Frameworks for Artificial Intelligence
Dr. Tamaro J. Green
2021-02-07 17:42:27 viewed: 750
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? described some challenges that artificial intelligence may present to society. Bender, Gebru, and McMillan-Major (2021) annotate characteristics of the data sets that are the foundation of some large language models that might not qualify as quality data. An artificial intelligence framework may provide guidance to improve the process of designing language models Floridi et al. (2018). Floridi et al. (2018) highlight some of the artificial intelligence frameworks that have been developed to address ethical principles and procedures. Abebe et al. (2020) describes roles that computing technologies like artificial intelligence can influence positive social change. Abebe et al. (2020) highlights that computing technologies can assist in diagnosing problems in society, formalizing solutions to those problems, serving as a tool for rebuttal in public discourse, and prioritizing long-standing existing problems. These solutions can be present in a framework for computing technologies such as artificial intelligence Abebe et al. (2020). Cath (2018) explains that an artificial intelligence framework can focus on ethical governance, explainability, interpretability, and ethical auditing.
Bender et al. (2021) describe the importance of carbon neutral energy sources for operating data centers. An artificial intelligence framework may also include evaluations of the technology that operate algorithms (Bender et al., 2021). Rudzicz and Saqur (2020) explain, for example, the dangers that opaque algorithms can present in medicine. An artificial intelligence framework that promotes fairness and privacy and reduces algorithmic bias may be the basis of an ethical artificial intelligence framework (Rudzicz & Saqur, 2020).
Abebe, R., Barocas, S., Kleinberg, J., Levy, K., Raghavan, M., & Robinson, D. G. (2020). Roles for computing in social change. Paper presented at the Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency.
Bender, E., Gebru, T., & McMillan-Major, A. (2021). On the dangers of stochastic parrots: Can language models be too big. Proceedings of FAccT.
Cath, C. (2018). Governing artificial intelligence: Ethical, legal and technical opportunities and challenges. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 376(2133), 20180080. doi:10.1098/rsta.2018.0080
Floridi, L., Cowls, J., Beltrametti, M., Chatila, R., Chazerand, P., Dignum, V., . . . Vayena, E. (2018). AI4People - An ethical framework for a good AI society: Opportunities, risks, principles, and recommendations. Minds and Machines, 28(4), 689-707. doi:10.1007/s11023-018-9482-5
Rudzicz, F., & Saqur, R. (2020). Ethics of artificial intelligence in surgery. arXiv e-prints.