Artificial intelligence (AI) is infiltrating just about every facet of society, from detecting fraud and surveillance to helping countries battle the current COVID-19 pandemic. But AI is a thorny subject, fraught with complex terminology, contradictory information, and general confusion about what it is at its most fundamental level. This is why the Oxford Internet Institute (OII), the social and computer science department of the U.K.’s University of Oxford, has partnered with Google to launch a portal with a series of explainers outlining what AI actually is — including the fundamentals, ethics, its impact on society, and how it’s actually created.
At launch, the A-Z of AI covers 26 topics, including bias, how it’s used in climate science, ethics, machine learning, human-in-the-loop, and Generative adversarial networks (GANs).
In terms of how the topics were selected, Google’s people and AI Research team (PAIR) worked with Gina Neff, a senior research fellow and associate professor at OII, and her team to select the specific subjects they felt were pivotal to understanding AI and its role in how the world operates today.
“The 26 topics chosen are by no means an exhaustive list, but they are a great place for first-timers to start,” the guide’s FAQ section explains. “The team carefully balanced their selections across a spectrum of technical understanding, production techniques, use cases, societal implications, and ethical considerations.”
For example, bias in data sets is a well-documented issue in developing AI algorithms, and the A-Z of AI guide briefly explains how the problem is created and how it can be addressed.
“Typically, AI forms a bias when the data it’s given to learn from isn’t fully comprehensive and, therefore, starts leading it toward certain outcomes,” the guide reads. “Because data is an AI system’s only means of learning, it could end up reproducing any imbalances or biases found within the original information. For example, if you were teaching AI to recognize shoes and only showed it imagery of sneakers, it wouldn’t learn to recognize high heels, sandals, or boots as shoes.”
The guide can be perused in its full A-Z form or filtered from one of four categories: AI fundamentals, Making AI, Society and AI, and Using AI.
Those with a decent background in AI will find this guide simplistic, but it’s a good starting point for anyone looking to grasp the key talking points they will be hearing about as AI continues to shape society in the years to come.
It’s also worth noting that this isn’t a static resource — the plan is to update it as AI evolves.
“The A-Z will be refreshed periodically as new technologies come into play and existing technologies evolve,” the guide explains.
Author: Paul Sawers.
Source: Venturebeat