P2P

47 I L T A N E T . O R G F irms that invested in standalone AI tools early may now find themselves faced with a problem: they've added an expensive new platform to their tech stack that can only deliver value if lawyers and legal teams invest time to learn how it works, perform the required steps to use it, and diligently monitor outputs effectively. Indeed, these tools can help perform tedious tasks like data and text summarization, extraction, and generation. Still, getting there is sometimes less intuitive than expected from systems designed to save time. Most Gen AI-powered tools use large language models (LLMs) that train themselves using a source set of documents. While this approach ensures that the system's outputs are based on relevant information you've provided, it also requires moving documents, sometimes thousands, out of one system into another. This is often a manual process that involves exporting and uploading documents, extracting the results, and then moving those findings back into your primary workspace. The time-savings AI tools offer likely balance out these efforts, but it adds another potentially frustrating step to already complex workflows. When you pair this barrier with a lack of transparency about how AI reaches conclusions and lawyers' justified concerns about accuracy, we see a recipe for distrust, frustratingly low adoption rates, and limited ROI. One approach to navigating this challenge is to create AI-powered technology that can be easily integrated into systems lawyers and legal professionals use in daily workflows. From case management to e-discovery, leading legal tech providers are incorporating Gen AI components into their solutions to provide more intuitive, practical, and approachable paths to Gen AI integration. "Invisible AI is not the future, it's the present," Marinela Profi, AI strategy advisor for SAS, told Forbes. Similarly, Oracle's senior vice president of AI and data management service, Greg Pavlik said, "AI is immensely powerful today but will not reach its true potential until it is fully integrated into the applications people use every day." AI's place on the evolution cycle The potential of today's AI and Gen AI-driven technology is vast and evolving quickly. However, as someone with thirty- five years of first-hand experience developing, designing, and implementing technologies we now identify as AI and Gen AI, the patterns this technology appears to follow are like those of other technologies. When investing in any new solution, firms must analyze current processes, understand how technology can add value, and implement it in a way that enhances, instead of disrupts, workflows. Understanding how it can drive efficiency and organization in existing workflows is even more critical. This path isn't unique to AI or Gen AI; every significant advancement in the technology sector follows the same cycle: • Overhyped: This new technology can do everything! Forget about everything else. • Disappointment: This tool cannot solve everything. • Realization: This tool delivers more value when thoughtfully implemented. How and where can it deliver the most value? • Analysis: What are the strengths and weaknesses of the technology? • Adoption: Incorporating the technology into core workflows and processes. More value is drawn from the product in the analysis and adoption phases, so don't let your thinking stay in the past, namely the overhyped phase.

• Cover