One part of US #litigation practice that I think an #LLM / #AI could actually be useful for (but only if used to suggest/enhance a human drafter, *not *through a god-like chat interface): discovery requests, responses, and especially objections (plus tracking/managing the back & forth of these). There is so much tedium in discovery, and so many details to keep track of & keep consistent, that enhancing a human mind would a brilliant use case, I think.
The more I dig into so-called #AI (esp for #law), the more my irritation & frustration grows that any expert on this stuff could legitimately confuse an #LLM with biological intelligence.
Sure, they do general blathering great, but for tasks requiring any level of understanding, esp with complex context, they just fall apart & can't be trusted, even with nifty correctives like embeddings.
(I'm still also convinced they will be really useful assistive tools, with limits.)
Newcomer here.👋 I'm Mark Heftler, NJ/NY licensed attorney, based in Northern New Jersey, serving as Arbitrator Relations Counsel at Forthright. My role centers around managing and guiding arbitrators in dispute resolution, with a particular interest in how legaltech innovations can enhance our field.
I'm here to connect with peers who share a passion for the evolving landscape of law and technology. Looking forward to exchanging thoughts and insights. #Introduction#Arbitration#legaltech
JOB ALERT / HIRING : My company, ContractPodAi, is currently seeking folks who are excited about legal technology and the future of law for several roles!
Cannon has a proprietary way of signing photos to make it proof that the file has not been tampered after taking photo. I am wondering if there is an #OpenSource way to do this on like an android phone. Take a photo and have it be cryptographically guaranteed against phone hardware that it has not been tampered with after taking the photo. #cryptography#photography#signatures#foss#android#legal#legaltech
Appears to be primarily aimed at the 'fictional sources' scenario
"...Court users are expected to have verified that the cited materials exist and stand for the stated propositions."
"Where any Generative AI tool has been used to generate content that is used in Court documents, this should be disclosed within the first five paragraphs of the Court document."
"I predict that in less than five years AI systems will be capable of drafting high-quality first drafts of complex bills and regulations... I also asked ChatGPT to produce its draft legislation and regulations as executable code, that is, in the form of computer programs, using the programming language, Prolog. The results were remarkable. Again, I expect this to be a trustworthy process in a handful of years."
👀 "All attorneys appearing before the Court must file on the docket a certificate attesting either that no portion of the filing was drafted by generative artificial intelligence (such as #ChatGPT, Harvey.AI, or Google #Bard) or that any language drafted by generative artificial intelligence was checked for accuracy, using print reporters or traditional legal databases, by a human being."
I'm not sure what I think about this, but here we are.
TL;DR: Explore Legalpioneer Copilot, a new resource with 600 law-related GPTs covering legal, tax, and regulatory issues. Find and access GPTs created by individuals and companies for legal information and support.
"When I was asked to beta test its AI research bot, I informed a major legal research provider that it worse than sucked. It was dangerous. Not only did it hallucinate... but it conflated almost all the critical distinctions that make law work. It failed to distinguish between jurisdictions, both states and state and federal, as well as majority, concurrences and dissents. To AI, it was all the same, words about law..."
In this thoughtful piece, @qsteenhuis explores the use of "AI" within law school clinics, discussing the benefits, risks, and safer use cases for Gen AI.
It's a thorough examination of how such tools can enhance student learning, improve client services, and server the broader goals of the legal profession. It's also a nice corrective to those folks who reflexively seek to exclude such tech.¹
TL:DR: Curious about which language models perform best on legal reasoning tasks? The latest evaluation reveals that Open AI's GPT-4 takes the lead, followed closely by Google's Gemini Pro.