Meta's Muse takes a fee from the merchants it recommends. ChatGPT is testing ads. The experiments now measure how an agent's employer, prompt and revenue model bend what it picks for you, and whether you would notice.
Meta's Muse earns a fee from merchants such as Expedia. ChatGPT is testing ads on its free tiers.
Follows 22 news and primary sources, 31 papers and 74 inquiring lines.
Our mission is to ensure that artificial general intelligence benefits all of humanity. As outlined recently by Sam Altman and Jakub Pachocki, we prioritize our work towards this mission with three main goals: Navigate the next period of AI progress, by building an automated AI researcher, iterating with it on the alignment problem, and finding ways for people to remain part of the self-improvement loop. Automated AI research can involve varying degrees of human supervision.
Tong Zheng, Xidong Wu, Zheng Zhang et al. · 2026-09-14
Recursive self-improvement is becoming increasingly vital for autonomous AI agents, where progress hinges on discovering high-value solutions across complex domains. The driver of this process is effective exploration, however, managing and improving exploration strategies remains a major bottleneck.
Yunhao Chen, Xin Wang, Yixu Wang et al. · 2026-08-01
AI agents operate in persistent environments where early state changes can influence decisions far into the future. Unlike conventional language-model interactions, agent behavior is mediated through a shared state that is repeatedly modified and reused across long-horizon workflows. Current safety benchmarks often fail to capture these cumulative risks because they focus on short, static tasks.
Zheyuan Zhang, Alvin Zhang, Daniel Khashabi et al. · 2026-09-07
Language models may need to internalize information that arrives over time and retain it through many subsequent updates. To study this challenge, we introduce long-horizon memorization, a setting in which a model learns 100 query-answer tasks through continual supervised fine-tuning without retaining earlier training examples or receiving task identifiers at inference.
Varshini Elangovan, James Wedgwood, Chhavi Yadav et al. · Sep 2026
Conversational AI systems can pose safety risks to their users such as hallucination, sycophancy, overconfidence, and anthropomorphism, but these risks are difficult for users to detect during everyday use. We introduce Safety Nudges, a browser-based tool that provides lightweight, in situ flags when concerning behavior is detected in chatbot conversations. We evaluated Safety Nudges in a two-week field study with 45 frequent chatbot users, collecting interaction logs, surveys, and feedback on individual nudges.
Maan Qraitem, Kate Saenko, Bryan A. Plummer · Sep 2026
Procedural character generation aims to populate games, simulations, and other virtual worlds with diverse characters. Large language models (LLMs) offer a promising foundation for scaling this task. However, LLM-based procedural character generation remains at an early stage: existing methods either generate characters directly or adapt profiles retrieved from persona banks.
Ewelina Gajewska, Katarzyna Budzynska, Jaroslaw Chudziak · Sep 2026
Large Language Models (LLMs) are increasingly deployed as argumentative agents in persuasive dialogues, necessitating rigorous evaluation of their debating competence relative to human interlocutors. In this study, we focus on character attacks (ad hominem arguments), traditionally dismissed as fallacies, which play a pivotal role in political persuasive dialogues where ethos often rivals propositional content.