Are AI-Generated Images Biased? – VPN Mentor (David Ngure | July 2023)
Artificial intelligence image generators use machine learning and mathematical algorithms to create images from a description written in natural language.
Artificial intelligence image generators use machine learning and mathematical algorithms to create images from a description written in natural language.
Any unbiased reflection on the ethics of parachute research will conclude that it is ethically challenging and concerning. The practice of conducting research that wouldn’t be acceptable in your own jurisdiction in a poorer and more desperate country is, at best exploitative and, at worst rampant colonialism in research treating populations as lab rats for the benefit primarily in more developed countries. Research institutions, funding bodies and publishers need to take a strong stand when a researcher is obviously engaging in this kind of morally bankrupt behaviour.
This development would do little to ease the rising tensions between these superpowers and it doesn’t settle question of whether the lab in Wuhan was the original source of the COVID-19 pandemic, but it is likely to add fuel to the international furore surrounding these topics. It does highlight an important point, if you’re going to accept funding from another country, be prepared to be judged by that country’s regulatory standards.
It is fair to say we have been and continue to be loud and enthusiastic proponents of the idea that scientific publications should be free to read and free to publish in. Nevertheless, this piece published in the LSE Impact Blog reflects on important questions. 1. What is the reasonable cost of scientific publishing? 2. If readers and authors paying, how are those costs to be met? 3. What does scientific publishing look like after an international drive towards open access publishing?
With several colleagues, I have recently stumbled into investigating what we call ‘untrustworthy’ data in pain. The story started when
We share the views of Aaron Ciechanover (Nobel prizewinning biologist), how we evaluate research outputs and the significance of a researcher’s work is badly off kilter. Rather than looking at the significance and impact of their work, we are focussing on the wrong thing we are exclusively on the number of times a researcher publishes in a high-profile publication. This effectively treats science as something that can be usefully judged by quantitative measures rather than on the basis of its quality.
This story that appeared in Times Higher Education is another callout criticising the fact that while many journal articles say that data will be available upon request when access is requested, it is often not made available. Research institutions and publishers need to start taking action when researchers for you to live up to their undertakings when it comes to data sharing.
Abstract Editorial Independence is an essential principle in scientific publishing, protecting the content from undue influence beyond the scientific data.