Recent AI model progress feels mostly like bullshit

Re:Recent AI model progress feels mostly like bullshit by drooby on Sunday April 06, 2025 @03:34PM (#43604240)

I asked GPT-4.5 and it searched the web and immediately gave me a "yes" with paragraphs of sources cited.

[ Parent | on HN ]

Re:Recent AI model progress feels mostly like bullshit by john2x on Sunday April 06, 2025 @03:44PM (#43604324)

Truth is a probability game. Just keep trying until you arrive.

[ Parent | on HN ]

Re:Recent AI model progress feels mostly like bullshit by Avicebron on Sunday April 06, 2025 @04:28PM (#43604693)

"man puts prompt into LLM" > "LLM ships bad code" > "machine fails" > "person dies" > "fire man" > "man puts prompt into LLM"

[ Parent | on HN ]

Re:Recent AI model progress feels mostly like bullshit by john2x on Monday April 07, 2025 @02:47AM (#43608531)

Such accountability

[ Parent | on HN ]

Re:Recent AI model progress feels mostly like bullshit by boznz on Sunday April 06, 2025 @07:08PM (#43605746)

3-4 hours is enough time for It to have crawled the hacker news comments section. That's about the frequency the AI bots crawl my little out of the way blog.

[ Parent | on HN ]