Aug 21, 2024 | AI

I’m excited to welcome the Safe Sign Technologies team and technology to Thomson Reuters! The acquisition of Safe Sign Technologies, a UK legal large language model (LLM) startup with a team of world-class AI experts and early-stage language models, could not be a better fit.

Below are five reasons this acquisition is a boon for our customers across all professions as we scale this technology across the Thomson Reuters portfolio.

1. LLMs are helping legal professionals conduct research more quickly and effectively, and Safe Sign Technologies brings expertise in training and developing specialized LLMs. Safe Sign Technologies has developed two legal specific LLMs. Based on our internal assessment, we believe Safe Sign’s models have demonstrated industry-leading performance across a number of domain-specific evaluations. By combining their early-stage technology and their expertise with our existing internal efforts, we hope to significantly accelerate our continued evolution as a content-enabled technology company.
2. We expect Safe Sign Technologies will help Thomson Reuters realize our CoCounsel vision even faster. CoCounsel is founded upon our ability to combine our data, expertise, and trusted content with cutting-edge technology. We expect this acquisition to help accelerate our ability to provide our customers with a professional grade AI experience through CoCounsel  – our genAI assistant – that enables professionals across industries to accelerate and streamline their workflows.
3. Alexander Kardos-Nyheim, Dr. Jonathan Schwarz and their team are strong additions to Thomson Reuters. Co-Founders Alexander Kardos-Nyheim and Dr. Jonathan Schwarz built the Safe Sign Technologies team with top law and AI professors and researchers from Cambridge, Oxford, MIT and DeepMind. They have a world-class team comprised of tenured professors, AI scholars, mathematicians, coders and influential business leaders. Their mix of domain and technical expertise brings a highly unique perspective to the development of technology, which perfectly aligns with our content and technology strategy.
4. This is good news for Thomson Reuters customers. Specialized language models have the potential to significantly increase the speed and accuracy of genAI applications. The accuracy of AI models, including LLMs, is dependent on the quality of content that they are trained on. We believe bringing together Safe Sign’s expertise and approaches to training LLMs with Thomson Reuters trusted content and insights will help us deliver greater accuracy, quality and performance from our AI solutions that will continue to drive benefits for our customers.
5. This deepens our relationships with the academic and AI research communities. Safe Sign developed their AI system using cutting-edge and nascent research paradigms enabled by their strong academic connections. Safe Sign focuses on deep-learning research with a particular emphasis on data-centric methods. As a result of this acquisition, we will be furthering our investments in the integration and development of these early-stage specialized language models, deepening our relationships with the academic and AI research communities, while continuing our ongoing investments in building AI skills across our product portfolio.

Read the press release for more about this exciting acquisition and please join me in welcoming the Safe Sign Technologies team to Thomson Reuters.