In this report, we argue that there is a realistic possibility that some AI systems will be conscious and/or robustly agentic in the near future. That means that the prospect of AI welfare and moral patienthood -- of AI systems with their own interests and moral significance -- is no longer an issue only for sci-fi or the distant future. It is an issue for the near future, and AI companies and other actors have a responsibility to start taking it seriously. We also recommend three early steps that AI companies and other actors can take: They can (1) acknowledge that AI welfare is an important and difficult issue (and ensure that language model outputs do the same), (2) start assessing AI systems for evidence of consciousness and robust agency, and (3) prepare policies and procedures for treating AI systems with an appropriate level of moral concern. To be clear, our argument in this report is not that AI systems definitely are -- or will be -- conscious, robustly agentic, or otherwise morally significant. Instead, our argument is that there is substantial uncertainty about these possibilities, and so we need to improve our understanding of AI welfare and our ability to make wise decisions about this issue. Otherwise there is a significant risk that we will mishandle decisions about AI welfare, mistakenly harming AI systems that matter morally and/or mistakenly caring for AI systems that do not.

As modern AI models become integral to everyday tasks, concerns about their inherent biases and their potential impact on human decision-making have emerged. While bias in models are well-documented, less is known about how these biases influence human decisions. This paper presents two interactive experiments investigating the effects of partisan bias in AI language models on political decision-making. Participants interacted freely with either a biased liberal, biased conservative, or unbiased control model while completing political decision-making tasks. We found that participants exposed to politically biased models were significantly more likely to adopt opinions and make decisions aligning with the AI's bias, regardless of their personal political partisanship. However, we also discovered that prior knowledge about AI could lessen the impact of the bias, highlighting the possible importance of AI education for robust bias mitigation. Our findings not only highlight the critical effects of interacting with biased AI and its ability to impact public discourse and political conduct, but also highlights potential techniques for mitigating these risks in the future.

"Designed to be deleted" is the tagline of one of the UK's most popular dating apps. Hinge promises that it is "the dating app for people who want to get off dating apps" – the place to find lasting love. But critics say modern dating is in crisis. They claim that dating apps, which have been downloaded hundreds of millions of times worldwide, are "exploitative" and are designed not to be deleted but to be addictive, to retain users in order to create revenue. An Observer investigation has found that dating apps are increasingly pushing users to buy extras that have been likened to "gambling products" and can cost hundreds of pounds a year.

A significant challenge in automating hate speech detection on social media is distinguishing hate speech from regular and offensive language. These identify an essential category of content that web filters seek to remove. Only automated methods can manage this volume of daily data. To solve this problem, the community of Natural Language Processing is currently investigating different ways of hate speech detection. In addition to those, previous approaches (e.g., Convolutional Neural Networks, multi-channel BERT models, and lexical detection) have always achieved low precision without carefully treating other related tasks like sentiment analysis and emotion classification. They still like to group all messages with specific words in them as hate speech simply because those terms often appear alongside hateful rhetoric. In this research, our paper presented the hate speech text classification system model drawn upon deep learning and machine learning. In this paper, we propose a new multitask model integrated with shared emotional representations to detect hate speech across the English language. The Transformer-based model we used from Hugging Face and sentiment analysis helped us prevent false positives. Conclusion. We conclude that utilizing sentiment analysis and a Transformer-based trained model considerably improves hate speech detection across multiple datasets.

Prior study shows that LLMs sometimes generate content that violates copyright. In this paper, we study another important yet underexplored problem, i.e., will LLMs respect copyright information in user input, and behave accordingly? The research problem is critical, as a negative answer would imply that LLMs will become the primary facilitator and accelerator of copyright infringement behavior. We conducted a series of experiments using a diverse set of language models, user prompts, and copyrighted materials, including books, news articles, API documentation, and movie scripts. Our study offers a conservative evaluation of the extent to which language models may infringe upon copyrights when processing user input containing protected material. This research emphasizes the need for further investigation and the importance of ensuring LLMs respect copyright regulations when handling user input to prevent unauthorized use or reproduction of protected content. We also release a benchmark dataset serving as a test bed for evaluating infringement behaviors by LLMs and stress the need for future alignment.

