Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

ChatGPT programmed to avoid giving accurate information if it may cause offense

#1
C C Offline
https://quillette.com/2023/02/21/artific...t-offense/

EXCERPTS: . . . Recently, Greg Giovanni, a student of Vinod Goel, Professor of Cognitive Neuroscience at York University, held the following “dialogue” with ChatGPT, on an issue of current topical interest, which he and Professor Goel have permitted me to reproduce here [see article for that]...

[...] A reasonable alternative guidance would have been not to spread claims that are unsupported by peer-reviewed studies or any other claims made without a verifiable basis in fact. But to claim that abstract information should be censored if it might be viewed by someone as harmful or offensive is symptomatic of the serious modern social disease prevalent at both ends of the current political spectrum.

While this used to be primarily the province of the extreme Right, one version of the disease is now a facet of the vocal secular religion I am now choosing to call “Fundamentalist Wokism” (to be distinguished from “Woke” as Fundamentalist Islam is distinguished from “Islam”—see my most recent Substack piece in Critical Mass).

Let’s be clear about this: Valid, empirically derived information is not, in the abstract, either harmful or offensive.

The reception of information can be offensive, and it can, depending upon the circumstances of the listener, potentially result in psychological or physical harm. But precisely because one cannot presume to know all such possible circumstances, following the OpenAI guidelines can instead sanction the censorship of almost any kind of information for fear that someone, somewhere, will be offended.

Even before ChatGPT, this was not a hypothetical worry...

[...] A more relevant example, perhaps, involves the loony guidelines recently provided to editors and reviewers for the journals of the Royal Society of Chemistry to “minimise the risk of publishing inappropriate or otherwise offensive content.”

As they describe it, “[o]ffence is a subjective matter and sensitivity to it spans a considerable range; however, we bear in mind that it is the perception of the recipient that we should consider, regardless of the author’s intention [emphasis mine] … Please consider whether or not any content (words, depictions or imagery) might have the potential to cause offence, referring to the guidelines as needed.”

Moreover, they define offensive content specifically as “Any content that could reasonably offend someone on the basis of their age, gender, race, sexual orientation, religious or political beliefs, marital or parental status, physical features, national origin, social status or disability...” (MORE - missing details)
Reply


Possibly Related Threads…
Thread Author Replies Views Last Post
  Article Fresh evidence of ChatGPT’s political bias revealed by comprehensive new study C C 2 97 Aug 17, 2023 12:45 PM
Last Post: stryder
  Many US students won’t find accurate climate science in their textbooks C C 1 84 Nov 17, 2021 05:42 PM
Last Post: Syne
  Information may be the core matter of the universe zhangjinyuan 1 115 Jul 28, 2020 12:25 AM
Last Post: Zinjanthropos



Users browsing this thread: 1 Guest(s)