ChatGPT (1 Viewer)

[CletiusMaximus]

Remarkable is not how I would describe it.

We haven’t even begun to see the damages technology will have on us. Imo, It’s a cancer and what’s making it worse is that we’re are helping it along.

In a world where 90% of the population can be manipulated and controlled by headlines and propaganda, I don’t see how we come out of this.

It’s cliche and straight out of terminator but once these machines don’t need us we’re done.

Yeah I tend to have this view. Don’t know if it’s just a sign I’m getting old but it is hard to see how this has a net positive effect on the world. It’s probably great for the <1% who control technology and are already super wealthy but for the masses, it seems they are going to be left farther and farther behind.

The deep fake (videos) we are going to see in our lifetime will likely cause some very dangerous problems.

I think it is because you’re getting old(er). When you were younger you likely would have thought this was an amazing idea. As you age and wisen up you figure out that life isn’t about computers or the rat race. It’s about living your life and experiencing it with people.

We need to try and get back to a life without technology having so much of a footprint.

good one I saw yesterday

 

[chet]

Remarkable is not how I would describe it.

We haven’t even begun to see the damages technology will have on us. Imo, It’s a cancer and what’s making it worse is that we’re are helping it along.

In a world where 90% of the population can be manipulated and controlled by headlines and propaganda, I don’t see how we come out of this.

It’s cliche and straight out of terminator but once these machines don’t need us we’re done.

Yeah I tend to have this view. Don’t know if it’s just a sign I’m getting old but it is hard to see how this has a net positive effect on the world. It’s probably great for the <1% who control technology and are already super wealthy but for the masses, it seems they are going to be left farther and farther behind.

The deep fake (videos) we are going to see in our lifetime will likely cause some very dangerous problems.

I think it is because you’re getting old(er). When you were younger you likely would have thought this was an amazing idea. As you age and wisen up you figure out that life isn’t about computers or the rat race. It’s about living your life and experiencing it with people.

We need to try and get back to a life without technology having so much of a footprint.

Deepfake vids give anyone plausible deniability which is equally as dangerous IMO.

 

[chet]

They have neutered this thing to the point where it's not usable IMO. A real shame.

 

[Judge Smails]

My pet peeve is all the 20 and 30 somethings trying to be influencers on LinkedIn. Many have a goal to post something, anything, every single day. Nauseating since they are supposed guru’s in Sales or something else even though they hardly have any real world experience. But anything to grow followers including endless recycled posts from others. Now with ChatGPT it’s going to make LinkedIn a complete cesspool.

 

[Capella]

My pet peeve is all the 20 and 30 somethings trying to be influencers on LinkedIn. Many have a goal to post something, anything, every single day. Nauseating since they are supposed guru’s in Sales or something else even though they hardly have any real world experience. But anything to grow followers including endless recycled posts from others. Now with ChatGPT it’s going to make LinkedIn a complete cesspool.

I’m glad I have no idea what any of this sentence means.

 

[fatness]

We haven’t even begun to see the damages technology will have on us. Imo, It’s a cancer and what’s making it worse is that we’re are helping it along.

In a world where 90% of the population can be manipulated and controlled by headlines and propaganda, I don’t see how we come out of this.

It’s cliche and straight out of terminator but once these machines don’t need us we’re done.

This seems like a quick over-reaction to me. First sentence seems like people who swear that emissions from power lines cause brain cancer. Second sentence sounds like just plain paranoia ("they're all insane sheep except us"). Third sentence seems like taking movies more seriously than real life.

 

[STEADYMOBBIN 22]

Remarkable is not how I would describe it.

We haven’t even begun to see the damages technology will have on us. Imo, It’s a cancer and what’s making it worse is that we’re are helping it along.

In a world where 90% of the population can be manipulated and controlled by headlines and propaganda, I don’t see how we come out of this.

It’s cliche and straight out of terminator but once these machines don’t need us we’re done.

