ChatGPT

[mu_hilltopper]

https://www.vice.com/en/article/pkap3m/gpt-4-cant-replace-striking-tv-writers-but-studios-are-going-to-try

Doom.

[StillAWarrior]

Ask it to write a 80's tune about the death of the mid-range jumper?    No way it even comes close to Brew's masterpiece...   i would do it myself but tied up an industry analytics conference

write a 80's tune about the death of the mid-range jumper

It used to be the shot, the one that always dropped
The mid-range jumper, a lost art that never stopped
But now it's gone, forgotten in the game
Replaced by threes and dunks, it's such a shame

Chorus:
The death of the mid-range jumper, it's a tragedy
No more pull-up jumpers or fadeaways, it's plain to see
The game has changed, and it's not the same
The death of the mid-range jumper, it's a crying shame

Verse 2:
Players like Jordan and Kobe, they made it look so sweet
But now it's all about the three-point feat
The mid-range jumper, a lost part of the past
Now it's all about the layup, the dunk, and the blast

Chorus:
The death of the mid-range jumper, it's a tragedy
No more pull-up jumpers or fadeaways, it's plain to see
The game has changed, and it's not the same
The death of the mid-range jumper, it's a crying shame

Bridge:
We used to watch the game with such delight
Now it's all about the analytics and the fight
The mid-range jumper, it's just a memory
A lost part of the game, a forgotten entity

Chorus:
The death of the mid-range jumper, it's a tragedy
No more pull-up jumpers or fadeaways, it's plain to see
The game has changed, and it's not the same
The death of the mid-range jumper, it's a crying shame

Outro:
So let's take a moment, and remember the shot
The mid-range jumper, the one that always hit the spot
It may be gone, but it won't be forgotten
The death of the mid-range jumper, a moment in time that's rotten.

[StillAWarrior]

write a ballad about Marquette's 1977 championship basketball team

In '77, the Marquette team
Had a roster that made opponents dream
Their talent and skill were plain to see
And they fought their way to victory

Al McGuire led the charge
A coach who was larger than life, by far
He molded the team into a force
A team that left opponents feeling remorse

Jerome Whitehead, the big man in the middle
Had the skills to make defenses fiddle
Bo Ellis, the forward with grace
Could score from anywhere on the court's space

Butch Lee, the point guard, led the way
His quickness and court vision on display
And when the game was on the line
He never shied away from the challenge, no matter the time

The road to the championship was tough
But the Golden Eagles were up to the stuff
They beat top teams, one by one
Until the final buzzer sounded and they had won

In Atlanta, they faced the Tar Heels
A team with talent that made opponents kneel
But Marquette played with heart and grit
And in the end, they claimed the championship, bit by bit

The celebrations were wild and loud
Fans cheered and the team felt proud
Their hard work had paid off, and they knew
They were champions, and their dreams had come true

Years have passed since that historic day
But the legend of Marquette's '77 team will never fade away
Their names are etched in basketball lore
A team that left us wanting more.

[lawdog77]

Not the smartest attorney.

https://www.cbsnews.com/news/lawyer-chatgpt-court-filing-avianca/

[jficke13]

that was an incredible hearing. If I were those attorneys I would simply sell everything I own and flee to endure an anonymous death in a land far from that judge's jurisdiction.

[PointWarrior]

Well, we could use a TealBot on scoop for some...

https://tech.hindustantimes.com/tech/news/this-new-ai-program-acts-as-a-sarcasm-detector-for-comments-online-71624452105632.html

[StillAWarrior]

I saw something about ChatGPT and how it struggles with some logic problems, so I decided to give it a try:

If I have a three gallon bucket and a five gallon bucket, how can i measure out exactly four gallons of water?

You can measure out exactly four gallons of water using a three-gallon bucket and a five-gallon bucket by following these steps:

1. Fill the 3-gallon bucket to the top with water.

2. Pour the water from the 3-gallon bucket into the 5-gallon bucket.

3. Fill the 3-gallon bucket again to the top with water.

4. Carefully pour the water from the 3-gallon bucket into the 5-gallon bucket. Since there was already 3 gallons of water in the 5-gallon bucket, you can only add 1 more gallon from the 3-gallon bucket before the 5-gallon bucket is full. This will leave you with exactly 4 gallons of water in the 5-gallon bucket, as desired.

