A Critical Examination of the Ethics of AI-Mediated Peer Review
Laurie A. Schintler, Connie L. McNeely, James Witte,
arXiv,
2023/09/29
I really like this paper (21 page PDF) for two reasons. First, it firmly places the ethics of using AI in contrast with the existing human systems, and in particular, it notes that "human peer review systems are also fraught with related problems, such as biases, abuses, and a lack of transparency, which already diminish credibility." Second, it approaches the question of AI ethics from a perspective of 'legitimacy' of use in scientific research (in contrast to, say, to a social justice perspective). "Science encompasses unique structures, dynamics, and factors that drive perceptions about legitimacy. Different fields and disciplines have their own value systems, priorities, and cultures, including notions of what is legitimate or not." While this article discusses peer review, I would say that the considerations apply to assessment in general across a wide variety of disciplines.
Web: [Direct Link] [This Post]
Technology as a bridge–not a shortcut–to student relationships
Julia Freeland Fisher, Dr. Heather D. Wathington,
eSchool News,
2023/09/29
As I've noted before, what the elite schools deliver that the rest don't is a network of high quality connections. This article underlines the importance of forming networks and offers five strategies for designing tech that supports this process (yeah, it's a listicle, mostly). It's generally good advice, and while students still won't be connected with the old-money elite from the Ivies, with strong networks of their own, they won't need to. For me, a major strength of online learning, and especially its role in supporting personal networking, is to render quaint the old school advantage.
Web: [Direct Link] [This Post]
The promising alliance of generative and discriminative AI
David Benigson,
TechTalks,
2023/09/29
This is interesting and a new (to me, at least) concept. "We already have a complementary tool that can mitigate generative AI's downsides while enhancing its creative potential. This tool, called discriminative AI, is another branch of machine learning specifically designed to evaluate content and categorize new information. If generative AI is your creative, artistic friend who spontaneously shoots wild ideas from the hip, discriminative AI is your no-nonsense buddy with a laser focus on facts." Here's some background on generative vs. discriminative machine learning models; the latter are engines that can learn inference rules, regressions, categorization, and the like, in other words, already familiar applications of AI. See also Fast Company, Turing, Ng & Jordan, Zheng, et al.
Web: [Direct Link] [This Post]
Evidence Is Mounting That Calculus Should Be Changed. Will Instructors Heed It?
Daniel Mollenkamp,
EdSurge,
2023/09/29
The 'direct instruction' crowd isn't going to like this. "The traditional lecture method of teaching calculus isn't as effective as active models. Those who learned from active methods did significantly better across race, gender and major, according to the study." I can believe it. Calculus was my biggest challenge in university. As a loner, I didn't join study groups, and of course the tutors in tutorial sessions were swamped, and so missed out on the active and personal learning that would have helped me succeed. Making active learning the primary mode of instruction would have helped me a lot. The study, published in Science, started with more than 1,000 students, and "followed the Consolidated Standards for Reporting Trials (CONSORT) guidelines."
Web: [Direct Link] [This Post]
New online SA archive aims to better empower citizens
Sane Dhlamini,
Polity,
2023/09/29
This captures both the spirit and purpose of online learning. "Hundreds of historical grassroots democracy documents from the past 29 years have been made public in a bid to empower South Africans and assist them, alongside activists, to hold government to account." It's more than just that, actually, because the resources can also be used by governments themselves; for example, this Local Government Ethics Committee Guidebook should be read by everyone involved. Now I'm sure someone will say "this resource by itself won't create better government." And of course it won't; people need to be encouraged to read the resources, and enabled to act on them. But having free and open resources like this helps a lot. Via Tara Calishain.
Web: [Direct Link] [This Post]
There are many ways to read OLDaily; pick whatever works best for you:
This newsletter is sent only at the request of subscribers. If you would like to unsubscribe, Click here.
Know a friend who might enjoy this newsletter? Feel free to forward OLDaily to your colleagues. If you received this issue from a friend and would like a free subscription of your own, you can join our mailing list. Click here to subscribe.
Copyright 2023 Stephen Downes Contact: stephen@downes.ca
This work is licensed under a Creative Commons License.