Yeah I tend to have this view. Don’t know if it’s just a sign I’m getting old but it is hard to see how this has a net positive effect on the world. It’s probably great for the <1% who control technology and are already super wealthy but for the masses, it seems they are going to be left farther and farther behind.

The deep fake (videos) we are going to see in our lifetime will likely cause some very dangerous problems.

I think it is because you’re getting old(er). When you were younger you likely would have thought this was an amazing idea. As you age and wisen up you figure out that life isn’t about computers or the rat race. It’s about living your life and experiencing it with people.

We need to try and get back to a life without technology having so much of a footprint.

And my father would say just about the same thing - when he wasn't shaking his head and tsk-tsking - if he was watching you type these posts on your phone or computer.

I’ve got ~15 years left of work and this guy will hand in his work phone and not buy another.

Remarkable is not how I would describe it.

We haven’t even begun to see the damages technology will have on us. Imo, It’s a cancer and what’s making it worse is that we’re are helping it along.

In a world where 90% of the population can be manipulated and controlled by headlines and propaganda, I don’t see how we come out of this.

It’s cliche and straight out of terminator but once these machines don’t need us we’re done.

Yeah I tend to have this view. Don’t know if it’s just a sign I’m getting old but it is hard to see how this has a net positive effect on the world. It’s probably great for the <1% who control technology and are already super wealthy but for the masses, it seems they are going to be left farther and farther behind.

The deep fake (videos) we are going to see in our lifetime will likely cause some very dangerous problems.

I think it is because you’re getting old(er). When you were younger you likely would have thought this was an amazing idea. As you age and wisen up you figure out that life isn’t about computers or the rat race. It’s about living your life and experiencing it with people.

We need to try and get back to a life without technology having so much of a footprint.

good one I saw yesterday

I’ve seen some that are better than that one.

We haven’t even begun to see the damages technology will have on us. Imo, It’s a cancer and what’s making it worse is that we’re are helping it along.

In a world where 90% of the population can be manipulated and controlled by headlines and propaganda, I don’t see how we come out of this.

It’s cliche and straight out of terminator but once these machines don’t need us we’re done.

This seems like a quick over-reaction to me. First sentence seems like people who swear that emissions from power lines cause brain cancer. Second sentence sounds like just plain paranoia ("they're all insane sheep except us"). Third sentence seems like taking movies more seriously than real life.

I’d argue with you but your retort didn’t actually say anything except rudely insinuate that I’m being a paranoid conspiracy theorist all because I think technology that allows people to put out incredibly realistic videos of political figures and world leaders is a probably a very bad idea.

Literally every single US government agency agrees that deep fakes are incredibly and increasingly more dangerous but they’re prolly nut jobs too.

 

[grateful zed]

gf is that you?

 

[Soootch]

Write an essay arguing that Eli Manning does not belong in the NFL Hall of Fame.

Eli Manning is a former American football quarterback who played for the New York Giants in the National Football League (NFL). While he had a successful career and won two Super Bowls, there are several reasons why he does not belong in the NFL Hall of Fame.
First and foremost, Manning was a consistent but not dominant quarterback throughout his career. He played in the NFL for 16 seasons and had a regular season record of 117-117. While he did lead the Giants to two Super Bowl victories, he also had several seasons where he struggled and the team missed the playoffs. In comparison, many of the quarterbacks in the Hall of Fame had much more impressive regular season records and consistently led their teams to the playoffs.
Another reason why Manning does not belong in the Hall of Fame is that he was not a statistically impressive quarterback. He had a career passer rating of 84.1, which is below the average for Hall of Fame quarterbacks. He also had a relatively low touchdown-to-interception ratio, with 366 touchdowns and 244 interceptions. These statistics are not on par with the other quarterbacks in the Hall of Fame, who typically have much higher passer ratings and better touchdown-to-interception ratios.
Additionally, Manning was not known for his ability to make plays with his feet. He was primarily a pocket passer and did not have the mobility or scrambling ability of many of the other quarterbacks in the Hall of Fame. This limited his effectiveness and made him less versatile as a quarterback.
Finally, Manning was not a leader on the field in the same way that many other quarterbacks in the Hall of Fame were. He did not have the charisma or presence that many of the other great quarterbacks had, and he was not known for being a vocal leader on the field. This is an important factor in determining whether a quarterback is worthy of induction into the Hall of Fame.
In conclusion, while Eli Manning had a successful career and won two Super Bowls, he does not belong in the NFL Hall of Fame. He was a consistent but not dominant quarterback, he was not statistically impressive, he was not a mobile quarterback, and he was not a strong leader on the field. These factors, combined with the high standards for induction into the Hall of Fame, make it clear that Manning does not deserve this honor.

