Large language models (LLMs) are increasingly being introduced in workplace settings, with the goals of improving efficiency and fairness. However, concerns have arisen regarding these models' potential to reflect or exacerbate social biases and stereotypes. This study explores the potential impact of LLMs on hiring practices. To do so, we conduct an algorithm audit of race and gender biases in one commonly-used LLM, OpenAI's GPT-3.5, taking inspiration from the history of traditional offline resume audits. We conduct two studies using names with varied race and gender connotations: resume assessment (Study 1) and resume generation (Study 2). In Study 1, we ask GPT to score resumes with 32 different names (4 names for each combination of the 2 gender and 4 racial groups) and two anonymous options across 10 occupations and 3 evaluation tasks (overall rating, willingness to interview, and hireability). We find that the model reflects some biases based on stereotypes. In Study 2, we prompt GPT to create resumes (10 for each name) for fictitious job candidates. When generating resumes, GPT reveals underlying biases; women's resumes had occupations with less experience, while Asian and Hispanic resumes had immigrant markers, such as non-native English and non-U.S. education and work experiences. Our findings contribute to a growing body of literature on LLM biases, in particular when used in workplace contexts.

Automatic speech recognition (ASR) systems, increasingly prevalent in education, healthcare, employment, and mobile technology, face significant challenges in inclusivity, particularly for the 80 million-strong global community of people who stutter. These systems often fail to accurately interpret speech patterns deviating from typical fluency, leading to critical usability issues and misinterpretations. This study evaluates six leading ASRs, analyzing their performance on both a real-world dataset of speech samples from individuals who stutter and a synthetic dataset derived from the widely-used LibriSpeech benchmark. The synthetic dataset, uniquely designed to incorporate various stuttering events, enables an in-depth analysis of each ASR's handling of disfluent speech. Our comprehensive assessment includes metrics such as word error rate (WER), character error rate (CER), and semantic accuracy of the transcripts. The results reveal a consistent and statistically significant accuracy bias across all ASRs against disfluent speech, manifesting in significant syntactical and semantic inaccuracies in transcriptions. These findings highlight a critical gap in current ASR technologies, underscoring the need for effective bias mitigation strategies. Addressing this bias is imperative not only to improve the technology's usability for people who stutter but also to ensure their equitable and inclusive participation in the rapidly evolving digital landscape.

The California judge presiding over Elon Musk's lawsuit against OpenAI and its CEO, Sam Altman, has removed himself from the case. Judge Ethan Schulman on Monday sustained a challenge from Musk's lawyers, which cited a California state law that allows plaintiffs and defendants to remove a judge they believe cannot grant an impartial trial. The law, known as California Code of Civil Procedure 170.6, does not require the person issuing the challenge to provide any factual basis for their claim that the judge is prejudiced against them. Each side in a case gets one such peremptory challenge, which is granted as long as it is filed with correct language and within a certain time frame. Lawyers for Altman and Musk did not respond to requests for comment.

When it comes to AI at Adobe, my team has taken a comprehensive approach that includes investment in foundational AI, strategic adoption, an AI ethics framework, legal considerations, security, and content authentication. The rollout follows a phased approach, starting with pilot groups and building communities around AI. This approach includes experimenting with and documenting use cases like writing and editing, data analysis, presentations and employee onboarding, corporate training, employee portals, and improved personalization across HR channels. The rollouts are accompanied by training podcasts and other resources to educate and empower employees to use AI in ways that improve their work and keep them more engaged. While there are innumerable ways that CIOs can leverage generative AI to help surface value at scale for knowledge workers, I'd like to focus on digital documents--a space in which Adobe has been a leader for over 30 years.

There are plenty of people like Sun who want to use AI to preserve, animate, and interact with lost loved ones as they mourn and try to heal. The market is particularly strong in China, where at least half a dozen companies are now offering such technologies and thousands of people have already paid for them. But some question whether interacting with AI replicas of the dead is truly a healthy way to process grief, and it's not entirely clear what the legal and ethical implications of this technology may be. Still, if only 1% of Chinese people can accept AI cloning of the dead, that's still a huge market. To read more about China's flourishing market for deepfakes that clone the dead, check out the latest edition of China Report, our weekly newsletter covering tech in China.

