AI in ADR
AI "Workslop" Undermining Productivity | Judge Preliminarily Approves $1.5B Anthropic Settlement | AAA Partners with Integra Ledger
This human-curated, AI-generated newsletter from the AAA-ICDR Institute and AAAiLab keeps you up on AI news over the past week that is relevant to alternative dispute resolution (ADR).
AI in ADR and Legal Services
The AAA Partners with Integra Ledger to Offer Blockchain-Based Document Authentication Service
American Arbitration Association
The American Arbitration Association and Integra Ledger have launched a blockchain-based service that enables users to register and authenticate documents, addressing growing risks of AI-powered forgeries. This tool offers tamper-proof, privacy-preserving verification, allowing individuals and organizations to confirm document integrity throughout a contract’s lifecycle. The service is accessible globally, with free and subscription options, supporting trust and transparency in legal and commercial documentation.
88% of Legal Aid Professionals See AI as Key for Access to Justice, New Everlaw Survey Reveals
Associated Press
Legal aid organizations across the U.S. are rapidly embracing AI tools to address overwhelming demand and resource shortages, with most already using AI for tasks like document review and client intake. Surveyed professionals overwhelmingly believe AI can help close the justice gap, increasing their capacity to serve underserved populations, though concerns about data privacy and ethics persist. AI is seen as a transformative force for expanding access to legal services.
Gen Z Is Getting Divorced, With the Help of AI
Cosmopolitan / AOL
Annabel Iwegbue
Gen Zers facing divorce are increasingly turning to digital tools like ChatGPT and online legal resources to navigate separations without expensive lawyers, especially in straightforward cases. This shift reflects their digital fluency and desire for autonomy, but legal experts caution that while AI can simplify and demystify the process, it cannot replace the nuanced guidance and advocacy of a qualified attorney, particularly in complex or high-stakes situations.
AI’s Rational Empathy Promotes Reconciliation in Conflict: Evidence from Behavioral Experiments, Linguistic Analysis, and Topic Modeling
Research Square
Tongtong Ye, et al.
In a controlled study comparing AI and human counselors in conflict resolution role-play, both were equally effective at reducing participants’ desire for retaliation and increasing understanding, but AI outperformed humans in boosting willingness to communicate. Linguistic analysis showed AI favored practical, solution-focused advice—termed “Rational Empathy”—while humans emphasized emotional affirmation.
An AI Fail By An Elite Litigation Firm
davidlat.substack.com
David Lat
A major law firm admitted to submitting a court brief with multiple erroneous and even fabricated legal citations generated by AI, despite having policies and training for responsible AI use. The supervising partner took full responsibility, highlighting that even elite legal teams are vulnerable to AI-related errors.
LexisNexis exec says it’s ‘a matter of time’ before attorneys lose their licenses over AI errors
Fortune
Nino Paoli
Attorneys are facing mounting penalties as courts uncover a surge of AI-generated errors in legal filings, including fabricated cases and improper citations from open-source language models. These tools also pose risks to client confidentiality, prompting some firms to develop in-house or proprietary AI systems, though these have their own limitations. Experts emphasize the urgent need for legal professionals to be rigorously trained in responsible AI use as the technology’s adoption outpaces understanding.
Stanford, USC, Duke study shows AI reads contracts better than lawyers
Mashable
Monica Chin
In a head-to-head test, AI outperformed experienced lawyers at reviewing NDAs, achieving higher accuracy and vastly greater speed. While this highlights AI’s growing capability in legal document analysis, experts emphasize that such tools are best seen as assistants, freeing lawyers for more complex, human-centered work. Familiarity with legal AI is becoming essential for lawyers, signaling a shift toward more efficient and technology-integrated legal services.
With AI, Junior Lawyers Will Excavate Insights, Not Review Docs
Bloomberg Law
Eric Dodson Greenberg
As AI becomes integral to legal work, the profession faces the dual challenge of leveraging its efficiency without compromising rigorous standards or undermining the training of junior lawyers. Success hinges on disciplined validation of AI outputs and strategic prompt engineering, with mentorship ensuring that technology augments rather than replaces core legal skills. This approach enables legal teams to innovate while safeguarding professional development and judgment.
Generative AI and LLM Developments
AI-Generated “Workslop” Is Destroying Productivity
Harvard Business Review
Kate Niederhoffer, et al.
Despite rapid adoption of generative AI in the workplace and increased reliance on AI-driven processes, most organizations are not experiencing tangible benefits or measurable returns on their investments. This disconnect highlights a gap between the enthusiasm for AI integration and its actual impact on productivity and value creation within companies.
Anthropic AI’s $1.5 Billion Copyright Settlement: A Game-Changer for AI and Content Creators
Opentools
A judge preliminarily approved a landmark $1.5 billion settlement between Anthropic and and a group of authors underscores the escalating financial and legal risks for AI firms using copyrighted materials without permission. The precedent-setting payout, awarding $3,000 per infringed work, signals a new era of accountability and compels AI developers to prioritize legal and ethical data sourcing to avoid costly repercussions and strengthen respect for intellectual property in AI development.