 

[Westerberg]

Did y'all have to give your phone number to sign up for this thing?

 

[The Z Machine]

Did y'all have to give your phone number to sign up for this thing?

Yeah, but I gave your number, not mine.

 

[Scoresman]

write a haiku about Joe Bryant, the owner of footballguys.com

Football insights flow
Joe Bryant's website shines bright
Touchdown knowledge found

 

[chet]

Has anyone done anything interesting with this AI recently?

 

[leftcoastguy7]

Has anyone done anything interesting with this AI recently?

I saw that a poker player had stumped the AI by asking, "When should you fold pocket aces before the flop in Texas Hold'em? "

The answers it gave were not good. Stuff like --- you should fold pocket aces because someone else might have pocket aces.

 

[The Commish]

You guys are slippin'

What is the value of -5^2 and explain your answer


In mathematics, the caret symbol (^) is used to represent exponentiation, so -5^2 would be read as "-5 to the power of 2."
To find the value of -5^2, we would calculate -5 * -5, which is equal to 25. Therefore, the value of -5^2 is 25.
It's important to note that the caret symbol should not be confused with the multiplication symbol (x) or the letter x. In those cases, the caret would be read as "raised to" rather than "to the power of." For example, 5^2 would be read as "5 raised to the power of 2," not "5 multiplied by 2."

 

[massraider]

A College Student Created an App to Detect If Essays Were Written by ChatGPT

 

[Maurile Tremblay]

Has anyone done anything interesting with this AI recently?

I tried a bunch of different ways to try to make it useful somehow in some aspect of writing a motion for summary judgment. Ultimately, I couldn't find a single way to use it for that. The closest I came was to feed it a paragraph at a time and ask it to identify spelling or other errors, but there are already more efficient ways for computers to help with that.

When I started playing around with it, though, it gave me some false hope at the beginning. I was asking it to summarize the legal rules in California on a bunch of issues, and it was pretty accurate. Then I asked it to cite some case law to support its claims. It gave me citations to cases and summarized them in ways that were quite useful. The more specific instructions I gave it, the more I was impressed. That is, until I looked the cases up and found that they didn't exist. The AI was just making up cases out of whole cloth and giving them fake (but realistic) citations in the case reporters.

 

[Maurile Tremblay]

You guys are slippin'

What is the value of -5^2 and explain your answer


In mathematics, the caret symbol (^) is used to represent exponentiation, so -5^2 would be read as "-5 to the power of 2."
To find the value of -5^2, we would calculate -5 * -5, which is equal to 25. Therefore, the value of -5^2 is 25.
It's important to note that the caret symbol should not be confused with the multiplication symbol (x) or the letter x. In those cases, the caret would be read as "raised to" rather than "to the power of." For example, 5^2 would be read as "5 raised to the power of 2," not "5 multiplied by 2."

Did Wikipedia have an entry on this back when we were discussing it? I don't remember anyone citing it.

Order of operations - Wikipedia

en.wikipedia.org

 

[Caveman33]

You guys are slippin'

What is the value of -5^2 and explain your answer


In mathematics, the caret symbol (^) is used to represent exponentiation, so -5^2 would be read as "-5 to the power of 2."
To find the value of -5^2, we would calculate -5 * -5, which is equal to 25. Therefore, the value of -5^2 is 25.
It's important to note that the caret symbol should not be confused with the multiplication symbol (x) or the letter x. In those cases, the caret would be read as "raised to" rather than "to the power of." For example, 5^2 would be read as "5 raised to the power of 2," not "5 multiplied by 2."

