OpenAI has abandoned its controversial restructuring plan. In a dramatic reversal, the company said Monday it would no longer try to separate control of its for-profit arm from the non-profit board that currently oversees operations. "We made the decision for the nonprofit to retain control of OpenAI after hearing from civic leaders and engaging in constructive dialogue with the offices of the Attorney General of Delaware and the Attorney General of California," said Bret Taylor, the chairman of OpenAI. OpenAI had originally argued its existing structure would not allow its nonprofit to "easily do more than control the for-profit." It also said it needed more money, a mere two months after securing 6.6 billion in new investment.

But Rus and many others I spoke with at the expo suggest that this hype just doesn't add up. Humanoids "are mostly not intelligent," she said. Rus showed a video of herself speaking to an advanced humanoid that smoothly followed her instruction to pick up a watering can and water a nearby plant. But when she asked it to "water" her friend, the robot did not consider that humans don't need watering like plants and moved to douse the person. "These robots lack common sense," she said.

OpenAI has dialed back a significant restructuring plan, with its nonprofit parent retaining control in a move that is likely to limit CEO Sam Altman's power over the pioneering maker of ChatGPT. The announcement follows a storm of criticism and legal challenges, including a high-profile lawsuit filed by rival and co-founder Elon Musk, who has accused OpenAI of straying from its founding mission to develop artificial intelligence for the benefit of humanity. "OpenAI was founded as a non-profit, is today a non-profit that oversees and controls the for-profit, and going forward will remain a non-profit that oversees and controls the for-profit. That will not change," Altman said in a blog post Monday.

Behind a set of rules in Deontic Defeasible Logic, there is a mapping process of normative background fragments. This process goes from text to rules and implicitly encompasses an explanation of the coded fragments. In this paper we deliver a methodology for \textit{legal coding} that starts with a fragment and goes onto a set of Deontic Defeasible Logic rules, involving a set of \textit{scenarios} to test the correctness of the coded fragments. The methodology is illustrated by the coding process of an example text. We then show the results of a series of experiments conducted with humans encoding a variety of normative backgrounds and corresponding cases in which we have measured the efforts made in the coding process, as related to some measurable features. To process these examples, a recently developed technology, Houdini, that allows reasoning in Deontic Defeasible Logic, has been employed. Finally we provide a technique to forecast time required in coding, that depends on factors such as knowledge of the legal domain, knowledge of the coding processes, length of the text, and a measure of \textit{depth} that refers to the length of the paths of legal references.

Legal practice requires careful adherence to procedural rules. In the United States, few are more complex than those found in The Bluebook: A Uniform System of Citation. Compliance with this system's 500+ pages of byzantine formatting instructions is the raison d'etre of thousands of student law review editors and the bete noire of lawyers everywhere. To evaluate whether large language models (LLMs) are able to adhere to the procedures of such a complicated system, we construct an original dataset of 866 Bluebook tasks and test flagship LLMs from OpenAI, Anthropic, Google, Meta, and DeepSeek. We show (1) that these models produce fully compliant Bluebook citations only 69%-74% of the time and (2) that in-context learning on the Bluebook's underlying system of rules raises accuracy only to 77%. These results caution against using off-the-shelf LLMs to automate aspects of the law where fidelity to procedure is paramount.

Artificial Intelligence (AI) is extensively used across various stages of the recruitment process, from automated candidate sourcing on social media platforms to asynchronous video recruitment methods. A study of Human Resources (HR) professionals representing 500 mid-sized organisations from diverse industries across five countries revealed that 24% of businesses have already implemented AI for recruitment purposes, while 56% of hiring managers plan to adopt it within the next year [Sage, 2020]. AI is employed to augment human decision-making regarding job candidates (such as determining who should receive a job offer) and to support the actions of human decision-makers throughout the process (such as data collection and analysis; Gonzalez, Liu, Shirase, Tomczak, Lobbe, Justenhoven, and Martin [2022]). Some applications incorporating AI algorithms are widely accepted and relatively uncontroversial.

