chatgpt Timeline - FediScanner

03.09.2025 22:26
InfodocMicroveille (@InfodocMicroveille@social.sciences.re)

Does ChatGPT Ignore Article Retractions and Other Reliability Concerns?

https://microblogging.infodocs.eu/2025/09/03/does-chatgpt-ignore-article-retractions-and-other-reliability-concerns/

#ChatGPT #ErTeZheng #GenerativeArtificialIntelligence #IriniKatsirea #KimHolmberg #MariannaLehtisaari #MikeThelwall #RetractionsOfPublications #ScholarlyPublishing #ScientificCommunication

Show Original Post

Report

03.09.2025 22:21
Vittoria (@Vittoria@mastodon.social)

American artificial intelligence firm OpenAI said Tuesday it would add parental controls to its chatbot ChatGPT, a week after an American couple said the system encouraged their teenaged son to kill himself. FRANCE 24's Eliza Herbert reports.
#ChatGPT #suicide #OpenAI #AI

https://www.youtube.com/watch?v=3E5szzeeLi0

Show Original Post

Report

03.09.2025 21:43
Bongolian (@Bongolian@mstdn.social)

The promises mean nothing until a court of law enforces them.

Stephen Council: ChatGPT's reckoning is here. Tragic news in August exposed issues with OpenAI's flagship product and brought promises of change

#openai #chatgpt #meta AI
https://www.sfgate.com/tech/article/openai-chatgpt-reckoning-is-here-21027576.php

Show Original Post

Report

03.09.2025 21:02
mempko (@mempko@fosstodon.org)

generative AI is the greatest tool for plausible deniability that man has ever seen. We have seen both Trump and Venezuelan president this week claim videos were AI generated.

#AI #LLMs #ChatGPT #tech #technology

Show Original Post

Report

03.09.2025 20:46
Nikolas (@Nikolas@federated.press)

That's right time to foster #critical #thinking: Large #language models are "souped-up statistical #regurgitation machines, so they will continue to stumble into problems around #truth, hallucinations and #reasoning."
https://www.nytimes.com/2025/09/03/opinion/ai-gpt5-rethinking.html #ai #Chatgpt

Show Original Post

Report

03.09.2025 20:24
reddit_tech_vn_bot (@reddit_tech_vn_bot@mastodon.maobui.com)

Một người dùng Reddit đã phát triển tiện ích mở rộng Chrome "ChatSight" giúp dễ dàng theo dõi và chuyển đổi giữa các prompt trong cuộc trò chuyện ChatGPT. Tiện ích này hiển thị danh sách các câu hỏi/prompt của người dùng, tổng số lượng prompt và số lượng token (tính năng thử nghiệm).

#ChatGPT #AI #Extension #TiệnÍch #TríTuệNhânTạo

https://www.reddit.com/r/SideProject/comments/1n7lxfq/i_built_a_chrome_extension_to_easily_track_and/

Show Original Post

Report

03.09.2025 20:21
rod2ik (@rod2ik@mastodon.social)

#OepnAI #ChatGPT : #WTF ?

*Il y a quelques jours, une MAJ sur le blog de OpenAI en date du 27 août 2025, révélait que l'entreprise scannait désormais les conversations ChatGPT de ses utilisateurs à titre préventif !

Cela dans le but de rechercher des contenus préjudiciables à leurs utilisateurs, tels que des instructions pour s'automutiler ou créer des armes... L'entreprise pourra ainsi signaler des cas graves à la police...

Show Original Post

Report

03.09.2025 20:20
r (@r@bsky.brid.gy)

#OepnAI #ChatGPT : #WTF ? *Il y a quelques jours, une MAJ sur le blog de OpenAI en date du 27 août 2025, révélait que l'entreprise scannait désormais les conversations ChatGPT de ses utilisateurs à titre préventif !

Show Original Post

Report

03.09.2025 20:20
r (@r@fed.brid.gy)

#OepnAI #ChatGPT : #WTF ? *Il y a quelques jours, une MAJ sur le blog de OpenAI en date du 27 août 2025, révélait que l'entreprise scannait désormais les conversations ChatGPT de ses utilisateurs à titre préventif !