Once again, hallucinated citations are not just for lawyers. A recent Minnesota case highlighted the issue of GenAI hallucinations when it come to making up things to fit the narrative of its users. In this situation, a prestigious Stanford University Professor, Jeff Hancock, founding director of the Stanford Social Media Lab submitted an affidavit to the court in support of Minnesota Attorney General Keith Ellison that contained two citations to academic articles that do not exist. To put a cherry on top of this story, the affidavit was about the effects of “deep fake” technology and its influences on elections. Whether this is laziness, or purposeful (my guess is the prior), my prediction is that fake quotes and citations are going to run rampant in 2025, so get your cite checkers out for both legal and scientific citations.
For those in the US, next week is Thanksgiving, so I’ll be taking a break and heading a bit north to windy and cold(er) Oklahoma to see family and friends. I hope many of you are doing the same.
Here are some other stories that caught my attention this week. Remember, you can track these stories with the AI Lawyer Talking Tech podcast (Spotify or Apple) which covers legal tech news and summarizes stories on a daily basis.
Misinformation Expert’s Report Contains Fabricated Citations
A Minnesota-based misinformation expert recently faced scrutiny after submitting a report to the state legislature that included fabricated citations. The expert acknowledged using artificial intelligence tools to draft the document, which resulted in references to non-existent sources. This incident highlights the potential pitfalls of relying on AI for content creation without thorough verification, emphasizing the necessity for human oversight to ensure accuracy and credibility in official communications.
Source: Yahoo! News
Lexis+ AI Faces Scrutiny Over Fabricated Citations
A Canadian law professor sparked debate by critiquing Lexis+ AI’s research capabilities, citing instances of fabricated case citations and mischaracterized summaries during testing. The professor described the AI’s performance as failing, raising concerns about the reliability of generative AI in legal research. LexisNexis defended its tool, emphasizing it is still in development and designed to augment, not replace, traditional research methods. This incident underscores the challenges of integrating AI into legal practice, particularly the need for robust validation processes and human oversight.
Source: Artificial Lawyer
University of Hawai‘i Develops AI Expungement Tool
The University of Hawai‘i launched an AI-driven expungement tool, automating the creation of court documents for clearing criminal records. The tool is expected to increase access to justice by making the process more efficient and cost-effective.
Source: University of Hawai’i News
Aderant Achieves SOC 2 Certification
Aderant, a leading global provider of business software for law firms and other legal organizations, has announced that its Software as a Service (SaaS) platforms have achieved Service Organization Control (SOC) 2 Type 2 certification with zero exceptions. This certification signifies that Aderant’s cloud-based offerings meet rigorous standards for security, availability, processing integrity, confidentiality, and privacy, assuring clients that their sensitive data is protected. The independent audit, conducted by KirkpatrickPrice, covered key platforms including Aderant Expert/Expert Sierra, handshakes by Aderant, iTimekeep, BillBlast, and CompuLaw. This achievement reinforces Aderant’s commitment to data security and underscores the reliability and trustworthiness of its cloud-based solutions for legal professionals.
Source: Legal Technology News
Brightflag Launches GenAI Assistant for Legal Spend Management
Brightflag has introduced the first generative AI assistant specifically for legal spend and matter management. Leveraging a database of over $250 billion in legal invoices and 3.5 million matters, the tool helps corporate legal teams make faster, data-driven decisions by providing insights and predictions in natural language. This innovative assistant aims to optimize budgets, improve efficiency, and enhance predictability for legal departments.
Source: Legal IT Professionals
Trellis Launches Trellis AI
Trellis unveiled Trellis AI, a generative AI tool designed to transform trial court litigation. The platform automates legal tasks such as document review and drafting, aiming to improve efficiency and reduce litigation costs.
Source: Legal Technology News
Hotshot Launches M&A Simulation Platform
A new merger and acquisition (M&A) simulation platform, developed by Hotshot, aims to boost associate training and development in law firms. This interactive tool provides a realistic, firm-specific M&A experience, allowing associates to hone practical skills in due diligence, negotiation, and deal execution. The platform emphasizes experiential learning and personalized feedback, addressing the limitations of traditional training methods. Hotshot believes this immersive approach fosters deeper understanding, improved decision-making, and greater confidence in handling complex transactions, ultimately benefiting both associates and the firms they work for.
Source: Bluefield Daily Telegraph
NetDocuments Showcases ndMAX Assist for Legal AI Innovation
NetDocuments has revealed its AI assistant, ndMAX Assist, which automates legal workflows such as document creation and research. Aimed at making legal AI more proactive, the technology could reshape legal practices, streamlining operations and reducing inefficiencies.
Source: Legal IT Insider
Beyond Billable Hours: Flatiron’s Conrad Everhard on a Fixed-Fee and Tech Based Approach to M&A
Flatirons Development, a Colorado-based real estate law firm, has embraced a fixed-fee, tech-driven approach to mergers and acquisitions, seeking to offer clients greater predictability and efficiency. By leveraging technology like Kira Systems for due diligence and custom-built software for document automation, the firm has streamlined processes and reduced the time spent on routine tasks. This allows attorneys to focus on higher-value work and provide more personalized client service, while the fixed-fee structure promotes transparency and eliminates surprise billing. The firm believes this model sets them apart in the competitive legal landscape and better serves the needs of their clients, particularly in the often complex and drawn-out M&A process.
Source: Technically Legal – A Legal Technology and Innovation Podcast
Theo AI Raises $2.2 Million for Litigation Analytics
Theo AI secured $2.2 million in funding to develop its litigation outcome prediction platform. The tool analyzes case data to forecast probabilities, damages, and timelines, offering strategic insights for lawyers.
Source: Artificial Lawyer
Buenos Aires Courts Adopt ChatGPT for Drafting Rulings
Buenos Aires has begun using ChatGPT to draft rulings on simple cases like payment orders, aiming to reduce judicial backlogs. While AI drafts are reviewed by judges, this move positions the city as a pioneer in integrating AI into the judiciary.
Source: Rest of the World
Tonic Textual Introduces AI for Legal Document De-Identification
Tonic.ai’s Textual tool uses large language models to redact sensitive information while preserving data integrity, ensuring compliance with privacy regulations like GDPR and HIPAA. This innovation simplifies privacy compliance for legal professionals.
Source: Security Boulevard