Ghazi @curiousmind@qoto.org
Tunisian 🇹🇳 Pharmacist 💊 Romanista 🐺
I'll toot about whatever
Joined Nov 2022
@surveillancepod I use Nunti as a RSS reader on Android. It has a cool feature called Adaptive learning, which allows the user to rate the articles and then the algorithm (dirty word, I know) sorts the articles based on his preferences.
You can find a detailed explanation on how the algorithm works in the link below :
https://gitlab.com/ondrejfoltyn/nunti/-/issues/28
Ghazi
boosted
Ghazi
boosted
RT @___z__
Il y a une semaine, c'était une caricature.
Aujourd'hui, c'est une réalité
#Saïed #raciste #fascisme #SOS #Tunisie
(Dessin immortalisé sur la Blockchain ici : https://opensea.io/assets/ethereum/0x495f947276749ce646f68ac8c248420045cb7b5e/78780522338020336869644616041657157587801539095442514548727814296582848774145/)
Ghazi
boosted
Dire que ce pays a donné son nom à l'Afrique.
Ghazi
boosted
RT @SKhmira
"There is a criminal plan set up since the beginning of the century to "Alter the demographic structure of Tunisia"...the flow of irregular sub saharan African migrants aims, discretely, to strip Tunisia of its Arab and Muslims identity..."
#Tunisia
@hszakher I'm extremely ashamed by the state of my country. I never expected him to stoop so low...
Funny how he doesn't mention this conspiracy theory in his #Twitter post.
Wow! Just wow!! #Tunisian president #Kais_Saied just published an extremely #racist #Facebook post, promoting the #Great_Replacement conspiracy theory !!! Extremely worrying
This recent wave of #racism against #subsaharans in #Tunisia is yet another example on how #social_media companies (mainly #Facebook) are enabling a minority of extreme right wing fascists to spread their hatred among the population. #Hate_speech is being amplified by the #algorithm and content moderation is inexistent. #Meta doesn't even care about the harm they're causing outside of Western countries.
Ghazi
boosted
Ghazi
boosted
My brother in JFC...automated AI medical advice (via systems that are almost certainly collecting/creating invasive health surveillance data on behalf of insurers/employers) is still not helpful (even if it were accurate, which it's not) to people who can't afford to follow it
@httpjames VOIP providers in countries other than the US and Canada
Ghazi
boosted
Here's a first... I just got an email because ChatGPT suggested an article I wrote to somebody. Could I send them a copy? Except, I never wrote the article, it doesn't exist. PLEASE realize right now that this tool isn't pulling out cool references for you. It's making plausible titles and matching them to authors names.
Ghazi
boosted
And this reminds me that I really wish we had classes & curriculum in basic education that gave a tour of flaws & limitations in our brains & sensory apparatuses.
Teach folks about optical illusions, sleight of hand, close-up magic, statistical reasoning, cognitive bias, logical fallacies - a survey of the many ways things might not be what they seem and show the utility of doubt & skepticism & verification. I feel like this stuff is really hard for folks in general
Ghazi
boosted
As AI tools for writing become more common, let me throw one more worry into the mix: that students who write well without AI assistance may be falsely accused of #plagiarism by teachers using imperfect tools to detect AI-assisted writing.
Ghazi
boosted
"After years of #FDA issuing guidance documents and launching programs to encourage sponsors to seek broader representation in #clinicaltrials, policy makers have added a “stick” to the process: research sponsors now will be required to tackle this issue by submitting Diversity Action Plans (DAPs) to the agency when proposing and seeking approval of Phase III and pivotal studies."
https://www.pharmtech.com/view/clinical-trial-diversity-continues-to-face-challenges
Ghazi
boosted
نشرت شبكة "فوربيدن ستوريز" للصحافيين الاستقصائيين الأربعاء نتائج تحقيق أجرته حول نشاط شركة إسرائيلية أطلقت عليها اسم "خلية خورخي". وخلص الصحافيون إلى أن الشبكة متورطة في القضية التي استهدفت مقدم البرامج في شبكة "بي إف إم تي في" رشيد مباركي المتهم بالخضوع لتأثير خارجي. وقال التقرير إن الشركة تدير مشروعا ضخما للتضليل وتورطت بالتلاعب بالانتخابات في 33 بلدا ثلثاها في أفريقيا.
#OpenAI Chief CTO Mira Murati admitted "to one of #ChatGPT's clearest flaws: that the device is often wrong, confidently spitting out falsehoods without a hint of caution".
https://futurism.com/the-byte/cto-chatgpt-concerns-releasing-publicly
Tunisian 🇹🇳 Pharmacist 💊 Romanista 🐺
I'll toot about whatever
Joined Nov 2022
