How AI is Changing Legal Services

In the vanguard of legal innovation, generative AI (GAI) is propelling the legal profession into a new era of efficiency and expanded capabilities. A recent survey reveals a significant embrace of GAI within the Am Law 200 firms, showcasing a shift from skepticism to adoption with compelling evidence of enhanced productivity, quality, and speed in legal tasks when leveraging tools like GPT-4. The impact of AI in law extends beyond mere productivity enhancements: -Evolving Lawyer Roles: As tasks become automated, lawyers are transitioning from doing to directing, focusing on integrating AI into workflows, managing automated processes, and ensuring precise engagement of resources. -Enhanced Predictive Analytics: Improved data quality from automation leads to more accurate case analysis and financial forecasting, enabling firms to optimize case selection and fee structures. -Access to Information Revolutionized: Natural Language Interfaces (NLIs) like Harvey offer direct, query-based access to legal information, simplifying research and compliance checks. -Emergence of New Legal Services: Nearly half of law firm leaders are exploring new billable services powered by GAI, including real-time regulatory compliance and data privacy services. As we stand at the precipice of this transformative period, it's clear that AI's role in legal practice is not just a fleeting trend but a cornerstone of future legal services. Transforming the Business of Law™ Let's explore how to future-proof your practice and offer innovative solutions to your clients. #LegalTech #AIInLaw #FutureOfLaw #LegalInnovation #GenerativeAI

I'm an applied scientist, I'm in the business of advancing humanity's knowledge of how to apply technology to advance society. Relativity's Applied Science team is heads-down on the next set of advances for legal technology, but I'm going to pause and reflect on some of the growing evidence demonstrating the validity of our approach and how AI is transforming the legal technology field. First, there's the excellent Vals Legal AI Report, benchmarking the ability of AI to perform various legal tasks, meeting and exceeding the ability of attorneys to do the same work: https://www.vals.ai/vlair. In this study, multiple AI tools were pitted against experienced lawyers across seven common tasks (from document Q&A to contract redlining). The outcome? In four of the tasks, at least one AI tool outscored the human lawyers, and in a fifth task the top AI equaled human performance . The AI excelled particularly at more formulaic or data-intensive tasks like document analysis and extraction, while humans retained an edge in a couple of more complex, reasoning-intensive scenarios. Additionally, in "AI-Powered Lawyering: AI Reasoning Models, Retrieval Augmented Generation, and the Future of Legal Practice", researchers conducted the first randomized controlled trial assessing these technologies, assigning upper-level law students to complete six legal tasks using a RAG-powered legal AI tool (Vincent AI), an AI reasoning model (OpenAI’s o1-preview), or no AI. They found that both AI tools significantly enhanced legal work quality, a marked contrast with previous research examining older large language models like GPT-4. Moreover, they found that these models maintained the efficiency benefits associated with use of older AI technologies. Their findings show that AI assistance significantly boosts productivity in five out of six tested legal tasks, with Vincent yielding statistically significant gains of approximately 38% to 115% and o1-preview increasing productivity by 34% to 140%, with particularly strong effects in complex tasks like drafting persuasive letters and analyzing complaints. What am I most excited about? Benjamin Sexton has published a recent collection of anecdotes from across the industry. The anecdotes included point estimates for precision and recall for 16 uses of generative AI to identify relevant documents for production. These users were able to build AI classifiers that achieved on average 80% precision and 95% recall, saving cost, effort, and time for their customers. Legal professionals should no longer be wondering, "Is generative AI ready for legal?" The evidence continues to grow every day well beyond doubt – the real question now is, "How can we integrate it well into our practice?" For those preparing to embrace this transformation, the path forward is clear: identify pilot projects, select trusted AI platforms, and start refining workflows to capitalize on these efficiency gains. #Legaltech #GenerativeAI #eDiscovery

