Episode 57 of the Legal Tech StartUp Focus Podcast -- An interview with Tanguy Chau, co-founder & CEO of Paxton and Michael Ulin, co-founder and CTO of Paxton
Mar 14, 2024
auto_awesome
Engineer and VC Tanguy Chau and AI expert Michael Ulin discuss Paxton, an award-winning legal tech startup. They focus on developing tailored legal language models for research, drafting, and document analysis. Tips include early product release, customer feedback, and innovation in the legal industry.
Paxton AI focuses on developing industry-specific legal language models for accuracy and security.
Paxton offers legal research, document drafting, and document analysis applications with a focus on data privacy.
Deep dives
Paxton AI's Unique Approach to LegalTech
Paxton AI focuses on developing industry-specific and application-specific legal language models to enhance accuracy, speed, and security. By training on legal-specific data, they aim to provide faster and more accurate responses to legal queries, ensuring no issues such as hallucination problems. This approach sets them apart in the legal tech field and allows for more tailored and reliable solutions for legal professionals and clients.
Paxton AI's Applications and Use Cases
Paxton AI offers three main applications: legal research, document drafting, and document analysis. Their platform indexes millions of legal documents daily to provide access to state and federal laws, court rulings, and more. They leverage their trained models on public documents to facilitate accurate document drafting, focusing on motions, complaints, and other legal documents. Additionally, Paxton offers document analysis for extracting insights and facilitating Q&A based on uploaded case files or internal knowledge sources, prioritizing data privacy and security for users.
Future Roadmap and Advice for LegalTech Startups
Paxton AI's future roadmap includes advanced language models, expanded data source connections, and multi-step workflows for enhanced user experience. The company emphasizes customer-centricity and feedback-driven development to continuously improve their platform. Their advice to legal tech startups includes the importance of engaging with the community, embracing early iterations for rapid feedback incorporation, and prioritizing customer needs to drive innovation and success in the legal tech industry.
Tanguy's and Mike's Respective Backgrounds and Path into Legal Tech:
Tanguy is an engineer with advanced degrees (including an MBA) from MIT and experience in venture capital, notably early-stage investing in legal tech startups (one of such startups being Ironclad)
Mike has a background that includes working at the Federal Reserve, McKinsey, and co-founding a company that applied AI in the insurance space. Mike’s experience with AI and legal/regulatory challenges contributed to starting Paxton.
Paxton's Recent Award: Celebrated being one of the winners at the 2024 ABA Tech Show Startup Alley
Paxton's Technology and Approach:
Focuses on developing industry-specific, application-specific, and firm-specific legal language models for greater accuracy, response speed, and security.
The data for model training includes public domain legal documents, emphasizing legal-specific training over general-purpose models.
Paxton enables customization for firms by allowing connections to internal knowledge sources without training the model on client-specific data unless requested.
Applications of Paxton:
Legal Research: Provides access to laws, regulations, and court rulings across all states and federal levels.
Document Drafting: Uses a vast corpus of legal documents to assist in drafting accurate first drafts of legal documents.
Document Analysis: Offers document analysis and Q&A capabilities for large volumes of documents, ensuring data privacy and governance for firms.
Use Case for Training Young Lawyers: Paxton aids in training younger lawyers by allowing them to ask questions and practice without fear of judgment, enhancing their learning and confidence.
Future Roadmap: Paxton plans to develop more advanced language models, connect to more data sources, and execute multi-step workflows for synthesized answers from various data sources.
Advice for Legal Tech Startups: Tanguy and Mike emphasize the importance of being customer-centric, seeking feedback, and iterating based on user input to improve and refine the product.
Get the Snipd podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share & Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode