The rapid rise of AI-based autonomous agents is transforming human society and economic systems, as these entities increasingly exhibit human-like or superhuman intelligence. From excelling at complex games like Go to tackling diverse general-purpose tasks with large language and multimodal models, AI agents are evolving from specialized tools into dynamic participants in social and economic ecosystems. Their autonomy and decision-making capabilities are poised to impact industries, professions, and human lives profoundly, raising critical questions about their integration into economic activities, potential ethical concerns, and the balance between their utility and safety. To address these challenges, this paper presents ten principles of AI agent economics, offering a framework to understand how AI agents make decisions, influence social interactions, and participate in the broader economy. Drawing on economics, decision theory, and ethics, we explore fundamental questions, such as whether AI agents might evolve from tools into independent entities, their impact on labor markets, and the ethical safeguards needed to align them with human values. These principles build on existing economic theories while accounting for the unique traits of AI agents, providing a roadmap for their responsible integration into human systems. Beyond theoretical insights, this paper highlights the urgency of future research into AI trustworthiness, ethical guidelines, and regulatory oversight. As we enter a transformative era, this work serves as both a guide and a call to action, ensuring AI agents contribute positively to human progress while addressing risks tied to their unprecedented capabilities.

LLMs have made impressive progress, but their growing capabilities also expose them to highly flexible jailbreaking attacks designed to bypass safety alignment. While many existing defenses focus on known types of attacks, it is more critical to prepare LLMs for unseen attacks that may arise during deployment. To address this, we propose a lifelong safety alignment framework that enables LLMs to continuously adapt to new and evolving jailbreaking strategies. Our framework introduces a competitive setup between two components: a Meta-Attacker, trained to actively discover novel jailbreaking strategies, and a Defender, trained to resist them. To effectively warm up the Meta-Attacker, we first leverage the GPT-4o API to extract key insights from a large collection of jailbreak-related research papers. Through iterative training, the first iteration Meta-Attacker achieves a 73% attack success rate (ASR) on RR and a 57% transfer ASR on LAT using only single-turn attacks. Meanwhile, the Defender progressively improves its robustness and ultimately reduces the Meta-Attacker's success rate to just 7%, enabling safer and more reliable deployment of LLMs in open-ended environments. The code is available at https://github.com/sail-sg/LifelongSafetyAlignment.

Political biases encoded by LLMs might have detrimental effects on downstream applications. Existing bias analysis methods rely on small-size intermediate tasks (questionnaire answering or political content generation) and rely on the LLMs themselves for analysis, thus propagating bias. We propose a new approach leveraging the observation that LLM sentiment predictions vary with the target entity in the same sentence. We define an entropy-based inconsistency metric to encode this prediction variability. We insert 1319 demographically and politically diverse politician names in 450 political sentences and predict target-oriented sentiment using seven models in six widely spoken languages. We observe inconsistencies in all tested combinations and aggregate them in a statistically robust analysis at different granularity levels. We observe positive and negative bias toward left and far-right politicians and positive correlations between politicians with similar alignment. Bias intensity is higher for Western languages than for others. Larger models exhibit stronger and more consistent biases and reduce discrepancies between similar languages. We partially mitigate LLM unreliability in target-oriented sentiment classification (TSC) by replacing politician names with fictional but plausible counterparts.

We investigate long-context vulnerabilities in Large Language Models (LLMs) through Many-Shot Jailbreaking (MSJ). Our experiments utilize context length of up to 128K tokens. Through comprehensive analysis with various many-shot attack settings with different instruction styles, shot density, topic, and format, we reveal that context length is the primary factor determining attack effectiveness. Critically, we find that successful attacks do not require carefully crafted harmful content. Even repetitive shots or random dummy text can circumvent model safety measures, suggesting fundamental limitations in long-context processing capabilities of LLMs. The safety behavior of well-aligned models becomes increasingly inconsistent with longer contexts. These findings highlight significant safety gaps in context expansion capabilities of LLMs, emphasizing the need for new safety mechanisms.

