Individuals and organizations are still struggling with how and how much to integrate AI into daily life. Rarely has that been more clear than a case out of Emory University in which the school went from awarding students with an entrepreneurship prize worth 10,000 for their AI-powered studying tool to suspending them for it, 404 Media reports. No, the students didn't suddenly misuse the tool, known as Eightball, in any way; they did just as they said they would, and all the while, Emory promoted them -- until they didn't. Eightball allowed students to turn any coursework or readings into practice tests or flashcards for studying. It also connected to Canvas -- the platform professors at Emory use to share course documents with their students.

On 21 May 2024, the Council of the European Union formally approved the artificial intelligence (AI) Act. The legislative act will come into force in about three weeks' time, with the new regulations being phased in over the course of the coming months and years. According to the Council, the new law aims to "foster the development and uptake of safe and trustworthy AI systems across the EU's single market by both private and public actors. At the same time, it aims to ensure respect of fundamental rights of EU citizens and stimulate investment and innovation on artificial intelligence in Europe." The legislation is designed to follow a risk-based approach, with the higher the risk a system poses, the stricter the rules relating to its use and/or release.

In a product demo last week, OpenAI showcased a synthetic but expressive voice for ChatGPT called "Sky" that reminded many viewers of the flirty AI girlfriend Samantha played by Scarlett Johansson in the 2013 film Her. One of those viewers was Johansson herself, who promptly hired legal counsel and sent letters to OpenAI demanding an explanation, according to a statement released later. In response, the company on Sunday halted use of Sky and published a blog post insisting that it "is not an imitation of Scarlett Johansson but belongs to a different professional actress using her own natural speaking voice." Johansson's statement, released Monday, said she was "shocked, angered, and in disbelief" by OpenAI's demo using a voice she called "so eerily similar to mine that my closest friends and news outlets could not tell the difference." Johansson revealed that she had turned down a request last year from the company's CEO, Sam Altman, to voice ChatGPT and that he had reached out again two days before last week's demo in an attempt to change her mind.

Safety and trustworthiness are indispensable requirements for real-world applications of AI systems using large language models (LLMs). This paper formulates human value alignment as an optimization problem of the language model policy to maximize reward under a safety constraint, and then proposes an algorithm, Stepwise Alignment for Constrained Policy Optimization (SACPO). One key idea behind SACPO, supported by theory, is that the optimal policy incorporating reward and safety can be directly obtained from a reward-aligned policy. Building on this key idea, SACPO aligns LLMs step-wise with each metric while leveraging simple yet powerful alignment algorithms such as direct preference optimization (DPO). SACPO offers several advantages, including simplicity, stability, computational efficiency, and flexibility of algorithms and datasets. Under mild assumptions, our theoretical analysis provides the upper bounds on optimality and safety constraint violation. Our experimental results show that SACPO can fine-tune Alpaca-7B better than the state-of-the-art method in terms of both helpfulness and harmlessness.

Scaling laws have driven remarkable progress across machine learning domains like language modeling and computer vision. However, the exploration of scaling laws in embodied AI and robotics has been limited, despite the rapidly increasing usage of machine learning in this field. This paper presents the first study to quantify scaling laws for Robot Foundation Models (RFMs) and the use of LLMs in robotics tasks. Through a meta-analysis spanning 198 research papers, we analyze how key factors like compute, model size, and training data quantity impact model performance across various robotic tasks. Our findings confirm that scaling laws apply to both RFMs and LLMs in robotics, with performance consistently improving as resources increase. The power law coefficients for RFMs closely match those of LLMs in robotics, resembling those found in computer vision and outperforming those for LLMs in the language domain. We also note that these coefficients vary with task complexity, with familiar tasks scaling more efficiently than unfamiliar ones, emphasizing the need for large and diverse datasets. Furthermore, we highlight the absence of standardized benchmarks in embodied AI. Most studies indicate diminishing returns, suggesting that significant resources are necessary to achieve high performance, posing challenges due to data and computational limitations. Finally, as models scale, we observe the emergence of new capabilities, particularly related to data and model size.

This paper presents an extended version of the SPADE platform, which aims to empower intelligent agent systems with normative reasoning and value alignment capabilities. Normative reasoning involves evaluating social norms and their impact on decision-making, while value alignment ensures agents' actions are in line with desired principles and ethical guidelines. The extended platform equips agents with normative awareness and reasoning capabilities based on deontic logic, allowing them to assess the appropriateness of their actions and make informed decisions. By integrating normative reasoning and value alignment, the platform enhances agents' social intelligence and promotes responsible and ethical behaviors in complex environments.

When using machine learning for automated prediction, it is important to account for fairness in the prediction. Fairness in machine learning aims to ensure that biases in the data and model inaccuracies do not lead to discriminatory decisions. E.g., predictions from fair machine learning models should not discriminate against sensitive variables such as sexual orientation and ethnicity. The training data often in obtained from social surveys. In social surveys, oftentimes the data collection process is a strata sampling, e.g. due to cost restrictions. In strata samples, the assumption of independence between the observation is not fulfilled. Hence, if the machine learning models do not account for the strata correlations, the results may be biased. Especially high is the bias in cases where the strata assignment is correlated to the variable of interest. We present in this paper an algorithm that can handle both problems simultaneously, and we demonstrate the impact of stratified sampling on the quality of fair machine learning predictions in a reproducible simulation study.

Our study addresses the challenges of building datasets to understand the risks associated with organized activities and human trafficking through commercial sex advertisements. These challenges include data scarcity, rapid obsolescence, and privacy concerns. Traditional approaches, which are not automated and are difficult to reproduce, fall short in addressing these issues. We have developed a reproducible and automated methodology to analyze five million advertisements. In the process, we identified further challenges in dataset creation within this sensitive domain. This paper presents a streamlined methodology to assist researchers Figure 1: Methodology to generate a pseudo-labeled in constructing effective datasets for combating dataset in human trafficking risk prediction and organized organized crime, allowing them to focus on activity detection tasks.

To ensure unbiased and ethical automated predictions, fairness must be a core principle in machine learning applications. Fairness in machine learning aims to mitigate biases present in the training data and model imperfections that could lead to discriminatory outcomes. This is achieved by preventing the model from making decisions based on sensitive characteristics like ethnicity or sexual orientation. A fundamental assumption in machine learning is the independence of observations. However, this assumption often does not hold true for data describing social phenomena, where data points are often clustered based. Hence, if the machine learning models do not account for the cluster correlations, the results may be biased. Especially high is the bias in cases where the cluster assignment is correlated to the variable of interest. We present a fair mixed effects support vector machine algorithm that can handle both problems simultaneously. With a reproducible simulation study we demonstrate the impact of clustered data on the quality of fair machine learning predictions.

Causal reasoning, in particular, counterfactual reasoning plays a central role in testing for discrimination. Counterfactual reasoning materializes when testing for discrimination, what is known as the counterfactual model of discrimination, when we compare the discrimination comparator with the discrimination complainant, where the comparator is a similar (or similarly situated) profile to that of the complainant used for testing the discrimination claim of the complainant. In this paper, we revisit the comparator by presenting two kinds of comparators based on the sort of causal intervention we want to represent. We present the ceteris paribus and the mutatis mutandis comparator, where the former is the standard and the latter is a new kind of comparator. We argue for the use of the mutatis mutandis comparator, which is built on the fairness given the difference notion, for testing future algorithmic discrimination cases.