The Need for Ethical, Responsible, and Trustworthy Artificial Intelligence for Environmental Sciences

@article{McGovern2022TheNF,
  title={The Need for Ethical, Responsible, and Trustworthy Artificial Intelligence for Environmental Sciences},
  author={Amy McGovern and Imme Ebert‐Uphoff and David John Gagne and Ann Bostrom},
  journal={ArXiv},
  year={2022},
  volume={abs/2112.08453}
}
Given the growing use of Artificial intelligence (AI) and machine learning (ML) methods across all aspects of environmental sciences, it is imperative that we initiate a discussion about the ethical and responsible use of AI. In fact, much can be learned from other domains where AI was introduced, often with the best of intentions, yet often led to unintended societal consequences, such as hard coding racial bias in the criminal justice system or increasing economic inequality through the… 

Figures from this paper

The future of artificial intelligence in the context of industrial ecology

Artificial intelligence (AI) applications and digital technologies (DTs) are increasingly present in the daily lives of citizens, in cities and in industries. These developments generate large

NSF AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES)

We introduce the National Science Foundation (NSF) AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES). This AI institute was funded in 2020 as part of

The Evolving Role of Humans in Weather Prediction and Communication

The article closes with the vision for the ways that humans can maintain an essential role in weather prediction and communication, highlighting the interdependent relationship between computers and humans.

A sensitivity analysis of a regression model of ocean temperature

There has been much recent interest in developing data-driven models for weather and climate predictions. However, there are open questions regarding their generalizability and robustness,

A Machine Learning Tutorial for Operational Meteorology, Part I: Traditional Machine Learning

A survey of some of the most common machine learning methods in meteorology, including linear regression, logistic regression, decision trees, random forest, gradient boosted decision Trees, naïve Bayes, and support vector machines is provided.

ESA-ECMWF Report on recent progress and research directions in machine learning for Earth System observation and prediction

A physics-augmented deep learning model, called “DeepTCNet”, is developed to estimate TC intensity and wind radii from infrared (IR) imagery over the North Atlantic Ocean, which shows better performance than several state of theart estimation techniques of TCs.

Segmenting across places: The need for fair transfer learning with satellite imagery

The increasing availability of high-resolution satellite imagery has enabled the use of machine learning to support land-cover measurement and inform policy-making. However, labelling satellite

Downscaling atmospheric chemistry simulations with physically consistent deep learning

Abstract. Recent advances in deep convolutional neural network (CNN)-based super resolution can be used to downscale atmospheric chemistry simulations with substantially higher accuracy than

References

SHOWING 1-10 OF 73 REFERENCES

AI for climate: freedom, justice, and other ethical and political challenges

This article outlines and discusses issues, with a focus on problems concerning freedom and justice at a global level, and calls for responsible use of AI for climate in the light of these challenges.

Artificial intelligence in the water domain: Opportunities for responsible use

  • N. Doorn
  • Computer Science
    Science of The Total Environment
  • 2020

A Survey on Bias and Fairness in Machine Learning

This survey investigated different real-world applications that have shown biases in various ways, and created a taxonomy for fairness definitions that machine learning researchers have defined to avoid the existing bias in AI systems.

The ethics of scientific communication under uncertainty

Communication by scientists with policy makers and attentive publics raises ethical issues. Scientists need to decide how to communicate knowledge effectively in a way that nonscientists can

Understanding Risk - Informing Decisions in a Democratic Society

This is the advice of a committee of leading American experts on risk characterisation, which recognises that involvement of all those with vested interests cannot ensure a rapid or consensus solution or preclude some groups `dropping out' and choosing the route of litigation.

Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy

Cathy O’Neil’s Weapons of Math Destruction is a timely reminder of the power and perils of predictive algorithms and model-driven decision processes and speaks forcefully to the cultural moment the authors share.

Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy

In 2008, Cathy O’Neil, a data scientist at the hedge fund D.E. Shaw, watched as the entire financial sector around her imploded.1 The financial crisis gave O’Neil, a self-proclaimed math nerd with a

Preventing undesirable behavior of intelligent machines

A general framework for algorithm design is introduced in which the burden of avoiding undesirable behavior is shifted from the user to the designer of the algorithm, and this framework simplifies the problem of specifying and regulating undesirable behavior.

Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead

  • C. Rudin
  • Computer Science
    Nat. Mach. Intell.
  • 2019
This Perspective clarifies the chasm between explaining black boxes and using inherently interpretable models, outlines several key reasons why explainable black boxes should be avoided in high-stakes decisions, identifies challenges to interpretable machine learning, and provides several example applications whereinterpretable models could potentially replace black box models in criminal justice, healthcare and computer vision.

Deepfakes and Disinformation: Exploring the Impact of Synthetic Political Video on Deception, Uncertainty, and Trust in News

Artificial Intelligence (AI) now enables the mass creation of what have become known as “deepfakes”: synthetic videos that closely resemble real videos. Integrating theories about the power of visual
...