Finite-difference methods are widely used for zeroth-order optimization in settings where gradient information is unavailable or expensive to compute. These procedures mimic first-order strategies by approximating gradients through function evaluations along a set of random directions. From a theoretical perspective, recent studies indicate that imposing structure (such as orthogonality) on the chosen directions allows for the derivation of convergence rates comparable to those achieved with unstructured random directions (i.e., directions sampled independently from a distribution). Empirically, although structured directions are expected to enhance performance, they often introduce additional computational costs, which can limit their applicability in high-dimensional settings. In this work, we examine the impact of structured direction selection in finite-difference methods. We review and extend several strategies for constructing structured direction matrices and compare them with unstructured approaches in terms of computational cost, gradient approximation quality, and convergence behavior. Our evaluation spans both synthetic tasks and real-world applications such as adversarial perturbation. The results demonstrate that structured directions can be generated with computational costs comparable to unstructured ones while significantly improving gradient estimation accuracy and optimization performance.

As Large Language Models (LLMs) become deeply integrated into human life and increasingly influence decision-making, it's crucial to evaluate whether and to what extent they exhibit subjective preferences, opinions, and beliefs. These tendencies may stem from biases within the models, which may shape their behavior, influence the advice and recommendations they offer to users, and potentially reinforce certain viewpoints. This paper presents the Preference, Opinion, and Belief survey (POBs), a benchmark developed to assess LLMs' subjective inclinations across societal, cultural, ethical, and personal domains. We applied our benchmark to evaluate leading open- and closed-source LLMs, measuring desired properties such as reliability, neutrality, and consistency. In addition, we investigated the effect of increasing the test-time compute, through reasoning and self-reflection mechanisms, on those metrics. While effective in other tasks, our results show that these mechanisms offer only limited gains in our domain. Furthermore, we reveal that newer model versions are becoming less consistent and more biased toward specific viewpoints, highlighting a blind spot and a concerning trend. POBS: https://ibm.github.io/POBS

State-of-the-art deep learning models for tabular data have recently achieved acceptable performance to be deployed in industrial settings. However, the robustness of these models remains scarcely explored. Contrary to computer vision, there are no effective attacks to properly evaluate the adversarial robustness of deep tabular models due to intrinsic properties of tabular data, such as categorical features, immutability, and feature relationship constraints. To fill this gap, we first propose CAPGD, a gradient attack that overcomes the failures of existing gradient attacks with adaptive mechanisms. This new attack does not require parameter tuning and further degrades the accuracy, up to 81\% points compared to the previous gradient attacks.

Vision transformers (ViTs) perform exceptionally well in various computer vision tasks but remain vulnerable to adversarial attacks. Recent studies have shown that the transferability of adversarial examples exists for CNNs, and the same holds true for ViTs. However, existing ViT attacks aggressively regularize the largest token gradients to exact zero within each layer of the surrogate model, overlooking the interactions between layers, which limits their transferability in attacking black-box models. Therefore, in this paper, we focus on boosting the transferability of adversarial attacks on ViTs through adaptive token tuning (ATT). Specifically, we propose three optimization strategies: an adaptive gradient re-scaling strategy to reduce the overall variance of token gradients, a self-paced patch out strategy to enhance the diversity of input tokens, and a hybrid token gradient truncation strategy to weaken the effectiveness of attention mechanism.

Drug-induced toxicity is one of the leading reasons new drugs fail clinical trials. Machine learning models that predict drug toxicity from molecular structure could help researchers prioritize less toxic drug candidates. However, current toxicity datasets are typically small and limited to a single organ system (e.g., cardio, renal, or liver). Creating these datasets often involved time-intensive expert curation by parsing drug labelling documents that can exceed 100 pages per drug. Here, we introduce UniTox, a unified dataset of 2,418 FDA-approved drugs with drug-induced toxicity summaries and ratings created by using GPT-4o to process FDA drug labels.

A UFO researcher has an ominous message for humanity as governments around the world begin releasing more information about alleged contact with extraterrestrials. Dr Julia Mossbridge is a cognitive neuroscientist and a researcher of unidentified aerial phenomena (UAP) - the new term for UFOs and alien sightings. After scientists in Colombia recovered a mysterious, sphere-shaped object that many now believe is a piece of UFO technology, Mossbridge said the world is moving into an era which may soon have to deal with the knowledge that aliens exist. 'We are entering a time when we are starting to recognize as humans we don't have the control that we thought we had over everything,' Dr Mossbridge told Fox News. However, Mossbridge, who studies how humans think and also attended the May 1 congressional hearing on UAPs, said the impending disclosure of alien life could throw the worldview of many people into chaos.