International law firm Pinsent Masons is to deploy a new genAI automation platform, ‘V7 Go’, created by London-based V7 Labs, which provides AI products and services to clients across multiple sectors, including Sony and Roche.
So, what is V7 Go? Well, the short answer is that it’s a highly structured genAI wrapper that helps users to apply, for example, an LLM’s abilities to handle text queries in a more focused way as part of a customised workflow that uses multiple prompts. It operates with OpenAI, Anthropic and Gemini models.
Orlando Conetta, Head of Research and Development at Pinsent Masons, put it this way: ‘V7 Go is the equivalent of Excel for natural language processing, enabling users to scale complex chains of prompts across large volumes of documents. All through an intuitive and powerful UI.’
Or, as the company explained: ‘Go is more accurate and robust than calling a model provider directly. By breaking down complex tasks into reasoning steps with Index Knowledge, Go enables LLMs to query your data more accurately than an out of the box API call. Combining this with conditional logic, which can route high sensitivity data to a human review, Go builds robustness into your AI powered workflows.’
And perhaps another way of putting it would be to say that although an LLM on its own can do a lot for you, an intermediary layer between the genAI capabilities and your range of legal texts that has chained prompts and rules-based logic steps should help to shape those outputs into exactly the end results you want.
‘V7 Go’s capabilities offer immediate opportunities for increasing efficiency, scale, and quality at Pinsent Masons. This is in line with the firm’s broader R&D strategy to pilot technologies that deliver tangible benefits to its people and clients,’ Pinsent Masons added.
One other aspect that stood out in relation to V7 Labs is their strength in visual analysis and they note that ‘V7 Go is capable of recognising both printed and handwritten text, as well as charts, diagrams and logos, by leveraging advanced OCR technologies.’
And in terms of how one pays for this service, the company explained that it uses something called ‘Go tokens’, which are are standardised units obtained by converting tokens from various model providers into a single measure.
‘More expensive models consume Go tokens at a faster rate than cheaper models. While Go tokens are metered and limited as part of a billing plan, the primary limit is on fields,’ they added.
Conetta concluded: ‘This investment reflects our view that we are witnessing a significant wave of creativity driven by Generative AI. Our R&D function provides us with the flexibility to rapidly examine new technologies and consider companies, like V7, that sit adjacent to the traditional legal tech marketplace. Challenging thinking as to the potential for AI in the legal industry and supporting services across our global offices.’
While Alberto Rizzoli, CEO and Co-Founder of V7, which is based in Soho, added: ‘We believe there is an enormous opportunity in empowering legal professionals to build and use tailored GenAI automation workflows.
‘This collaboration will bring generative AI workflows to a wide number of legal professionals, helping to accelerate better efficiency and productivity, and in doing so, bring greater value to Pinsent Masons’ clients.’
So, there you go. Is this a big deal? Well, the proof will be in the pudding and how much value Pinsents can get out of this collaboration by applying it to a range of workflows and client needs. One can see this more structured approach being especially useful for large data sets, such as for due diligence reviews, for example.
More broadly, it’s a further sign that larger law firms are fully embracing genAI, whether it’s via bringing in one, or several, vendors to help with a range of tasks, or in this case taking a more ‘Build Your Own’ (BYO) approach that allows for a lot of customisation, albeit with additional work for the innovation team to handle as genAI capabilities are fitted to specific tasks. And of course, firms can (and do) take both routes at the same time.
That said, innovation teams have limited resources, so we will likely see some firms focus more on BYO approaches first, and others go into deep POCs with particular vendors as their first major genAI moves, beyond using LLM tech that has already been integrated into tools they use, such as may be the case in legal research systems.