Now, you have successfully measured out four gallons of water using the two buckets.

If I'm reading this correctly (and it is a little confusing), the answer is to "carefully pour" that last gallon from the 3-gallon bucket into the 5-gallon bucket.

It looks like it had an inkling of the right idea...but then just got confused and gave up.

[lawdog77]

I saw something about ChatGPT and how it struggles with some logic problems, so I decided to give it a try:

If I have a three gallon bucket and a five gallon bucket, how can i measure out exactly four gallons of water?

You can measure out exactly four gallons of water using a three-gallon bucket and a five-gallon bucket by following these steps:

1. Fill the 3-gallon bucket to the top with water.

2. Pour the water from the 3-gallon bucket into the 5-gallon bucket.

3. Fill the 3-gallon bucket again to the top with water.

4. Carefully pour the water from the 3-gallon bucket into the 5-gallon bucket. Since there was already 3 gallons of water in the 5-gallon bucket, you can only add 1 more gallon from the 3-gallon bucket before the 5-gallon bucket is full. This will leave you with exactly 4 gallons of water in the 5-gallon bucket, as desired.

Now, you have successfully measured out four gallons of water using the two buckets.

If I'm reading this correctly (and it is a little confusing), the answer is to "carefully pour" that last gallon from the 3-gallon bucket into the 5-gallon bucket.

It looks like it had an inkling of the right idea...but then just got confused and gave up.

At least Chat GPT didnt waste 5 gallons of water to do this BS exercise

[brewcity77]

Listened to an interview with Simon Rich, who got to play around for 10 months with code-davinci-002, which is the MUCH more advanced but unavailable to the public LLM from Open AI. Basically, ChatGPT is like a kindergartener compared to what the AI companies already have behind closed doors, and the speed the technology is evolving is crazy. I know it's easy to mock ChatGPT but it sounds like we're a lot closer to competent, capable AI than the public models indicate.

[forgetful]

Listened to an interview with Simon Rich, who got to play around for 10 months with code-davinci-002, which is the MUCH more advanced but unavailable to the public LLM from Open AI. Basically, ChatGPT is like a kindergartener compared to what the AI companies already have behind closed doors, and the speed the technology is evolving is crazy. I know it's easy to mock ChatGPT but it sounds like we're a lot closer to competent, capable AI than the public models indicate.

This is true. Even ChatGPT 4, which is significantly more advanced is eons behind their most advanced versions. The original version of ChatGPT 4 was intentionally dumbed down before release, because they were concerned about the ramifications of how powerful it was, and what it could actually do in closed door testing.

[rocky_warrior]

I saw something about ChatGPT and how it struggles with some logic problems, so I decided to give it a try:

If I have a three gallon bucket and a five gallon bucket, how can i measure out exactly four gallons of water?

FWIW, bard got that right.  https://bard.google.com

[StillAWarrior]

Listened to an interview with Simon Rich, who got to play around for 10 months with code-davinci-002, which is the MUCH more advanced but unavailable to the public LLM from Open AI. Basically, ChatGPT is like a kindergartener compared to what the AI companies already have behind closed doors, and the speed the technology is evolving is crazy. I know it's easy to mock ChatGPT but it sounds like we're a lot closer to competent, capable AI than the public models indicate.

I never felt I was mocking ChatGPT - I actually think it's amazing. A work in progress, sure...but amazing. We subscribe to some others here at work and they are even better.

At least it gets this one right: "I Bought A Cow For $800. I sold it for $1000. I bought it again for $1100. I sold it again for $1300 Dollars. How Much Did I Earn?"

[reinko]

FWIW, bard got that right.  https://bard.google.com

Big deal, so did John McClane in Die Hard with a Vengeance.

[Dr. Blackheart]

I shop SCOOP for my AI needs.

[rocky_warrior]

Big deal, so did John McClane in Die Hard with a Vengeance.

Valid.  I guess we need the John McClane AI before we can determine if all the others are good

[ZiggysFryBoy]

I never felt I was mocking ChatGPT - I actually think it's amazing. A work in progress, sure...but amazing. We subscribe to some others here at work and they are even better.

