Artificial Intelligence, AI Safety, and AI Governance
I have long-standing interests in understanding the history, growth trajectories, and long-term consequences of artificial intelligence. I am interested in the applications of machine learning and artificial intelligence to the biomedical sciences, and conversely, the relevance of advances in neuroscience and biophysics to the design of AI systems. I am also actively involved in the emerging disciplines of AI safety and AI ethics, including applications of affective neuroscience and comparative neuroanatomy to the development of safe AI goal structures, advancing robust software engineering practices in areas ranging from biological simulation to computer algebra, and foundational research into theories of superintelligence.
Computational Modeling, Data Science, and Software Engineering in Biomedicine
From high-throughput gene sequencing, to novel biological sensors, to new multi-modal imaging technologies, growth in the biomedical sciences is so expansive that we are inundated with new advances coming from many fronts. The ultimate consequence of introducing computational modeling, data science, and software engineering into the biomedical sciences will be no less than a fundamental reconceptualization in the classification, diagnosis, and treatment of disease.
I am interested in all aspects of the intersection of these fields, with a focus on realistic biophysical simulations of model organisms such as Caenorhabditis elegans and applications to the study of disease processes, drug design, and artificial intelligence. Related topics that I have pursued include the broad set of organizational and cultural issues confronting the biomedical sciences as a result of widespread automation and computational methods.
The present state of science, its organizational principles, cultural norms, prevailing paradigms of thought, and many other characteristics have often been the result of incidental, unusual, and surprising historical events. The same can be said of scientific prejudices and of those forces that have at times prevented important trains of thought from developing. As we confront the consequences of unprecedented growth of the scientific enterprise, it is essential that we continually re-examine the basic culture and infrastructure of the scientific establishment.
Metascience encompasses a broad range of topics ranging from traditional epistemology, to science policy, to data driven investigation of the research corpus. Although the term itself has been in use for at least several decades, its connotation in recent years has been shaped by efforts to address the reproducibility crisis. I am interested in a wide range of metascientific issues including innovations and reforms to the journal system, scientific funding, graduate education, large-scale collaborative endeavors (“open science”), and re-examining the historical and philosophical foundations of basic scientific methodology.