🚨 New Research Alert! 🚨 How will AI transform the legal profession? Our latest study provides the first randomized controlled trial (RCT) of AI reasoning models & retrieval-augmented generation (RAG) in legal work—and the findings are striking. 🔗 Read the full paper here: https://lnkd.in/gKK44W7B Key Takeaways: 📌 AI dramatically improved legal work quality—challenging past research on GPT-4’s limitations. 📌 We tested two cutting-edge AI innovations: ✅ Vincent AI (a RAG-powered legal AI) ✅ OpenAI’s o1-preview (a reasoning model) 📌 Using 127 law students from @UMNLawSchool & @UMichLaw, we found AI boosted performance in 5 out of 6 legal tasks, with productivity gains of up to 140%! 📈🔥 The Trade-offs: ⚖️ Vincent AI reduced hallucinations and improved factual accuracy. ⚖️ o1-preview enhanced analytical depth but sometimes hallucinated. 💡 Together, these technologies could complement each other for even greater gains. The Bigger Picture: 🚀 AI is no longer just about speed—it’s enhancing legal work quality. 📜 Next-gen legal AI will likely combine reasoning models & RAG for even bigger leaps. ⚖️ The legal profession is on the brink of major AI-driven transformation. Our research suggests litigation-related tasks benefit most from AI, raising important questions about how AI will reshape different legal practice areas. Want to dive deeper? Read the full paper and explore what this means for the future of lawyering! #LegalAI #LawTech #ArtificialIntelligence #FutureOfLaw Of special interest to all those working on legal tech and AI, including Mark Williams, Olga V. Mack, Tim Sullivan, Aamar Hussain, Legal AI, Michael Ulin, Michał Jackowski Prof., 杰克夫斯基, Jamie, Brian W Tang, Kevin Frazier, Ibrahim M. Saleh, Colin Levy, Heather Stevenson, I. Glenn Cohen, Daniel I. Levy, Luiza Jarovsky, Colin Levy, The Law of Tech, Nick Coleman, Olusola Babatunde Adegbite, Paul Jurcys, PhD, Shreya Vajpei, Ülkü Özkürkçü, Michael Ulin, Warris B., Daniel W. Linna Jr., Tiffany Li, Daniel Schwarz, Marcela Escobar-Alava, Rowena Rix, Cecy Graf, Damien Riehl, Marta Ostrowska, Erin Conway Johnsen, Brittany Johnson,, Aditya Sivakumar, Ilkin Gurbanov, LL.M., Emile Loza de Siles, Kyle Bahr, Christina Ayiotis, Esq., CRM, CIPP/E, AIGP