Did Wikipedia have an entry on this back when we were discussing it? I don't remember anyone citing it.

Order of operations - Wikipedia

en.wikipedia.org

Yea, isn't this basic order of operations and the answer is -25? Why is the GPT saying 25?

 

[Maurile Tremblay]

Did Wikipedia have an entry on this back when we were discussing it? I don't remember anyone citing it.

Order of operations - Wikipedia

en.wikipedia.org

Yea, isn't this basic order of operations and the answer is -25? Why is the GPT saying 25?

Current AI powered by deepnets can be impressively creative, but they sometimes make dumb errors. That's why we can use them to generate college essays, but not to drive cars.

 

[IvanKaramazov]

So we zoomed past the Turing test and now have AI that write about as well at the typical undergraduate. Does anybody want to try to set an O/U on how much longer it's going to be before we have AI that produces writing indistinguishable from someone with an 800 Verbal? I'm thinking two years.

My colleagues are currently worried about their students using ChatGPT to cheat. They don't realize that this is going to do writing what calculators did to arithmetic and/or what twentieth-century computing did to statistics. Students are still going to need to learn basic writing just like they need to learn how to do long division, but that's just so you can learn basic skills and understand what's going on under the hood of the computer. Nobody in real life ever needs to calculate OLS estimates by inverting a matrix by hand -- you just need to know that that's what Stata is doing, and you need to understand the process enough to spot coding errors when the results don't make any sense.

In the near future, people will "write" reports by telling the AI to write it for them, and then they'll go back and polish it a little, or tailor it for a specific audience. The skill of knowing how to prompt the AI and how to clean up its work is going to be more important than knowing whether it's okay to end a sentence with a preposition, which is sometimes looked down on. I think that's going to happen before most of us retire. Almost certainly before the end of the decade.

Edit: By "report," I don't mean a book report, or something you can easily Google like "Explain in five paragraph essay why J. Alfred Prufrock had such a hankering for peaches." More like "Write a report to our VP of marketing explaining the quarter-over-quarter FY23 sales figures as they appear in FY23sales.xls."

 

[pmedina]

College essays are going to be a huge problem. Took me about 30 seconds to more or less recreate my daughter's college essay with ChatGPT...or at least an average to slightly above average quality version of the same concept. Her real essay (which she worked on for months) was better - but not by that much, and mostly because of very specific information that she included...the ChatGPT version was just as well written and almost as entertaining.

Used it on an essay already done to see how it would edit it down from 280 to 250 words for a college essay. Result was pretty good!

 

[pmedina]

College essays are going to be a huge problem. Took me about 30 seconds to more or less recreate my daughter's college essay with ChatGPT...or at least an average to slightly above average quality version of the same concept. Her real essay (which she worked on for months) was better - but not by that much, and mostly because of very specific information that she included...the ChatGPT version was just as well written and almost as entertaining.

Used it on an essay already done to see how it would edit it down from 280 to 250 words for a college essay. Result was pretty good!

“Edit the following text to 250 words:”

 

[Dinsy Ejotuz]

which is sometimes looked down on

ISWYDT

 

[Dinsy Ejotuz]

Santiago@svpino

I just gave ChatGPT 200+ lines of JavaScript code and asked it to translate the code into Python. The code parses text using regular expressions. It was complex.

2 minutes later and I had Python code that worked flawlessly.

This thing just saved me a couple of days of work.

 

[STEADYMOBBIN 22]

Santiago@svpino

I just gave ChatGPT 200+ lines of JavaScript code and asked it to translate the code into Python. The code parses text using regular expressions. It was complex.

2 minutes later and I had Python code that worked flawlessly.

This thing just saved me a couple of days of work.

