The prompt we use to get the news, from which we make the prediction, doesn't ask about oil or gas prices so the info AI has to make the estimation doesn't take that into account. The oil/gas prices are pulled from separate free APIs (Yahoo Finance) and overlaid, as kind of a sanity-check to see how good the analysis actually was.
Thought it would be fun to see how accurate AI was at analyzing this, given its volatility. I wanted to see if the graph would "converge" like the ones showing the chance of a sports team winning over the course of the game - they usually bounce around before trending one direction at the end of the game.
Every morning a small Lambda checks Google News (war/peace/ceasefire/Hormuz queries), gives the headlines to Claude Haiku, and gets back a probability (0-100%) plus ~500 words (Claude is so generous, need to constrain that) on what moved and why. It's a static site on S3 + CloudFront, and we keep a daily history of it
Anyway the graph was all over the place. So to help gauge its accuracy yesterday I added oil and gas prices overlaid. Turns out they correlate pretty closely *most* of the time. If a deal is reached I'll be interested to see if the graph converges up to that point (showing AI was somewhat smart at predicting), or if it just jumps to 100 in which case it was less... predictive.
Take a look, happy to hear feedback or suggestions.
That is super cool - if passing a logo, do you recommend a simple black and white image? I tried some emojis (basically yellow and black) but they didn't show up as well as your default settings.
I don't think older people lack creativity, but I think a lot of the creativity drop is due to starting families and prioritizing that over work as people age. Guess we'll find out when Spaceballs 2 drops though :)
For an example of a really horrible audio "upgrade", check out the Ninja Gaiden Trilogy (remake) on the SNES where they replaced the well-crafted NES instruments with SNES samples just triggered for the same notes and beats. Theoretically could have sounded amazing but is like nails on a chalkboard. Not apples to apples comparison with these SNES->SNES restorations of course.
Users will learn to be more efficient when the prices go up too. We love to casually chat with AI whether we're coding, researching, etc. But we'll tighten up these conversations, just like when you're talking to a lawyer you generally prep a little more before the meeting so you don't incur extra billing at $300/hour
Wonder if this could be extended to a model in a physical simulation of a fruit fly flying. Even if it took a minute to process and simulate every "frame" (30fps) it would be cool to have a minute or so of simulation "footage" after a day