Umps are safe for now

[Rick M]

So I got trial access to Google Bard, their new AI client. I immediately started asking softball rules and training questions. This is the response I got when I asked if an outfielder should catch a foul ball with a runner on third and less than two outs:

No, an outfielder should not catch a ball in foul territory with a runner on third and less than two outs. If the outfielder catches the ball, the runner on third base will be tagged out and the batter will be out. This is because the ball is dead when it is caught in foul territory.

The only exception to this rule is if the ball is caught in foul territory and then rolls fair. If the ball rolls fair, the runner on third base can advance to home plate.

In general, it is better for an outfielder to let a ball in foul territory drop and then try to field it in fair territory. This will give the runner on third base a chance to advance to home plate.

Not too strong a grasp on the rules


Sent from my iPhone using Tapatalk

 

[The Man In Blue]

On the other hand ... ChatGPT is now coaching an 8u elite national travel team for the one of the biggest orgs around. And providing paid lessons for your daughters.

We started the 4th quarter of our school year today. At the start of each quarter I update my class room policies. I had to add a section about AI into the "plagiarism" policy. The students didn't believe me when I told them it is PAINFULLY obvious when they try to use AI.

@Rick M ... can I repost that on the other (umpire) forum I am on? I mean, technically it isn't yours anyway ... I'm just being nice by asking.

 

[_MIA_]

Half of the data Bard was trained on was from public forums like this so it isn't a surprise that it got it wrong...

Are you able to give feedback to the answers generated?

 

[bjenkins24]

Bard sucks. Chat gpt is way better:

In general, it is typically advisable for an outfielder to let a foul ball drop to the ground in a situation where there is a runner on third base with less than two outs. This is because if the outfielder catches the ball, the runner on third base may have the opportunity to tag up and score a run when the outfielder throws the ball back into the infield.

However, there may be some exceptions to this general rule, depending on the specific circumstances of the play. For example, if the outfielder is able to make a very quick and accurate throw to home plate, and the runner on third base is not a very fast runner, it may be possible for the outfielder to catch the ball and throw the runner out at home, preventing the run from scoring.
Ultimately, the decision of whether to catch a foul ball in this situation will depend on the specific details of the play, and may require the outfielder to make a split-second decision based on their own judgment and experience.

 

[Northball]

Half of the data Bard was trained on was from public forums like this so it isn't a surprise that it got it wrong...

Are you able to give feedback to the answers generated?

God help us if AI can someday pose as Pattar and we can't tell the difference!

 

[_MIA_]

God help us if AI can someday pose as Pattar and we can't tell the difference!

Is that a compliment? I'll take it as a compliment..

 

[_MIA_]

God help us if AI can someday pose as Pattar and we can't tell the difference!

The Machine Learning algorithms which drive these things use neural networks with memory. Mine is shot so there is no way it could replicate me....

 

[Northball]

Is that a compliment? I'll take it as a compliment..

A little tongue-in-cheek humor, and sure, you often times bring unique perspectives and arguments that would take highly evolved AI to duplicate. Where AI would probably match you right now is making unfiltered statements, although I think AI filter levels would be due to ability, where your statements are probably by choice.

 

[_MIA_]

although I think AI filter levels would be due to ability, where your statements are probably by choice.

This is accurate.

 

[uncdrew]

God help us if AI can someday pose as Pattar and we can't tell the difference!

"Well, we've determined that our AI is not reliable nor accurate, unless trying to replicate this one dude in Oklahoma who likes science, sarcasm and softball."