
- AI-driven innovation is filling the gap in legal software solutions.
- The advent of the LLM has completely changed the landscape of legal technology.
- Legal professionals are increasingly relying on AI to solve complex problems.
- The under-service of legitimate software highlights the potential for advances in AI.
The need for better legal AI products
- For legal AI products to be accepted, they must outperform the underlying model.
-
If you show up with a legitimate AI product, it has to be better than the underlying model, otherwise they’ll just say, why are you entitled to my money?
— Max Junestrand
- Tech-savvy lawyers are looking for great AI solutions.
- For legal AI products to be adopted, they must demonstrate great value.
- The competitive environment of the legal tech industry requires high-quality AI products.
- Lawyers expect legal AI solutions to offer clear advantages over existing models.
- The need for great products drives innovation in legal AI.
- Legal AI solutions must meet the high expectations of informed users.
Differences in the structure of AI software companies
- AI software companies are structurally different from traditional software companies.
-
One of the unique things about AI software companies is that they are not built tactically like traditional software companies. You need to deeply understand the model’s capabilities and deliver it to your customers in a radically differentiated way.
— Max Junestrand
- The rapid evolution of model capabilities impacts the operations of AI companies.
- AI companies must consider unique structural and strategic considerations.
- For AI software companies, understanding how models work is critical.
- To be successful, AI companies must offer differentiated products.
- The operating dynamics of AI companies are different from traditional companies.
- AI software companies need to adapt to the faster pace of technology evolution.
Impact of rapid advances in AI on product strategy
- As AI models improve, certain features can quickly become less relevant.
-
As the model improves, the feature may no longer be important after 6 months.
— Max Junestrand
- Rapid advances in AI will impact product development strategies.
- The pace of AI development is rapid, impacting feature relevance and strategy.
- Product strategies must adapt to the evolving capabilities of AI models.
- The speed of AI evolution requires agile product development.
- Stakeholders need to understand the implications of rapid advances in AI.
- AI product strategies must be flexible and responsive to changes in technology.
The importance of investing in products and engineering
- Investment in product and engineering is essential to success in a competitive market.
-
I think if you want to be the best, you need to invest in the product, you need to invest in the engineering, and you need to build a culture of reliability first.
— Max Junestrand
- A culture of trust is critical to leadership in the legal tech market.
- Investing in product development is essential to achieving competitive advantage.
- Engineering excellence is a key element of Legal Technology’s success.
- For companies to succeed in legal tech, they must prioritize product readiness.
- Investing in products and engineering drives innovation and market success.
- A focus on quality and reliability is essential to long-term success.
Balancing product preparation and market entry
- Focusing on product preparation can delay sales to ensure quality and reliability.
-
In fact, we went through a six-month period where we basically had no sales because we weren’t ready to hire 1,000 lawyers a day.
— Max Junestrand
- Prioritizing product quality can affect your immediate sales strategy.
- Start-ups face the challenge of balancing product development and market entry.
- Ensuring product readiness is key to successful market entry.
- Delaying sales in favor of quality can lead to long-term success.
- Strategic decisions to prioritize product readiness can impact growth.
- Companies must balance product development with market demand.