Awesome and terrifying

 

[zoobird]

So we zoomed past the Turing test and now have AI that write about as well at the typical undergraduate. Does anybody want to try to set an O/U on how much longer it's going to be before we have AI that produces writing indistinguishable from someone with an 800 Verbal? I'm thinking two years.

My colleagues are currently worried about their students using ChatGPT to cheat. They don't realize that this is going to do writing what calculators did to arithmetic and/or what twentieth-century computing did to statistics. Students are still going to need to learn basic writing just like they need to learn how to do long division, but that's just so you can learn basic skills and understand what's going on under the hood of the computer. Nobody in real life ever needs to calculate OLS estimates by inverting a matrix by hand -- you just need to know that that's what Stata is doing, and you need to understand the process enough to spot coding errors when the results don't make any sense.

In the near future, people will "write" reports by telling the AI to write it for them, and then they'll go back and polish it a little, or tailor it for a specific audience. The skill of knowing how to prompt the AI and how to clean up its work is going to be more important than knowing whether it's okay to end a sentence with a preposition, which is sometimes looked down on. I think that's going to happen before most of us retire. Almost certainly before the end of the decade.

Edit: By "report," I don't mean a book report, or something you can easily Google like "Explain in five paragraph essay why J. Alfred Prufrock had such a hankering for peaches." More like "Write a report to our VP of marketing explaining the quarter-over-quarter FY23 sales figures as they appear in FY23sales.xls."

I've already started to play around with having it write executive summaries for data analysis results from work, as well as having it synthesize long email threads into short summaries. I'd say the quality of its work is similar to what our entry level analysts (engineering, computer science, math and economics majors from top schools) can do during their first month, before we give them much coaching on how to do more useful summaries.

 

[Dinsy Ejotuz]

I've already started to play around with having it write executive summaries for data analysis results from work

Can you unpack this? Are you feeding it data results and getting written summaries?

 

[zoobird]

I've already started to play around with having it write executive summaries for data analysis results from work

Can you unpack this? Are you feeding it data results and getting written summaries?

Yes, but the data results are already in sentence form. So basically taking 10-20 bullet pointed sentences describing the data results and getting a much more compact summary. What it fails to do (as do our new analysts) is really homing in on what's important and what might be actionable, versus just being essentially interesting trivia.

 

[Keerock]

Has anyone done anything interesting with this AI recently?

I have actually used it to help remediate a couple technical issues at work. Probably nothing that Google couldn't have done, but it was much better compiled and certainly faster

 

[AAABatteries]

Has anyone done anything interesting with this AI recently?

I have actually used it to help remediate a couple technical issues at work. Probably nothing that Google couldn't have done, but it was much better compiled and certainly faster

Can you share more details on this?

 

[CletiusMaximus]

So we zoomed past the Turing test and now have AI that write about as well at the typical undergraduate. Does anybody want to try to set an O/U on how much longer it's going to be before we have AI that produces writing indistinguishable from someone with an 800 Verbal? I'm thinking two years.

I'll take the under. I remember it was back in the late '80s to mid '90s when chess grandmasters went from scoffing at the notion of even playing against a computer, to a few high-profile games with Kasparov beating IBM's Deep Blue, and then a short time later the computers were unbeatable by any human. That was over 20 years ago. I'm far from an expert, but I read a while back that one of the breakthroughs was when they stopped trying to teach computers and just let them learn through self-play. The best AI chess programs are only given the very basic game rules, then instructed to play themselves over and over. The first few hundreds of thousands of games are stupid - many not even ending in checkmate, just kind of ending when the computer couldn't figure out what to do. After a few hundred billion games or so, the top programs are now rated ELO 3,500 where as the top grandmasters are in the high 2's.

The pace this thing is moving is hard to track unless you follow it regularly. I think your prediction is likely spot-on, but who knows? Check back in a few months and it could be going in a direction we've not contemplated.

 

[STEADYMOBBIN 22]

So we zoomed past the Turing test and now have AI that write about as well at the typical undergraduate. Does anybody want to try to set an O/U on how much longer it's going to be before we have AI that produces writing indistinguishable from someone with an 800 Verbal? I'm thinking two years.

