Anthropic has partly resolved a legal disagreement that saw the AI startup draw the ire of the music industry. The group alleged that the company had trained its Claude AI model on at least 500 songs to which they held rights and that, when promoted, Claude could reproduce the lyrics of those tracks either partially or in full. Among the song lyrics the publishers said Anthropic had infringed on included Beyoncé's "Halo" and "Moves Like Jagger" by Maroon 5. In cases where the company intends not to address an issue, it must clearly state its intent to do so. "Our decision to enter into this stipulation is consistent with those priorities.

In San Francisco, meanwhile, a lawsuit is underway against the operators of a number of nudify apps. In some instances, the complaint identifies the defendants by name, but in the case of Clothoff, the accused is only listed as "Doe," the name frequently used in the U.S. for unknown defendants. According to the website's imprint, Clothoff is operated out of the Argentinian capital Buenos Aires. But the company has concealed the true identities of its operators through the use of shell companies and other methods. For a time, operators even sought to mislead the public with a fake image, presumably generated by AI, of the purported head of Clothoff.

Apple has agreed to pay 95 million to settle a lawsuit accusing the privacy-minded company of deploying its virtual assistant Siri to eavesdrop on people using its iPhone and other trendy devices. The proposed settlement filed Tuesday in an Oakland, California, federal court would resolve a 5-year-old lawsuit revolving around allegations that Apple surreptitiously activated Siri to record conversations through iPhones and other devices equipped with the virtual assistant for more than a decade. The alleged recordings occurred even when people didn't seek to activate the virtual assistant with the trigger words, "Hey, Siri." Some of the recorded conversations were then shared with advertisers in an attempt to sell their products to consumers more likely to be interested in the goods and services, the lawsuit asserted. The allegations about a snoopy Siri contradicted Apple's long-running commitment to protect the privacy of its customers -- a crusade that CEO Tim Cook has often framed as a fight to preserve "a fundamental human right."

The Sixth Circuit US Court of Appeals ruled yesterday that the FCC does not have the "statutory authority" to implement net neutrality rules. Since the rules were established in 2015, the FCC argued that classifying ISPs as "telecommunication services" gives it broad authority to regulate them. The decision to redefine ISPs as "information services" during the first Trump Administration led to the repeal of net neutrality in 2017. The current FCC voted to restore net neutrality on April 25 last year. The difference between 2015 and now is the Supreme Court's recent, radical reinterpretation of an important legal doctrine.

In the preliminary settlement, the tech firm denies any wrongdoing, as well as claims that it "recorded, disclosed to third parties, or failed to delete, conversations recorded as the result of a Siri activation" without consent. Apple's lawyers also say they will confirm they have "permanently deleted individual Siri audio recordings collected by Apple prior to October 2019". But the claimants say the tech firm recorded people who activated the virtual assistant unintentionally - without using the phrase "Hey, Siri" to wake it. They say advertisers who received the recordings could then look for keywords in them to better target ads. The lead plaintiff Fumiko Lopez claims she and her daughter were both recorded without their consent.

As the impact of global climate change intensifies, corporate carbon emissions have become a focal point of global attention. In response to issues such as the lag in climate change knowledge updates within large language models, the lack of specialization and accuracy in traditional augmented generation architectures for complex problems, and the high cost and time consumption of sustainability report analysis, this paper proposes CarbonChat: Large Language Model-based corporate carbon emission analysis and climate knowledge Q&A system, aimed at achieving precise carbon emission analysis and policy understanding.First, a diversified index module construction method is proposed to handle the segmentation of rule-based and long-text documents, as well as the extraction of structured data, thereby optimizing the parsing of key information.Second, an enhanced self-prompt retrieval-augmented generation architecture is designed, integrating intent recognition, structured reasoning chains, hybrid retrieval, and Text2SQL, improving the efficiency of semantic understanding and query conversion.Next, based on the greenhouse gas accounting framework, 14 dimensions are established for carbon emission analysis, enabling report summarization, relevance evaluation, and customized responses.Finally, through a multi-layer chunking mechanism, timestamps, and hallucination detection features, the accuracy and verifiability of the analysis results are ensured, reducing hallucination rates and enhancing the precision of the responses.

