Related papers: The Shutdown Problem: An AI Engineering Puzzle for…
Artificial Intelligence (AI) achieved super-human performance in a broad variety of domains. We say that an AI is made Artificially Stupid on a task when some limitations are deliberately introduced to match a human's ability to do the…
Under the assumption $\mathcal{P} \neq \mathcal{NP}$, we prove that two natural problems from the theory of synchronizing automata cannot be solved in polynomial time. The first problem is to decide whether a given reachable partial…
Artificially intelligent agents deployed in the real-world will require the ability to reliably \textit{cooperate} with humans (as well as other, heterogeneous AI agents). To provide formal guarantees of successful cooperation, we must make…
This paper focuses on a dynamic aspect of responsible autonomy, namely, to make intelligent agents be responsible at run time. That is, it considers settings where decision making by agents impinges upon the outcomes perceived by other…
AI agents that take actions in their environment autonomously over extended time horizons require robust governance interventions to curb their potentially consequential risks. Prior proposals for governing AI agents primarily target…
Defining artificial intelligence (AI) is a persistent challenge, often muddied by technical ambiguity and varying interpretations. Commonly used definitions heavily emphasize technical properties of AI but neglect the human purpose of it.…
In this paper, we consider one-to-one matchings between two disjoint groups of agents. Each agent has a preference over a subset of the agents in the other group, and these preferences may contain ties. Strong stability is one of the…
Decision-making algorithms are being used in important decisions, such as who should be enrolled in health care programs and be hired. Even though these systems are currently deployed in high-stakes scenarios, many of them cannot explain…
This position paper argues that AI agents should be regulated by the extent to which they operate autonomously. AI agents with long-term planning and strategic capabilities can pose significant risks of human extinction and irreversible…
Runtime enforcement refers to the theories, techniques, and tools for enforcing correct behavior with respect to a formal specification of systems at runtime. In this paper, we are interested in techniques for constructing runtime enforcers…
Modern AI assistants are trained to follow instructions, implicitly assuming that users can clearly articulate their goals and the kind of assistance they need. Decades of behavioral research, however, show that people often engage with AI…
Since Artificial Intelligence (AI) software uses techniques like deep lookahead search and stochastic optimization of huge neural networks to fit mammoth datasets, it often results in complex behavior that is difficult for people to…
Autonomy is a double-edged sword for AI agents, simultaneously unlocking transformative possibilities and serious risks. How can agent developers calibrate the appropriate levels of autonomy at which their agents should operate? We argue…
Imposing constraints on the output of a Deep Neural Net is one way to improve the quality of its predictions while loosening the requirements for labeled training data. Such constraints are usually imposed as soft constraints by adding new…
Using a model in which agents compete to develop a potentially dangerous new technology (AI), we study how changes in the pricing of factors of production (computational resources) affect agents' strategies, particularly their spending on…
Arthur [1,2] provided a model to explain the circumstances that lead to technological lock-in into a specific trajectory. We contribute substantially to this area of research by investigating the circumstances under which technological…
Notifications provide a unique mechanism for increasing the effectiveness of real-time information delivery systems. However, notifications that demand users' attention at inopportune moments are more likely to have adverse effects and…
As AI systems become more advanced, companies and regulators will make difficult decisions about whether it is safe to train and deploy them. To prepare for these decisions, we investigate how developers could make a 'safety case,' which is…
This book-length article combines several peer reviewed papers and new material to analyze the issues of ethical artificial intelligence (AI). The behavior of future AI systems can be described by mathematical equations, which are adapted…
We consider the problem of designing agents able to compute optimal decisions by composing data from multiple sources to tackle tasks involving: (i) tracking a desired behavior while minimizing an agent-specific cost; (ii) satisfying safety…