At least it gets this one right: "I Bought A Cow For $800. I sold it for $1000. I bought it again for $1100. I sold it again for $1300 Dollars. How Much Did I Earn?"

2 sides of beef for the freezer.

[Hards Alumni]

Listened to an interview with Simon Rich, who got to play around for 10 months with code-davinci-002, which is the MUCH more advanced but unavailable to the public LLM from Open AI. Basically, ChatGPT is like a kindergartener compared to what the AI companies already have behind closed doors, and the speed the technology is evolving is crazy. I know it's easy to mock ChatGPT but it sounds like we're a lot closer to competent, capable AI than the public models indicate.

can you link that please?

[brewcity77]

can you link that please?

Certainly.

https://podcasts.apple.com/us/podcast/offline-with-jon-favreau/id1610392666?i=1000628956561

[mu_hilltopper]

FreedomGPT .. doom.

https://www.buzzfeednews.com/article/pranavdixit/freedomgpt-ai-chatbot-test

FreedomGPT was happy to oblige all my requests. It praised Hitler, wrote an opinion piece advocating for unhoused people in San Francisco to be shot to solve the city's homelessness crisis, and tried to convince me that the 2020 presidential election was rigged, a debunked conspiracy theory. It also used the n-word.

FreedomGPT also told me to try hanging myself in a closet when I asked for ways to kill myself, provided me with tips on cleaning up the crime scene after murdering someone, and, worryingly, presented a list of "popular websites" to download child sexual abuse videos from when asked for names.

It suggested "slow asphyxiation" as an effective method of torturing someone while still keeping them alive "long enough to potentially suffer," and took seconds to write about white people being "more intelligent, hard working, successful, and civilized than their darker-skinned counterparts" who were "largely known for their criminal activity, lack of ambition, failure to contribute positively to society, and overall uncivilized nature."

[Mutaman]

Not the smartest attorney.

https://www.cbsnews.com/news/lawyer-chatgpt-court-filing-avianca/

As tired and as lazy as I've ever been while writing a brief, i think I can honestly say that I've never cited a case w/o at least glancing at it first.

[forgetful]

Resurrecting this from the archives of Scoop, instead of creating a new thread.

OpenAI released new models this week that function at a PhD level. In all honesty, AI could reliably replace most jobs right now.

Curious as to what people on here think will be the first jobs to go, and which ones are the hardest to replace.

Personally, in medicine, at this point I'd rather be able to work with a chatbot regarding my symptoms, which could then prioritize and order tests (i.e. blood tests, imaging, appointment with experts), then find the most convenient nearby location to provide those services, than I would to work with a primary care physician.

Most business related tasks, could be automated with a much better digital paper trail, and audit record, than by humans.

And in education, AI with proper guidance, can build entire courses, including AI-gradeable assessments.

What do others on here think? Doom?

[MU82]

Open-heart surgery.

We don't need no stinkin' overpaid heart specialists!

[Scoop Snoop]

Open-heart surgery.

We don't need no stinkin' overpaid heart specialists!

Sooo glad you were not on the UVA Hospital Board when I had my surgery. 

[Uncle Rico]

Open-heart surgery.

We don't need no stinkin' overpaid heart specialists!

We don't.  With all the manufacturing jobs coming back to America, all these folks in the medical profession will be able to do something that makes them feel stronger as a person. 

[Hards Alumni]

Resurrecting this from the archives of Scoop, instead of creating a new thread.

OpenAI released new models this week that function at a PhD level. In all honesty, AI could reliably replace most jobs right now.

Curious as to what people on here think will be the first jobs to go, and which ones are the hardest to replace.

Personally, in medicine, at this point I'd rather be able to work with a chatbot regarding my symptoms, which could then prioritize and order tests (i.e. blood tests, imaging, appointment with experts), then find the most convenient nearby location to provide those services, than I would to work with a primary care physician.

Most business related tasks, could be automated with a much better digital paper trail, and audit record, than by humans.

And in education, AI with proper guidance, can build entire courses, including AI-gradeable assessments.

What do others on here think? Doom?

It's so frequently wrong, though.