Employee attrition poses significant costs for organizations, with traditional statistical prediction methods often struggling to capture modern workforce complexities. Machine learning (ML) advancements offer more scalable and accurate solutions, but large language models (LLMs) introduce new potential in human resource management by interpreting nuanced employee communication and detecting subtle turnover cues. This study leverages the IBM HR Analytics Attrition dataset to compare the predictive accuracy and interpretability of a fine-tuned GPT-3.5 model against traditional ML classifiers, including Logistic Regression, k-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree, Random Forest, AdaBoost, and XGBoost. While traditional models are easier to use and interpret, LLMs can reveal deeper patterns in employee behavior. Our findings show that the fine-tuned GPT-3.5 model outperforms traditional methods with a precision of 0.91, recall of 0.94, and an F1-score of 0.92, while the best traditional model, SVM, achieved an F1-score of 0.82, with Random Forest and XGBoost reaching 0.80. These results highlight GPT-3.5's ability to capture complex patterns in attrition risk, offering organizations improved insights for retention strategies and underscoring the value of LLMs in HR applications.

LLM agents have the potential to revolutionize defensive cyber operations, but their offensive capabilities are not yet fully understood. To prepare for emerging threats, model developers and governments are evaluating the cyber capabilities of foundation models. However, these assessments often lack transparency and a comprehensive focus on offensive capabilities. In response, we introduce the Catastrophic Cyber Capabilities Benchmark (3CB), a novel framework designed to rigorously assess the real-world offensive capabilities of LLM agents. Our evaluation of modern LLMs on 3CB reveals that frontier models, such as GPT-4o and Claude 3.5 Sonnet, can perform offensive tasks such as reconnaissance and exploitation across domains ranging from binary analysis to web technologies. Conversely, smaller open-source models exhibit limited offensive capabilities. Our software solution and the corresponding benchmark provides a critical tool to reduce the gap between rapidly improving capabilities and robustness of cyber offense evaluations, aiding in the safer deployment and regulation of these powerful technologies.

While `jailbreaks' have been central to research on the safety and reliability of LLMs (large language models), the underlying mechanisms behind these attacks are not well understood. Some prior works have used linear methods to analyze jailbreak prompts or model refusal. Here, however, we compare linear and nonlinear methods to study the features in prompts that contribute to successful jailbreaks. We do this by probing for jailbreak success based only on the portions of the latent representations corresponding to prompt tokens. First, we introduce a dataset of 10,800 jailbreak attempts from 35 attack methods. We then show that different jailbreaking methods work via different nonlinear features in prompts. Specifically, we find that while probes can distinguish between successful and unsuccessful jailbreaking prompts with a high degree of accuracy, they often transfer poorly to held-out attack methods. We also show that nonlinear probes can be used to mechanistically jailbreak the LLM by guiding the design of adversarial latent perturbations. These mechanistic jailbreaks are able to jailbreak Gemma-7B-IT more reliably than 34 of the 35 techniques that it was trained on. Ultimately, our results suggest that jailbreaks cannot be thoroughly understood in terms of universal or linear prompt features alone.

Hosted by Ben Byford, The Machine Ethics Podcast brings together interviews with academics, authors, business leaders, designers and engineers on the subject of autonomous algorithms, artificial intelligence, machine learning, and technology's impact on society. In this episode we're chatting to Lisa about: Data and AI literacy, data sharing, data governance and data wallets, design values, selling in ethics to organisations, contractual agreements and ethical frameworks, AI unlearning, what organisations needs to know about ethics, and an AI ethics consultant directory… Lisa Talia Moretti is a Digital Sociologist based in the UK. She holds a MSc Digital Sociology and 17 years of experience working at the intersection of design research, social theory and technology. Lisa is the Chair of the AI Council at BIMA and a board member of the Conversation Design Institute Foundation. In 2020, Lisa was named one of Britain's 100 people who are shaping the digital industry in the category Champion for Change.

Vision-language models, like CLIP (Contrastive Language Image Pretraining), are becoming increasingly popular for a wide range of multimodal retrieval tasks. However, prior work has shown that large language and deep vision models can learn historical biases contained in their training sets, leading to perpetuation of stereotypes and potential downstream harm. In this work, we conduct a systematic analysis of the social biases that are present in CLIP, with a focus on the interaction between image and text modalities. We first propose a taxonomy of social biases called So-B-IT, which contains 374 words categorized across ten types of bias. Each type can lead to societal harm if associated with a particular demographic group. Using this taxonomy, we examine images retrieved by CLIP from a facial image dataset using each word as part of a prompt. We find that CLIP frequently displays undesirable associations between harmful words and specific demographic groups, such as retrieving mostly pictures of Middle Eastern men when asked to retrieve images of a "terrorist". Finally, we conduct an analysis of the source of such biases, by showing that the same harmful stereotypes are also present in a large image-text dataset used to train CLIP models for examples of biases that we find. Our findings highlight the importance of evaluating and addressing bias in vision-language models, and suggest the need for transparency and fairness-aware curation of large pre-training datasets.