In this paper, we introduce the Interpretable Cross-Examination Technique (ICE-T), a novel approach that leverages structured multi-prompt techniques with Large Language Models (LLMs) to improve classification performance over zero-shot and few-shot methods. In domains where interpretability is crucial, such as medicine and law, standard models often fall short due to their "black-box" nature. ICE-T addresses these limitations by using a series of generated prompts that allow an LLM to approach the problem from multiple directions. The responses from the LLM are then converted into numerical feature vectors and processed by a traditional classifier. This method not only maintains high interpretability but also allows for smaller, less capable models to achieve or exceed the performance of larger, more advanced models under zero-shot conditions. We demonstrate the effectiveness of ICE-T across a diverse set of data sources, including medical records and legal documents, consistently surpassing the zero-shot baseline in terms of classification metrics such as F1 scores. Our results indicate that ICE-T can be used for improving both the performance and transparency of AI applications in complex decision-making environments.

With the rapid advancement of natural language processing technologies, generative artificial intelligence techniques, represented by large language models (LLMs), are gaining increasing prominence and demonstrating significant potential for applications in safety engineering. However, fundamental LLMs face constraints such as limited training data coverage and unreliable responses. This study develops a vector database from 117 explosion accident reports in China spanning 2013 to 2023, employing techniques such as corpus segmenting and vector embedding. By utilizing the vector database, which outperforms the relational database in information retrieval quality, we provide LLMs with richer, more relevant knowledge. Comparative analysis of LLMs demonstrates that ChatSOS significantly enhances reliability, accuracy, and comprehensiveness, improves adaptability and clarification of responses. These results illustrate the effectiveness of supplementing LLMs with an external database, highlighting their potential to handle professional queries in safety engineering and laying a foundation for broader applications.

The trustworthiness of AI applications has been the subject of recent research and is also addressed in the EU's recently adopted AI Regulation. The currently emerging foundation models in the field of text, speech and image processing offer completely new possibilities for developing AI applications. This whitepaper shows how the trustworthiness of an AI application developed with foundation models can be evaluated and ensured. For this purpose, the application-specific, risk-based approach for testing and ensuring the trustworthiness of AI applications, as developed in the 'AI Assessment Catalog - Guideline for Trustworthy Artificial Intelligence' by Fraunhofer IAIS, is transferred to the context of foundation models. Special consideration is given to the fact that specific risks of foundation models can have an impact on the AI application and must also be taken into account when checking trustworthiness. Chapter 1 of the white paper explains the fundamental relationship between foundation models and AI applications based on them in terms of trustworthiness. Chapter 2 provides an introduction to the technical construction of foundation models and Chapter 3 shows how AI applications can be developed based on them. Chapter 4 provides an overview of the resulting risks regarding trustworthiness. Chapter 5 shows which requirements for AI applications and foundation models are to be expected according to the draft of the European Union's AI Regulation and Chapter 6 finally shows the system and procedure for meeting trustworthiness requirements.

Large language models (LLMs) have emerged as an integral part of modern societies, powering user-facing applications such as personal assistants and enterprise applications like recruitment tools. Despite their utility, research indicates that LLMs perpetuate systemic biases. Yet, prior works on LLM harms predominantly focus on Western concepts like race and gender, often overlooking cultural concepts from other parts of the world. Additionally, these studies typically investigate "harm" as a singular dimension, ignoring the various and subtle forms in which harms manifest. To address this gap, we introduce the Covert Harms and Social Threats (CHAST), a set of seven metrics grounded in social science literature. We utilize evaluation models aligned with human assessments to examine the presence of covert harms in LLM-generated conversations, particularly in the context of recruitment. Our experiments reveal that seven out of the eight LLMs included in this study generated conversations riddled with CHAST, characterized by malign views expressed in seemingly neutral language unlikely to be detected by existing methods. Notably, these LLMs manifested more extreme views and opinions when dealing with non-Western concepts like caste, compared to Western ones such as race.

Predictive modeling of criminal recidivism, or whether people will re-offend in the future, has a long and contentious history. Modern causal inference methods allow us to move beyond prediction and target the "treatment effect" of a specific intervention on an outcome in an observational dataset. In this paper, we look specifically at the effect of incarceration (prison time) on recidivism, using a well-known dataset from North Carolina. Two popular causal methods for addressing confounding bias are explained and demonstrated: directed acyclic graph (DAG) adjustment and double machine learning (DML), including a sensitivity analysis for unobserved confounders. We find that incarceration has a detrimental effect on recidivism, i.e., longer prison sentences make it more likely that individuals will re-offend after release, although this conclusion should not be generalized beyond the scope of our data. We hope that this case study can inform future applications of causal inference to criminal justice analysis.