I'll take the under. I remember it was back in the late '80s to mid '90s when chess grandmasters went from scoffing at the notion of even playing against a computer, to a few high-profile games with Kasparov beating IBM's Deep Blue, and then a short time later the computers were unbeatable by any human. That was over 20 years ago. I'm far from an expert, but I read a while back that one of the breakthroughs was when they stopped trying to teach computers and just let them learn through self-play. The best AI chess programs are only given the very basic game rules, then instructed to play themselves over and over. The first few hundreds of thousands of games are stupid - many not even ending in checkmate, just kind of ending when the computer couldn't figure out what to do. After a few hundred billion games or so, the top programs are now rated ELO 3,500 where as the top grandmasters are in the high 2's.

The pace this thing is moving is hard to track unless you follow it regularly. I think your prediction is likely spot-on, but who knows? Check back in a few months and it could be going in a direction we've not contemplated.

I’ll also take the under. Moores Law and all.

 

[Keerock]

Has anyone done anything interesting with this AI recently?

I have actually used it to help remediate a couple technical issues at work. Probably nothing that Google couldn't have done, but it was much better compiled and certainly faster

Can you share more details on this?

Sure.

How can I determine if a sql server query is using query store?


To determine if a query is using the Query Store, you can run the following statement:

SELECT qs.query_id, qs.query_sql_text
FROM sys.query_store_query qs
WHERE qs.query_sql_text LIKE '%YOUR_QUERY_HERE%';

Replace YOUR_QUERY_HERE with the text of the query you want to check. This will return a row for each instance of the query that is in the Query Store, along with the query's ID.
You can also check for the presence of the query in the Query Store by looking at the sys.query_store_query_text view. This view contains a row for each query text that is in the Query Store, along with the ID of the query.
If the query is not present in either of these views, then it is not being tracked by the Query Store.
Note that the Query Store must be enabled and configured for your database in order for it to track queries. You can check if the Query Store is enabled for a database by querying the is_query_store_on column of the sys.databases catalog view.

How can I resolve an ORA-01555 error


An ORA-01555 error indicates that an old snapshot of data is being used by a query. This can occur when a long-running query is attempting to access data that has been modified by another transaction that has not yet been committed.
Here are some steps you can take to resolve this error:

1. Identify the query that is causing the error. This can usually be done by looking in the Oracle alert log or by using tools such as Oracle Enterprise Manager.
2. Make sure that the query is using the most up-to-date data by committing any outstanding transactions.
3. If the query is running too long and is causing the ORA-01555 error, you may need to optimize the query or break it up into smaller pieces.
4. If the problem persists, you may need to increase the value of the UNDO_RETENTION parameter in the initialization file for the database. This will increase the amount of undo data that is retained in the UNDO tablespace, which may help to avoid the ORA-01555 error.
5. If none of these steps help, you may need to contact Oracle Support for further assistance.

 

[Nathan R. Jessep]

NYC Bans Students and Teachers from Using ChatGPT

The machine learning chatbot is inaccessible on school networks and devices, due to "concerns about negative impacts on student learning," a spokesperson said.

www.vice.com

 

[Scoresman]

NYC Bans Students and Teachers from Using ChatGPT

The machine learning chatbot is inaccessible on school networks and devices, due to "concerns about negative impacts on student learning," a spokesperson said.

www.vice.com

List the steps required to get around a website being banned on my network


Here are a few steps you can try to get around a website being banned on your network:

1. Use a proxy server: A proxy server acts as an intermediary between your device and the internet. You can use a proxy server to access websites that are blocked on your network by connecting to the proxy server and then visiting the website through the proxy server's connection.
2. Use a VPN: A virtual private network (VPN) encrypts your internet connection and routes it through a server in a different location. This can allow you to access websites that are blocked on your network, as the VPN makes it appear as if you are connecting from a different location.
3. Use the website's IP address instead of its URL: Websites are often blocked by URL, but the website's IP address may not be blocked. You can find a website's IP address by using the "ping" command in a command prompt or terminal window.
4. Use a mobile data connection: If the website is only blocked on your network, you may be able to access it using a mobile data connection (e.g., by connecting to your phone's hotspot).
5. Use a different device: If the website is blocked on your device, you may be able to access it using a different device (e.g., a friend's laptop).