𝗪𝗵𝘆 𝗱𝗼𝗲𝘀 𝗲𝘃𝗲𝗿𝘆𝗼𝗻𝗲 𝗮𝘀𝘀𝘂𝗺𝗲 𝘁𝗵𝗮𝘁 𝗔𝗜 𝘄𝗶𝗹𝗹 𝗿𝗲𝗱𝘂𝗰𝗲 𝘁𝗵𝗲 𝗱𝗲𝗺𝗮𝗻𝗱 𝗳𝗼𝗿 𝗹𝗲𝗴𝗮𝗹 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀? I've been seeing a flood of posts confidently predicting that AI will drastically reduce or eliminate the need for lawyers by enabling clients to handle their own legal work. While it's true that AI may automate certain legal tasks, this view completely ignores basic economics – specifically, the demand side of the equation. Consider the 𝗝𝗲𝘃𝗼𝗻𝘀 𝗘𝗳𝗳𝗲𝗰𝘁 (or Jevons Paradox): 𝗪𝗵𝗲𝗻 𝘁𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝘆 𝗺𝗮𝗸𝗲𝘀 𝗮 𝗿𝗲𝘀𝗼𝘂𝗿𝗰𝗲 𝗺𝗼𝗿𝗲 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝘁𝗼 𝘂𝘀𝗲, 𝘄𝗲 𝗼𝗳𝘁𝗲𝗻 𝗲𝗻𝗱 𝘂𝗽 𝘂𝘀𝗶𝗻𝗴 𝗺𝗼𝗿𝗲 𝗼𝗳 𝘁𝗵𝗮𝘁 𝗿𝗲𝘀𝗼𝘂𝗿𝗰𝗲, 𝗻𝗼𝘁 𝗹𝗲𝘀𝘀. William Stanley Jevons first observed this in 1865 when more efficient steam engines led to increased coal consumption rather than decreased. This effect appears throughout the technology world: • When email made communication faster and cheaper, we didn't communicate less – we communicated far more. • When word processing made writing easier, we didn't write less – we produced more written content than ever before. • When smartphones made computing more accessible, we didn't compute less – we dramatically increased our computing usage. 𝗦𝗼 𝘄𝗵𝘆 𝘄𝗼𝘂𝗹𝗱 𝗹𝗲𝗴𝗮𝗹 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀 𝗯𝗲 𝗮𝗻𝘆 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁? Consider: • When AI makes certain legal services more affordable and accessible, more individuals and businesses will likely seek legal help for matters they previously handled informally. • As legal tasks become easier to perform, companies may pursue legal strategies that were previously too costly or complex to consider. • AI may enable entirely new legal services that weren't previously feasible, creating new markets and demand. In other words, 𝘁𝗵𝗲 𝗮𝘀𝘀𝘂𝗺𝗽𝘁𝗶𝗼𝗻 𝘁𝗵𝗮𝘁 𝗔𝗜 𝘄𝗶𝗹𝗹 𝗿𝗲𝗱𝘂𝗰𝗲 𝘁𝗵𝗲 𝗻𝗲𝗲𝗱 𝗳𝗼𝗿 𝗹𝗮𝘄𝘆𝗲𝗿𝘀 𝗿𝗲𝘀𝘁𝘀 𝗼𝗻 𝗮 𝗳𝘂𝗻𝗱𝗮𝗺𝗲𝗻𝘁𝗮𝗹𝗹𝘆 𝗳𝗹𝗮𝘄𝗲𝗱 𝗽𝗿𝗲𝗺𝗶𝘀𝗲: that demand for legal services is fixed and inelastic. The Jevons Paradox demonstrates exactly the opposite – when services become more efficient and accessible, demand often expands dramatically. AI won't just automate existing legal work; it will likely create entirely new markets for legal services by making them more affordable and accessible to previously underserved segments of society and business. Those who focus solely on AI's ability to automate existing legal tasks are missing half the story. The real transformation may come not from reducing demand for legal services, but from dramatically expanding access to and creating new markets for legal expertise. #AI #legaltech #futureoflaw

👀 Fascinating insights about legal AI from my friend Ate A Pi on Twitter, who spoke with an big firm partner about AI. See her tweet below. I agree with *most* of it. My take - the future is AI-enabled super-lawyers, and putting the AI in-house. It'll be here FAST. More on what I've seen doing legal AI, especially since founding GC AI six months ago this week! 🏡 In-house is the best place for legal AI. We are relatively cheaper and don't bill by the hour. We are more operational. Oh and psst - we hire and fire firms. CEOs think of their lawyer as the GC and their team. Now, there are some outside counsel with deep relationships over time. But the days of the outside guy or gal every CEO wants are ... not the norm anymore. Law firms have to do RFPs like any other vendor. 📈 In-house lawyers have grown 7.5x times the rate of other kinds of lawyers the last 25 years. The role of "product counsel" boomed, just like the role of product manager in this time. Today, Google employs 828 "product counsel" per a search I just did here on LinkedIn. That's more than only the biggest law firms.  AI accelerates that trend. Just like lawyers with more context about your business are more useful, so too is AI. Companies will want to own that and have their best judgment lawyers on it.  So what about the tech? Who's winning? 🏆 OpenAI's GPT-4 is the only model good enough for legal work, and it's not even close. I say this having taught 1000s of lawyers how to use AI. But ChatGPT and vanilla tools have been instructed to be too friendly and tuned to be chatty and golden retriever-esque.  🤡 Gemini is nerfed. It refuses most legal questions.  ❌ Claude's tone is weird, and it won't search the web. 💄 Meta's Llama misses the point entirely, often. It thinks Revlon duties from the famous corporate law case - are to put on make-up.  🍫 PS the bar exam is legit hard! It tests grinding, speed, and memorization. It's a good proxy for many legal tasks. GPT-4 earns my respect for passing it.  But ... I disagree the legal profession is decimated. The future I see is: 🦸🏻♀️ Empowered super-lawyers, like the 100X AI-enabled software engineers Amjad Masad talks about and we see with AI for software. 🤝 Insider and niche knowledge will keep rates for certain expertise high. Sometimes you want the lawyer who's been before this judge or worked for the relevant regulator. You want to read the room, to weigh factors, to know what other of their clients are doing. AI can't do that yet. 🏀 Legal AI will raise the bar for the entire profession. Like how basketball play got better when everyone started lifting weights. You have to use AI if your opponent does. ⚖️ People who could never afford lawyers or just didn't hire them - will get basic but good legal advice. Think - represented divorces, lawyers on smaller deals, startups in regulated industries. AI will lawyer the unlawyered. Thoughts? #LegalAI

