Hah! Idiot. Well, I guess we don’t have to worry about Google Bard!
At least it said it with confidence.
I look forward to AI official scorers.
“It must be a human error, Dave.”
The scorer last night could have used an infusion of intelligence, artificial or otherwise.
IBM’s Watson used a formula to calculate its confidence levels whenever it came up with a Jeopardy answer. When it actually competed on Jeopardy against Ken Jennings and Brad Rutter, you could see its potential answers, selected answer, and confidence level on the screen in real time, as if you had a window into its thought process. Watson had a decent but not impenetrable lead going into Final Jeopardy, made a terrible guess that elicited gasps from the audience, but won anyway - since its confidence level was low it wagered almost nothing.
The confidence indicator is one thing that is sorely missing from these current chat bots and LLMs. As Jim used to say, they are often wrong but never in doubt.
Of course, Fox will jigger theirs so that it always reports 100% confidence.
You’re probably one letter off at Fox.
Ouch.
Damn that’s scary.
Eeesh.
Hope he came out ahead.
It’s a f’ing conspiracy!
That’s awesome. I didn’t know that rule existed. I wish more scorers would be so bold.
Tampa is en fuego.
Starting in the 7th inning they should raise the price of beers $1 each inning, or maybe $.50 each half inning.
Yankees putting on a disaster clinic, giving up 9 runs to the Twins in the top of the 1st. Very satisfying.
Must be windy there.