This paper explores the complex tradeoffs between various fairness metrics such as equalized odds, disparate impact, and equal opportunity and predictive accuracy within COMPAS by building neural networks trained with custom loss functions optimized to specific fairness criteria. This paper creates the first fairness-driven implementation of the novel Group Accuracy Parity (GAP) framework, as theoretically proposed by Gupta et al. (2024), and applies it to COMPAS. To operationalize and accurately compare the fairness of COMPAS models optimized to differing fairness ideals, this paper develops and proposes a combinatory analytical procedure that incorporates Pareto front and multivariate analysis, leveraging data visualizations such as violin graphs. This paper concludes that GAP achieves an enhanced equilibrium between fairness and accuracy compared to COMPAS's current nationwide implementation and alternative implementations of COMPAS optimized to more traditional fairness definitions. While this paper's algorithmic improvements of COMPAS significantly augment its fairness, external biases undermine the fairness of its implementation. Practices such as predictive policing and issues such as the lack of transparency regarding COMPAS's internal workings have contributed to the algorithm's historical injustice. In conjunction with developments regarding COMPAS's predictive methodology, legal and institutional changes must happen for COMPAS's just deployment.

Algorithmic fairness of machine learning (ML) models has raised significant concern in the recent years. Many testing, verification, and bias mitigation techniques have been proposed to identify and reduce fairness issues in ML models. The existing methods are model-centric and designed to detect fairness issues under static settings. However, many ML-enabled systems operate in a dynamic environment where the predictive decisions made by the system impact the environment, which in turn affects future decision-making. Such a self-reinforcing feedback loop can cause fairness violations in the long term, even if the immediate outcomes are fair. In this paper, we propose a simulation-based framework called FairSense to detect and analyze long-term unfairness in ML-enabled systems. Given a fairness requirement, FairSense performs Monte-Carlo simulation to enumerate evolution traces for each system configuration. Then, FairSense performs sensitivity analysis on the space of possible configurations to understand the impact of design options and environmental factors on the long-term fairness of the system. We demonstrate FairSense's potential utility through three real-world case studies: Loan lending, opioids risk scoring, and predictive policing.

Existing legal frameworks on AI rely on training compute thresholds as a proxy to identify potentially-dangerous AI models and trigger increased regulatory attention. In the United States, Section 4.2(a) of Executive Order 14110 instructs the Secretary of Commerce to require extensive reporting from developers of AI models above a certain training compute threshold. In the European Union, Article 51 of the AI Act establishes a presumption that AI models above a certain compute threshold have high impact capabilities and hence pose systemic risk, thus subjecting their developers to several obligations including capability evaluations, reporting, and incident monitoring. In this paper, we examine some enhancement techniques that are capable of decreasing training compute usage while preserving, or even increasing, model capabilities. Since training compute thresholds rely on training compute as a metric and trigger for increased regulatory attention, these capability-enhancing and compute-saving techniques could constitute a legal loophole to existing training compute thresholds. In particular, we concentrate on four illustrative techniques (fine-tuning, model reuse, model expansion, and above compute-optimal inference compute) with the goal of furthering the conversation about their implications on training compute thresholds as a legal mechanism and advancing policy recommendations that could address the relevant legal loopholes.

Recent work has investigated the concept of adversarial attacks on explainable AI (XAI) in the NLP domain with a focus on examining the vulnerability of local surrogate methods such as Lime to adversarial perturbations or small changes on the input of a machine learning (ML) model. In such attacks, the generated explanation is manipulated while the meaning and structure of the original input remain similar under the ML model. Such attacks are especially alarming when XAI is used as a basis for decision making (e.g., prescribing drugs based on AI medical predictors) or for legal action (e.g., legal dispute involving AI software). Although weaknesses across many XAI methods have been shown to exist, the reasons behind why remain little explored. Central to this XAI manipulation is the similarity measure used to calculate how one explanation differs from another. A poor choice of similarity measure can lead to erroneous conclusions about the stability or adversarial robustness of an XAI method. Therefore, this work investigates a variety of similarity measures designed for text-based ranked lists referenced in related work to determine their comparative suitability for use. We find that many measures are overly sensitive, resulting in erroneous estimates of stability. We then propose a weighting scheme for text-based data that incorporates the synonymity between the features within an explanation, providing more accurate estimates of the actual weakness of XAI methods to adversarial examples.