Generative AI is changing how litigators approach trials by providing better insights and more efficient workflows. With these tools, lawyers can make faster, more informed decisions and strategize more effectively for their clients' cases. 📝 💡 Here’s what’s making a difference: • 🔍 AI-Driven Insights: These tools don’t just organize data—they reveal trends, predict outcomes, and suggest strategies based on historical rulings. • 📖 More Efficient Prep: Automated analysis reduces time spent on research, letting legal teams focus on higher-value work. • 🎯 Smarter Decisions: Insights into judges’ rulings and opposing counsel tactics help lawyers prepare more effectively. • 🛡️ Compliance Matters: Vet providers thoroughly to ensure they meet privacy and ethical standards—especially when AI models use client data. If your firm isn’t using AI-powered litigation analytics tools yet, it’s worth exploring options that can increase efficiency and improve courtroom outcomes. Tools to consider include Trellis Law, LegalMation, Premonition Analytics, and Pre/Dicta. I recently wrote about litigation analytics for the ABA Journal—check out the link to the full article in the comments! 🔗 #legaltech #AI #litigationanalytics #litigation

𝐀𝐈 𝐢𝐧 𝐋𝐞𝐠𝐚𝐥 𝐒𝐞𝐫𝐯𝐢𝐜𝐞𝐬: 𝐖𝐡𝐞𝐫𝐞 𝐀𝐫𝐞 𝐭𝐡𝐞 𝐂𝐨𝐬𝐭 𝐒𝐚𝐯𝐢𝐧𝐠𝐬? #GenerativeAI is gaining traction in big law firms, yet corporate clients aren’t seeing the expected drop in legal bills. In fact, rates continue to rise. Why? While AI is being used for research, review, and due diligence, law firms haven’t fundamentally transformed how they deliver legal services. Efficiency gains remain difficult to measure, and tech investments can sometimes increase costs rather than reduce them. Clients like British Telecommunications are pushing for transparency wanting to know exactly where AI is applied and how it drives cost reductions. Yet many firms struggle to demonstrate clear savings, leaving in-house teams caught between the push to innovate and the reality of rising fees. This tension signals a pivotal moment: ↳ Clients must take the lead in demanding measurable ROI and clarity on AI’s impact on billing. ↳ Law firms need to move beyond mere adoption of technology to reimagine workflows, fee structures, and collaboration models. ↳ Alternative fee arrangements and AI-powered subscription services could challenge the entrenched billable hour model. The legal industry stands at a crossroads where client expectations, technological potential, and legacy practices collide. For legal leaders and corporate counsel, the path forward is clear: demand transparency, align on strategic goals, and leverage AI intentionally. 👉If your organization works with law firms or manages legal budgets, understanding this dynamic is essential for shaping the future of legal services. 👉 Driving AI adoption in legal or professional services? Now’s the time to demand real impact and transparency. Let’s connect! https://lnkd.in/g37-HCRX #JeffEyet #TheBerkeleyInnovationGroup #LegalTech #AI #GenerativeAI #LawFirms #CorporateCounsel #Innovation #CostSavings #LegalOperations #DigitalTransformation

