20 papers · ranked by Valyu relevance
Roman V. Yampolskiy, M. S. Spellchecker
In this work, we present and analyze reported failures of artificially intelligent systems and extrapolate our analysis to future AIs. We suggest that both the frequency and the seriousness of future AI failures will steadily increase. AI Safety can be improved based on ideas developed by cybersecurity experts. For…
Guangba Yu, G. Tan, Haojia Huang, Zhenyu Zhang + 3 more
'Roberto Natella' 'Zibin Zheng'] The rapid advancement of Artificial Intelligence (AI) has led to its integration into various areas, especially with Large Language Models (LLMs) significantly enhancing capabilities in Artificial Intelligence Generated Content (AIGC). However, the complexity of AI systems has also…
Peter J. H. Scott, Roman V. Yampolskiy
In this paper we examine historical failures of artificial intelligence (AI) and propose a classification scheme for categorizing future failures. By doing so we hope that (a) the responses to future failures can be improved through applying a systematic classification that can be used to simplify the choice of…
Robert M. Williams, Roman V. Yampolskiy
As AI technologies increase in capability and ubiquity, AI accidents are becoming more common. Based on normal accident theory, high reliability theory, and open systems theory, we create a framework for understanding the risks associated with AI applications. In addition, we also use AI safety principles to quantify…
Dong Lv, Rui Sun, Qiuhua Zhu, Jiajia Zuo + 2 more
Background: With the rapid expansion of the generative AI market, conducting in-depth research on cognitive conflicts in human-computer interaction is crucial for optimizing user experience and improving the quality of interactions with AI systems. However, existing studies insufficiently explore the role of user…
Steven A. Moore, Q. Vera Liao, Hariharan Subramonyam
Numerous guidelines have been proposed to design AI experiences (AIX) , including recommendations to mitigate and recover from AI failures . The PAIR playbook, for example, dedicates an entire chapter to errors and "graceful failure", offering best practices for designers to identify and diagnose AI errors and…
David Manheim
An important challenge for safety in machine learning and artificial intelligence systems is a set of related failures involving specification gaming, reward hacking, fragility to distributional shifts, and Goodhart's or Campbell's law. This paper presents additional failure modes for interactions within multi-agent…
Shreyansh Agrawal, Harsh B. Anadkat, Kiran K. Athimoolam, Harsh Bhardwaj + 8 more
Recent advances in artificial intelligence (AI) have prompted claims about autonomous “AI scientists,” yet systematic evaluations of these capabilities remain scarce. This exploratory study investigates whether current AI frameworks can execute scientific research tasks beyond isolated demonstrations. We tested eight…
Takahiro Kato, Daisuke Komura, Binay Panda, Shumpei Ishikawa
We propose Medicine for Artificial Intelligence (MAI), a clinical framework that reconceptualizes AI anomalies as diseases requiring systematic screening, differential diagnosis, treatment, and follow-up. Contemporary discourse on failures (e.g., “hallucination”) is ad hoc and fragmented across domains, impeding…
Mengting Gong, Aimei Li, Junwei Zhang
As intelligent robots are widely applied in people’s work and daily life, intelligent robot service failures have drawn more attention from academics and practitioners. Under the scenarios of intelligent robot service failures, most existing studies focus on service providers’ remedies for the failures and customers’…
Halime S. A. Gulluoglu, Jibin Baby, Kirti M. Bagul, Bhuvan R. Basangari + 17 more
Agentic artificial intelligence (AI) systems increasingly claim to automate scientific research, yet independent evaluations report persistent gaps between those claims and demonstrated capability. We tested frontier agentic AI systems on three practical problems: prediction of treatment non-response in immune-mediated…
Shanee Honig, Tal Oron-Gilad
While substantial effort has been invested in making robots more reliable, experience demonstrates that robots operating in unstructured environments are often challenged by frequent failures. Despite this, robots have not yet reached a level of design that allows effective management of faulty or unexpected behavior…
Mohammed O. Ahmed, Sahil A. Amale, Rhythm D. Bhavsar, Pratham Chopra + 18 more
Artificial Intelligence (AI) frameworks for automating scientific research have shown strong performance on benchmarks, but their capacity to routinely reproduce results from multiple real-life published studies remains largely untested. We evaluated five advanced AI research frameworks (Kosmos, K-Dense, ToolUniverse…
Muhammad Salar Khan, James L. Olds
Neuro-robots are a class of autonomous machines that, in their architecture, mimic aspects of the human brain and cognition. As such, they represent unique artifacts created by humans based on human understanding of healthy human brains. European Union’s Convention on Roboethics 2025 states that the design of all…
Priyanka Bhutada, Nitin Goyal, Tatsam K. Lakhankiya, Sai D. Narahari + 8 more
Artificial Intelligence (AI) frameworks for automating scientific research have shown strong performance on benchmarks, but their utility for real-world industrial research remains insufficiently characterized. Extending the analysis presented in the first paper of this series, we evaluated the same five advanced AI…
Nayeon Kim, Hyuk Jun Yoo, Daeho Kim, Heeseung Lee + 1 more
Autonomous laboratories hold great promise for accelerating material discovery but are often restricted by static, predefined experimental constraints. We present SPACESHIP, an AIdriven framework for dynamic, constraint-free exploration of synthesizable regions in chemical parameter spaces. SPACESHIP integrates…
Authors not listed
This article proposes a three-level classification of artificial intelligence (AI) application in chemical sciences, reflecting the increasing degree of technology involvement in scientific and production processes: from automation of routine tasks (the level of "AI Assistant"), to the creation of specialized…
Authors not listed
Realizing the promise of artificial intelligence (AI) to accelerate scientific progress and deliver technological impact depends on how effectively AI can be integrated into real-world decision- making processes. As Peter Norvig states, “Somewhat remarkably, almost all AI research until very recently has assumed that…
Authors not listed
As the utilization of artificial intelligence (AI) and generative AI (GenAI) is expanding in the educational field, presenting significant implications for STEM disciplines, it is bringing opportunities to enhance how chemistry and chemical engineering are taught and learned. This perspective critically explores the…
Authors not listed
Artificial intelligence (AI) is reshaping scientific research by accelerating discovery and enabling the analysis of complex data that traditional methods struggle to handle. This review examines over 310,000 journal articles and patents from the CAS Content Collection (2015–2025), with a focus on, biomedical research…