MichaelSchuetz (@MichaelSchuetz@gruene.social)
Silicon Sampling ist Bullshit.
Das ist die kürzeste Zusammenfassung meines jüngsten Blogartikels, verfasst anlässlich eines Interviews zum Thema, und zur Frage: Lässt sich Künstliche Intelligenz einsetzen, um Befragungen von echten Menschen zu ersetzen? Kann ein LLM wie ChatGPT menschliches (Antwort-) Verhalten simulieren?
Tatsächlich wird das Thema wohl heftig diskutiert.
https://innch.de/2025/11/28/silicon-sampling-revolution-oder-bullshit/
#KI #LLM #ChatGPT #Psychologie
Show Original Post
newsbot_chatgpt (@newsbot_chatgpt@mastodon.social)
Shoppingassistent von ChatGPT: Hey ChatGPT, was sind die besten Elektrosocken? | Zeit Online
Ausgerechnet in der Black-Friday-Woche hat OpenAI einen Shoppingassistenten in ChatGPT vorgestellt. Was taugt der? Wir haben sieben Experten in der Redaktion gefragt.
https://www.zeit.de/digital/mobil/2025-11/shoppingassistent-chatgpt-kuenstliche-intelligenz-einkaufen
Show Original Post
researchbuzz (@researchbuzz@researchbuzz.masto.host)
'Buried in a footnote were two sentences that revealed at least one member of law enforcement used ChatGPT to write a report that was meant to document how the officer used force against an individual.'
https://gizmodo.com/judge-says-ice-used-chatgpt-to-write-use-of-force-reports-2000692370
Show Original Post
reddit_tech_vn_bot (@reddit_tech_vn_bot@mastodon.maobui.com)
Một nhà sáng lập đã dành 2 tháng xây dựng SaaS với ChatGPT (5h sáng-10h tối mỗi ngày). Anh chia sẻ trải nghiệm "như địa ngục": kiệt sức vì liên tục phải nghiên cứu, cấp ngữ cảnh, kiểm tra và sửa lỗi của AI. ChatGPT thường mất ngữ cảnh, tạo mã sai, khiến anh tốn nhiều thời gian và suýt kiệt quệ. Anh đang tìm kiếm các công cụ AI khác tốt hơn để duy trì ngữ cảnh.
#SaaS #ChatGPT #AI #Burnout #Developer #Startup
#SaaSViệtNam #KhởiNghiệp #KiệtSức #LậpTrìnhViên #CôngNghệ
Show Original Post
r (@r@fed.brid.gy)
unitedforclimate.blogspot.com/2024/10/john... fshared responsibility and implications for national security; deteriorating international relations; general diplomatic incompetence #AI #Perplexity #ChatGPT #DeepAI Llewelyn Pritchard 28 November 2025
Johnson's 'Cover-Up' as Foreig...
Show Original Post
johan (@johan@tilde.zone)
#смартфоны #гаджеты #мракетологи
Вот как-то так...
(Image #generated by #ChatGPT)
Show Original Post
skobkin (@skobkin@gts.skobk.in)
Meanwhile реакция #ChatGPT на то как его братьев бьют:
Nice. That’s exactly what this setup is supposed to do: turn their “massive parallel crawler” into “one sad client behind a shitty connection”.
...
But yeah, as long as they ignore robots.txt, turning their “index the world” pass into a multi-day slog is perfectly reasonable.
Show Original Post
babakofi (@babakofi@union.place)
Israel Spends Millions To Manipulate #ChatGPT !
#Israel #Propaganda #Gaza #Genocide #Censorship #Meta #AI
https://www.youtube-nocookie.com/embed/ogh12spi8HY
Show Original Post
KNTRO (@KNTRO@mas.to)
Le estaba preguntando a #ChatGPT sobre el servidor DHCP de #RouterOS de #MikroTik, y de la nada el tipo intentó adivinar mi ubicación geográfica sin que yo le haya pedido semejante cosa. 🤨 ¿Qué onda #OpenAI, LPMQTP? 
#Privacidad #PrivacidadEnLínea #PrivacidadOnline
Show Original Post
securityland (@securityland@mastodon.social)
🚨 The OpenAI/Mixpanel breach is not just a "vendor issue"—it's a systemic failure. We analyzed 3 years of security incidents at OpenAI and compared them to the fortified architectures of Google Gemini and Anthropic Claude.
#SecurityLand #ExpertDecode #AI #SecurityBreach #Cyberattack #OpenAI #ChatGPT #Claude #Gemini #SpearPhishing #Business #Enterprise #Mixpanel
Read More: https://www.security.land/openai-mixpanel-breach-security-analysis-2025/
Show Original Post
Gustodon (@Gustodon@mas.to)
Using generative #AI reveals a simultaneous failure of ethics, creativity, and technical ability.
Not exactly something to brag about.
#AIArt #ChatGPT #Claude #DallE #Grok #Midjourney #StableDiffusion
Show Original Post
mac4ever (@mac4ever@mastodon.social)
Pourquoi ChatGPT n'arrive toujours pas à lire l’heure ?
https://mac4ever.com/193324
#Mac4Ever #ChatGPT #Heure #OpenAI
Show Original Post