AI is quietly slashing legal bills...and not just for the big guys. At the companies my family owns, we’ve seen a ~30% reduction in legal costs over the past year to 18 months, and we’re not alone. I’ve talked to several other small and medium-sized business owners who’ve seen similar savings, mostly by shifting routine legal work in-house with help from AI. We’re not talking about replacing lawyers altogether, but rather using AI to handle the repetitive, expensive stuff, first draft communications, contract analysis, compliance checks etc. The kind of work that legal teams have traditionally billed by the hour (often billed at lawyer's rates and done by paralegals)… but that LLMs can now do in seconds. And the numbers are staggering: -A recent study (link in comments) found that AI could complete invoice review 99.97% cheaper than a human reviewer, with higher accuracy. -The same holds true for contract reviews: faster, more consistent, and dramatically less expensive. These aren’t just theoretical gains. They’re happening now, and businesses that embrace these tools are freeing up both time and capital. I’m curious, if you run or advise a business, are you seeing similar savings? Or are you still on the fence about using AI in legal workflows? Let’s talk real-world results. #AI #legalspending #paralegal Disclaimer: This written content was entirely human-generated. The image was AI-generated.

🚨 “Why Legal Teams Are Pumping the Brakes on AI Adoption – And What Consultants Can Do About It" 🚨 As a consultant working at the intersection of tech and law, I’ve seen firsthand the glaring gap between the promise of AI solutions (including generative AI) and the cautious reality of in-house legal teams. While AI could revolutionize contract review, compliance, and risk management, many legal departments remain skeptical—and their hesitations are far from irrational. Here’s what’s holding them back: 1. "We Can’t Afford a Hallucination Lawsuit" Legal teams live in a world where accuracy is non-negotiable. One AI-generated error (like the fake citations in the Mata v. Avianca case) could mean sanctions, reputational ruin, or regulatory blowback. Until AI tools consistently deliver flawless outputs, “trust but verify” will remain their mantra. 2. "Our Data Isn’t Just Sensitive – It’s Existential" Confidentiality is the lifeblood of legal work. The fear of leaks (remember Samsung’s ChatGPT code breach?) or adversarial hacks makes teams wary of inputting case strategies or client data into AI systems—even “secure” ones. 3. "Bias + Autonomy = Liability Nightmares" Legal ethics demand fairness, but AI’s hidden biases (e.g., flawed sentencing algorithms) and the “black box” nature of agentic AI clash with transparency requirements. As one GC mentioned recently: “How do I explain to a judge that an AI I can’t audit made the call?” 4. "Regulators Are Watching… and We’re in the Crosshairs" With the EU AI Act classifying legal AI as high-risk and global frameworks evolving daily, legal teams fear adopting tools that could become non-compliant overnight. Bridging the Trust Gap: A Consultant’s Playbook To move the needle, consultants must: ✅ Start small: Pilot AI on low-stakes tasks (NDA drafting, doc review) to prove reliability without existential risk. ✅ Demystify the tech: Offer bias audits, explainability frameworks, and clear liability protocols. ✅ Partner, don’t push: Co-design solutions with legal teams—they know their pain points better than anyone. The future isn’t about replacing lawyers with bots; it’s about augmenting human expertise with AI precision. But until we address these fears head-on, adoption will lag behind potential. Thoughts? How are you navigating the AI-legal trust gap?👇 #LegalTech #AIEthics #FutureOfLaw #LegalInnovation #cmclegalstrategies

If you're not using Generative AI--or at least experimenting with it--you're behind. On today's and next week's episodes of The Grace Period, I give a very, very high-level overview of what GenAI is and how I use it (almost daily) in my practice. Generative AI is transforming legal practice by creating efficiencies that allow attorneys to focus on strategic work and client relationships. This tech functions more like an assistant than a search engine, helping with drafting, summarizing, and creating content based on prompts. In today's episode, I touch on: • Generative AI tools including ChatGPT, Grok, Claude, and lawyer-specific options like CoCounsel and Harvey • How AI can help draft discovery requests, summarize depositions, and create project plans in significantly less time • Claims about AI replacing associates are overstated—it complements rather than replaces legal judgment • Ethical obligations include maintaining confidentiality by avoiding open-universe AI for client information • AI can "hallucinate" information, so all output requires verification, especially legal citations • ABA rules require technological competence, making AI familiarity a requirement • Enterprise AI solutions like Husch Blackwell's "Prompt Composer" maintain attorney-client privilege in a closed system Next week, I'll share practical examples of how I use generative AI in my daily practice, including what works, what doesn't, and how to get started safely. 🤖♥️🔥✌🏻