india (@india@mastodonapp.uk)
#India has launched a #pilotprogramme allowing #consumers to #shop and #pay directly through #AIchatbots, starting with #OpenAI’s #ChatGPT and integrations with #Google’s #Gemini and #Anthropic’s #Claude in development. The pilot, powered by the #NationalPaymentsCorporation of India’s #UPI #ReservePay and #UPICircle protocols, aims to deepen engagement with India’s rapidly expanding #digitaleconomy. https://techcrunch.com/2025/10/09/india-pilots-ai-chatbot-led-e-commerce-with-chatgpt-gemini-claude-in-the-mix/?eicker.news #India #Tech #IndiaTech #Technews
Show Original Post
stacksize (@stacksize@mastodon.social)
🚨 New research reveals something surprising:
ChatGPT actually answers more accurately when prompted rudely—beating polite prompts in head-to-head tests.
Penn State researchers found accuracy rose from 80.8% (very polite) to 84.8% (very rude). Tone matters.
https://dropletdrift.com/chatgpt-responds-better-to-harsh-prompts-than-respectful-ones-study-finds/
#AI #LLM #ChatGPT #ArtificialIntelligence #MachineLearning #DataScience #TechNews #AIResearch #Innovation #PromptEngineering #Research #Ethics #AIFuture #GenerativeAI #AITrends #Science #Technology #OpenAI
Show Original Post
nando161 (@nando161@partyon.xyz)
"I asked #ChatGPT" well I asked Hermaeus Mora and that's why I'm trapped in Apocrypha for all eternity
Show Original Post
nando161 (@nando161@partyon.xyz)
"I asked #ChatGPT" well I asked M'aiq and he knows much, tells some. M'aiq knows many things others do not
Show Original Post
sajal24x7 (@sajal24x7@mastodon.social)
ChatGPT OSS runs on phone with 16GB memory
We are getting closer to the dream I have of on-device smart enough AI. This means two things - my next Mac needs to have maybe 32GB memory. And Apple needs to add more than 16GB to the iPhone Pros. I know they need to manage power - RAM eats power. Or, we get an even better model which runs on even less memory. Less likely though.
📖 Read more: https://sajalchoudhary.net/micro/chatgpt-oss-runs-on-phone-with-16gb-memory/
Show Original Post
jos1264 (@jos1264@social.skynetcloud.site)
Your LLM Won’t Stop Lying Any Time Soon https://hackaday.com/2025/10/10/your-llm-wont-stop-lying-any-time-soon/ #ArtificialIntelligence #machinehallucination #ChatGPT #LLM #ai
Show Original Post
RoundSparrow (@RoundSparrow@autistics.life)
#OutGroupDumb #TheELIZAeffect #FWakeELIZA #FWakeDumb
The SMARTEST person you know, EVERYONE YOU KNOW is being fed egoism / egomania. #EgomaniaOpera /\ #FWakeChatGPT #RFP19 #ChatGPT
"The ELIZA effect" was described in 1974 by MIT university, people who believe computer rewards. Video games, chat bots, etc. Screen reality.
https://old.reddit.com/r/memes/comments/1o35nyc/unfortunately_this_is_true/
Show Original Post
itnewsbot (@itnewsbot@schleuss.online)
Your LLM Won’t Stop Lying Any Time Soon - Researchers call it “hallucination”; you might more accurately refer to it as conf... - https://hackaday.com/2025/10/10/your-llm-wont-stop-lying-any-time-soon/ #artificialintelligence #machinehallucination #chatgpt #llm #ai
Show Original Post
crow (@crow@cville.online)
Can #Sora 2 or #ChatGPT draw graphs of their own respective carbon footprints?
Show Original Post
cesarpose (@cesarpose@infosec.exchange)
They're going to bring that shit to my country. It barely produces enough energy for normal purposes, but their dirty business dealings are more important.
#ai #argentina #openai #chatgpt
@slashdot https://mastodon.cloud/@slashdot/115352444151897134
Show Original Post
