Hacker News: tana_shahhhttps://news.ycombinator.com/user?id=tana_shahhHacker News RSShttps://hnrss.org/hnrss v2.1.1Sun, 24 May 2026 20:08:06 +0000<![CDATA[New comment by tana_shahh in "SpaceX launches Starship v3 rocket"]]>Thanks for the summary It was a good read and you nailed it!

]]>Sun, 24 May 2026 16:46:09 +0000https://news.ycombinator.com/item?id=48258846tana_shahhhttps://news.ycombinator.com/item?id=48258846https://news.ycombinator.com/item?id=48258846<![CDATA[New comment by tana_shahh in "Anthropic blames dystopian sci-fi for training AI models to act "evil""]]>magine an outer-space AI civilization invading Earth and seeing humans use AI this way.

]]>Sun, 24 May 2026 14:08:48 +0000https://news.ycombinator.com/item?id=48257402tana_shahhhttps://news.ycombinator.com/item?id=48257402https://news.ycombinator.com/item?id=48257402<![CDATA[New comment by tana_shahh in "OpenAI Is Preparing to File for an IPO Soon"]]>We got a non-profit filing for IPO before GTA-6!

]]>Sun, 24 May 2026 14:05:48 +0000https://news.ycombinator.com/item?id=48257382tana_shahhhttps://news.ycombinator.com/item?id=48257382https://news.ycombinator.com/item?id=48257382<![CDATA[New comment by tana_shahh in "Local LLMs perform better when you teach them to ask before they answer"]]>Absolutely True not only for Local LLMs but for cloud ones too. Clarifying the intention, the type of output we want improves the model's response multiple folds.

]]>Sun, 24 May 2026 13:58:26 +0000https://news.ycombinator.com/item?id=48257330tana_shahhhttps://news.ycombinator.com/item?id=48257330https://news.ycombinator.com/item?id=48257330