69 million page views

If you do not like science, just change it

Reader comment on item: Covering Muslims
in response to reader comment: For all you know, Erik Bleich and Maurits van der Veen are right

Submitted by Prashant (United States), Jun 13, 2022 at 16:46

Dear Dr Pipes, Authors Erik Bleich and Maurits van der Veen did a study of news reports and found that these reports linked Islam to violence and so are biased against Islam. Instead of treating this study as an objective indicator that Islam was more associated with violence, they chose to assume that the media was biased against Islam. In short, if you do not like objective science, you can malign the people who created the data. This is the good-old shoot the messenger strategy!!

Let me illustrate it with an easy to understand example. A lot of computer models work by identifying word associations. We all know that the world famous Taj Mahal was built by the Muslim king Shahjahan. People have written hundreds of essays and books stating that Shahjahan built Tajmahal. If we feed these books and essays to computers, the computers will soon "learn" that Tajmahal and Sahjahan have something to do with each other. It is as simple as that. If a computer program produces something to this effect, we wont say that this program is biased in favor of Muslims or Shahjahan. We will just say that the computer program rediscovered the known association between the the Taj and the king.

But when the same science discovers a link between Islam and violence, we start to manipulate the science and say that media is biased. I am writing this note to protect the scientific methods. Protecting scientific methodologies is more important for humanity than protecting Islam or any other philosophy. People should not start distrusting science because it links Islam with violence. I would rather call Erik Bleich and Maurits van der Veen dishonest charlatans than honest scientists.

Unfortunately Bleich and Veen are not alone. In this day and age of political correctness a lot of others are doing the same. And, Muslims are taking advantage of that. Please check the reference ref1 below. In this article a Muslim sounding scientist Abubakar Abid paints AI technology GPT-3 as biased against Muslims.

To establish the claim that a scientific method is biased against Muslims, we will have to show that it produces anti-Muslims info even when we feed it unbiased articles about Islam. To establish that the articles fed to GPT-3 were biased, we have to show that they were biased independently of the scientific method. These authors did neither.

If we buy the Bleich and Veen's methodology, I have a suggestion for scientists in Israel. They should feed the IDF related articles written by the Palestinian media to GPT-3. GPT-3 will soon learn that IDF does atrocities. This will kill two birds with one stone. First, Israel can then claim that IDF is as gentle as Florence Nightingale was. And, second, they can establish once and for all that Palestinian media is biased. This is exactly what Erik Bleich and Maurits van der Veen and Abubakar Abid have done. Unfortunately, IDF is too honest and too intelligent to do that.

ref1: Article suggesting GPT-3's bias against Islam: https://www.vox.com/future-perfect/22672414/ai-artificial-intelligence-gpt-3-bias-muslim

Dislike
Submitting....

Note: Opinions expressed in comments are those of the authors alone and not necessarily those of Daniel Pipes. Original writing only, please. Comments are screened and in some cases edited before posting. Reasoned disagreement is welcome but not comments that are scurrilous, off-topic, commercial, disparaging religions, or otherwise inappropriate. For complete regulations, see the "Guidelines for Reader Comments".

Submit a comment on this item

Reader comments (13) on this item

Title Commenter Date Thread
Computational topic modeling, and human bias [128 words]PeterJul 19, 2022 16:20283942
1For all you know, Erik Bleich and Maurits van der Veen are right [262 words]PrashantJun 9, 2022 16:19282450
1If you do not like science, just change it [504 words]PrashantJun 13, 2022 16:46282450
1How to get published and be popular? [464 words]
w/response from Daniel Pipes
PrashantJun 5, 2022 11:56282359
1Does Said's own life debunk his thesis? [432 words]PrashantJun 6, 2022 11:23282359
1Your arguments hold entirely [133 words]JeffJun 7, 2022 21:16282359
1I don't oppose Islam or Muslims; I oppose hypocrisy and selfishness [135 words]PrashantJun 8, 2022 17:13282359
2The dangerous trend today is from... [41 words]GinaJun 8, 2022 19:22282359
1Fully agreed with Gina [63 words]PrashantJun 9, 2022 13:02282359
Just wait [22 words]JeffJun 4, 2022 14:25282344
1More and more teachers and journalists are propageting fake news even in democratic countries. [29 words]Amos ZotJun 4, 2022 04:36282340
What news from the East? [140 words]John HarveyJun 2, 2022 12:23282313
Lack of Comparison with Counterintuative Belief System Defeats Analysis [310 words]M ToveyJun 6, 2022 16:51282313

Follow Daniel Pipes

Facebook   Twitter   RSS   Join Mailing List

All materials by Daniel Pipes on this site: © 1968-2024 Daniel Pipes. daniel.pipes@gmail.com and @DanielPipes

Support Daniel Pipes' work with a tax-deductible donation to the Middle East Forum.Daniel J. Pipes

(The MEF is a publicly supported, nonprofit organization under section 501(c)3 of the Internal Revenue Code.

Contributions are tax deductible to the full extent allowed by law. Tax-ID 23-774-9796, approved Apr. 27, 1998.

For more information, view our IRS letter of determination.)