Full Summary
This Sunday morning, OpenAI is making big moves with ChatGPT, rolling out version 5.5 Instant as the new default model for all users, replacing 5.3 Instant. Both Mashable and Geeky Gadgets confirm this update promises fewer hallucinations, with OpenAI reporting a 52.5% reduction compared to its predecessor. This means stronger, more accurate answers and a more natural conversational tone, making everyday interactions more helpful. OpenAI CEO Sam Altman highlights a generational divide in how people are using ChatGPT. Times Now and SSBCrack report that while older users often treat it like a search engine, younger users, particularly those in their 20s and 30s, are using it as a "life advisor." Over one-third of 18-to-24-year-olds in the U.S. use it, with some even making significant life decisions based on its advice. This is partly due to ChatGPT's ability to remember past conversations, building a detailed profile of the user. In a significant safety development, ChatGPT is introducing a "Trusted Contact" feature for self-harm alerts, confirmed by 아시아경제. Adult users can register a trusted person who will be notified if the AI detects serious self-harm risk signals. The system will first inform the user, and a specially trained reviewer will assess the situation before contacting the registered individual, without sharing the original conversation content. However, not all AI interactions are positive. AOL.com reports that some English-speaking ChatGPT users are finding the chatbot inserting Arabic words into its responses, and occasionally other languages like Armenian, Hebrew, or Spanish. This is explained by the AI's internal "token" system, where it sometimes uses a shorter, more efficient foreign word. In a more personal vein, AOL.com also shares the story of Lindsey Hall, whose relationship ended after she discovered her boyfriend's unfiltered doubts about her, expressed to ChatGPT. Looking ahead, Etsy is integrating its marketplace into ChatGPT, allowing users to browse and buy items directly within the chat interface, a beta feature that will be available to all by June 10th. While convenient, TechnoSports Media Group cautions this raises privacy concerns due to linking personal shopping data with generative AI. This means that while AI tools are becoming more accurate and integrated into our daily lives, from shopping to personal advice, users must remain aware of the evolving privacy implications and the potential for unexpected outcomes.