The field of AI safety seeks to prevent or reduce the harms caused by AI systems. A simple and appealing account of what is distinctive of AI safety as a field holds that this feature is constitutive: a research project falls within the purview of AI safety just in case it aims to prevent or reduce the harms caused by AI systems. Call this appealingly simple account The Safety Conception of AI safety. Despite its simplicity and appeal, we argue that The Safety Conception is in tension with at least two trends in the ways AI safety researchers and organizations think and talk about AI safety: first, a tendency to characterize the goal of AI safety research in terms of catastrophic risks from future systems; second, the increasingly popular idea that AI safety can be thought of as a branch of safety engineering. Adopting the methodology of conceptual engineering, we argue that these trends are unfortunate: when we consider what concept of AI safety it would be best to have, there are compelling reasons to think that The Safety Conception is the answer. Descriptively, The Safety Conception allows us to see how work on topics that have historically been treated as central to the field of AI safety is continuous with work on topics that have historically been treated as more marginal, like bias, misinformation, and privacy. Normatively, taking The Safety Conception seriously means approaching all efforts to prevent or mitigate harms from AI systems based on their merits rather than drawing arbitrary distinctions between them.

This paper presents a domain-specific implementation of Retrieval-Augmented Generation (RAG) tailored to the Fair Use Doctrine in U.S. copyright law. Motivated by the increasing prevalence of DMCA takedowns and the lack of accessible legal support for content creators, we propose a structured approach that combines semantic search with legal knowledge graphs and court citation networks to improve retrieval quality and reasoning reliability. Our prototype models legal precedents at the statutory factor level (e.g., purpose, nature, amount, market effect) and incorporates citation-weighted graph representations to prioritize doctrinally authoritative sources. We use Chain-of-Thought reasoning and interleaved retrieval steps to better emulate legal reasoning. Preliminary testing suggests this method improves doctrinal relevance in the retrieval process, laying groundwork for future evaluation and deployment of LLM-based legal assistance tools.

Information on the web, such as scientific publications and Wikipedia, often surpasses users' reading level. To help address this, we used a self-refinement approach to develop a LLM capability for minimally lossy text simplification. To validate our approach, we conducted a randomized study involving 4563 participants and 31 texts spanning 6 broad subject areas: PubMed (biomedical scientific articles), biology, law, finance, literature/philosophy, and aerospace/computer science. Participants were randomized to viewing original or simplified texts in a subject area, and answered multiple-choice questions (MCQs) that tested their comprehension of the text. The participants were also asked to provide qualitative feedback such as task difficulty. Our results indicate that participants who read the simplified text answered more MCQs correctly than their counterparts who read the original text (3.9% absolute increase, p<0.05). This gain was most striking with PubMed (14.6%), while more moderate gains were observed for finance (5.5%), aerospace/computer science (3.8%) domains, and legal (3.5%). Notably, the results were robust to whether participants could refer back to the text while answering MCQs. The absolute accuracy decreased by up to ~9% for both original and simplified setups where participants could not refer back to the text, but the ~4% overall improvement persisted. Finally, participants' self-reported perceived ease based on a simplified NASA Task Load Index was greater for those who read the simplified text (absolute change on a 5-point scale 0.33, p<0.05). This randomized study, involving an order of magnitude more participants than prior works, demonstrates the potential of LLMs to make complex information easier to understand. Our work aims to enable a broader audience to better learn and make use of expert knowledge available on the web, improving information accessibility.

Generative large language models as tools in the legal domain have the potential to improve the justice system. However, the reasoning behavior of current generative models is brittle and poorly understood, hence cannot be responsibly applied in the domains of law and evidence. In this paper, we introduce an approach for creating benchmarks that can be used to evaluate the reasoning capabilities of generative language models. These benchmarks are dynamically varied, scalable in their complexity, and have formally unambiguous interpretations. In this study, we illustrate the approach on the basis of witness testimony, focusing on the underlying argument attack structure. We dynamically generate both linear and non-linear argument attack graphs of varying complexity and translate these into reasoning puzzles about witness testimony expressed in natural language. We show that state-of-the-art large language models often fail in these reasoning puzzles, already at low complexity. Obvious mistakes are made by the models, and their inconsistent performance indicates that their reasoning capabilities are brittle. Furthermore, at higher complexity, even state-of-the-art models specifically presented for reasoning capabilities make mistakes. We show the viability of using a parametrized benchmark with varying complexity to evaluate the reasoning capabilities of generative language models. As such, the findings contribute to a better understanding of the limitations of the reasoning capabilities of generative models, which is essential when designing responsible AI systems in the legal domain.