Reddit’s Lawsuit Could Change How Much AI Knows About You
Best Lawyers
Justin Smulison
Reddit’s lawsuit against Anthropic centers on claims that the AI company used Reddit’s user-generated content to train its models without permission or payment, bypassing standard licensing deals. Instead of copyright infringement, Reddit alleges breach of contract, testing whether terms of service can effectively limit AI data scraping. The case’s mediation outcome could set crucial precedents for how user content is accessed, monetized, and ethically protected in the AI era.
The AI-energy apocalypse might be a little overblown
The Verge
Justine Calma
Surging electricity demand from AI-driven data centers is pressuring US power grids, prompting utilities to propose a major expansion of fossil fuel infrastructure. However, much of this projected growth may be speculative or overstated, risking unnecessary costs and increased pollution. Greater transparency, improved forecasting, and stronger commitments to renewable energy from both tech companies and utilities are essential to align AI expansion with climate and affordability goals.
AI Regulation and Policymaking
‘Humanity’s Fate Cannot Be Left to Algorithm,’ Warns Secretary-General in Security Council Debate on Artificial Intelligence
United Nations
The UN Secretary-General emphasized the urgent need for global cooperation to ensure artificial intelligence advances peace and security rather than fueling conflict. He called for strict human oversight of AI in military decisions, robust international regulations, protection against AI-driven misinformation, and efforts to bridge technological divides between nations. He urged member states to support new UN initiatives fostering inclusive, human-centered, and accountable AI governance.
SuperAI—A Supervisory Framework for AI Governance and Alignment
The AI Journal
Blade Kotelly
As AI systems become increasingly complex and influential, their risks—such as bias, manipulation, and unpredictable interactions—demand robust oversight. The proposed “SuperAI” framework introduces a supervisory AI layer that continuously monitors, audits, and intervenes in other AI systems to enforce ethical, legal, and societal alignment. Grounded in open-source transparency and informed by evolving legal codes, SuperAI aims to provide adaptive, accountable governance, ensuring AI serves human interests across diverse cultures and jurisdictions.
Meta’s Llama AI Gets Green Light for U.S. Government Use
Modern Diplomacy / Reuters
U.S. federal agencies now have official approval to use Meta’s Llama AI system, alongside offerings from Amazon, Microsoft, Google, Anthropic, and OpenAI. These tools, vetted for security and legal compliance, are intended to streamline government operations such as contract analysis and technical support. The move reflects a broader government effort to modernize services by adopting advanced commercial AI technologies.
AI News from Other Fields
The New York Times builds custom AI tools for investigative reporting
mediacopilot.substack.com
Christopher Allbritton
The New York Times has integrated AI into its newsroom to supercharge investigative journalism, enabling reporters to analyze vast troves of data—such as hours of video or thousands of documents—that would have been unmanageable manually. Custom-built tools help journalists sift through complex information and identify leads, while staff are trained to use AI cautiously, leveraging its strengths in research and analysis but not for writing articles.
Oxford gives students access to AI platform
BBC News / AOL
Ethan Gudge
The University of Oxford is granting all students and staff access to ChatGPT Edu, a specialized AI tool for education, after a successful year-long pilot. This move, part of a five-year collaboration with OpenAI, aims to enhance learning, research, and innovation while prioritizing privacy and responsible use. Training will accompany the rollout, focusing on ethical and effective application of AI in academic settings.
Financial firms fuel a surge in AI research, adoption
CIO Dive
Matt Ashare
Major global banks are accelerating AI research, shifting focus toward practical applications like agentic AI to drive real-world banking innovations. Leading firms have invested in talent, cloud infrastructure, and cybersecurity to overcome regulatory and technical barriers, rapidly deploying AI tools for coding, customer service, and advisory roles. A handful of banks, notably JPMorgan Chase and Capital One, are setting industry standards by quickly translating research into operational solutions.
AI can now pass the hardest level of the CFA exam in a matter of minutes
CNBC
Leslie Picker
Advanced large language models can now pass the most challenging parts of the CFA exam, a feat that previously required years of human study, demonstrating rapid progress in AI’s ability to handle complex financial reasoning and essay-based assessments. While these models show promise for transforming financial services, human expertise remains important for interpreting context and nuanced client needs.
Air Force Grapples with Limitations of AI
Air & Space Forces Magazine
Shaun Waterman
As the Air Force begins using AI in real operations, leaders are confronting challenges beyond algorithmic performance, including data quality, technical infrastructure, interoperability, and workforce skills. While AI can dramatically speed up decision-making, its outputs often require human oversight due to issues like bias, accuracy, and context awareness. Building reliable systems will require careful system design, improved data management, and ongoing investment in both technology and personnel.
The AAA-ICDR and AAAiLab do not endorse, control, or assume responsibility for the accuracy, availability, or content of any summaries, articles, or external sites or links referenced in this newsletter. Neither the AAA-ICDR nor AAAiLab are compensated or otherwise incentivized by any third parties for including content in this newsletter. All copyrights, trademarks, and other content referenced in the articles or article summaries remain the property of their respective owners. If you believe we have linked to material that infringes your copyright rights, please contact us at institute@adr.org, and we will review and address the matter promptly. This newsletter does not constitute legal advice.
The AAA-ICDR respects your privacy. If you no longer wish to receive these emails, you may unsubscribe at any time by clicking the link below.
We also do not share your personal information without your consent. For more information, please refer to the AAA-ICDR’s full privacy policy.