Please note that attempting to bypass a website ban may violate the terms of service of your network or internet service provider, and it may also be illegal in some cases. You should carefully consider the consequences of attempting to bypass a website ban before doing so.

 

[CletiusMaximus]

Talking with my son about this today - his school has blocked ChatGPT and are probably going to follow the trend of banning it - it is apparent that this goes way beyond writing and will be useful in many other disciplines as several have noted in this discussion. He's putting physics questions into it and the program is creating perfect answers with narrative analysis, finding and using the right formulas and doing the math. There are surely many other programs in the works that will address specific uses and areas in the arts and sciences.

He also asked it to describe a 1 v 1 basketball game to eleven between Ang (Avatar) and Anakin Skywalker. Ang won 13-11 using his airbending powers to sky a 3 pointer. Anakin had taken an 11-10 lead using the force to clear Ang out of the way for an easy dunk. The program assumed the 'win by 2' rule applied even though that was not specified.

 

[Corporation]

Talking with my son about this today - his school has blocked ChatGPT and are probably going to follow the trend of banning it - it is apparent that this goes way beyond writing and will be useful in many other disciplines as several have noted in this discussion. He's putting physics questions into it and the program is creating perfect answers with narrative analysis, finding and using the right formulas and doing the math. There are surely many other programs in the works that will address specific uses and areas in the arts and sciences.

He also asked it to describe a 1 v 1 basketball game to eleven between Ang (Avatar) and Anakin Skywalker. Ang won 13-11 using his airbending powers to sky a 3 pointer. Anakin had taken an 11-10 lead using the force to clear Ang out of the way for an easy dunk. The program assumed the 'win by 2' rule applied even though that was not specified.

With some of the responses I've seen it give I'd be wary of it giving "perfect" answers for a physics test or homework. But I do agree with the overall sentiment that a lot of students will try to get out of writing using this thing.

 

[CletiusMaximus]

This guy is working on an AI lawyer. Seems to have somewhat serious funding.

Save Time and Money with DoNotPay!

Continue surfing free trials, beating parking tickets, suing robocallers for cash, and more.

donotpay.com

 

[Dinsy Ejotuz]

serious funding

If the Chainsmokers are in, I'm in!

Pretty interesting idea.

 

[Psychopav]

Something that came up at work today, figured I'd see what ChatGPT had to say about it. Good, nuanced response...

What is the benefit of being able to conveniently lock your account with one credit reporting agency but not all three?


Locking your credit report with one of the three credit reporting agencies, Equifax, Experian, or TransUnion, can help protect you against identity theft by making it more difficult for someone to open a new credit account in your name. However, it is important to note that locking your credit report with one credit reporting agency does not prevent all forms of identity theft. For example, if someone has already obtained your personal information and uses it to open a new credit account, that account could still appear on your credit report from a different credit reporting agency. Therefore, it is generally advisable to lock your credit report with all three credit reporting agencies to provide the maximum level of protection against identity theft.

 

[Keerock]

I've had a song in my mind that I wanted to write... asked chatGPT to do it. Massive failure. Probably not asking/providing the right way, but was a fun yet frustrating exercise.

 

[zoobird]

I've had a song in my mind that I wanted to write... asked chatGPT to do it. Massive failure. Probably not asking/providing the right way, but was a fun yet frustrating exercise.

I had a similar experience with an idea for a short story that I thought of a few years back. Might give it another shot at some point though.

 

[Dinsy Ejotuz]

I had a similar experience with an idea for a short story that I thought of a few years back. Might give it another shot at some point though.

I plugged in a bunch of sequential scene setters with a prompt on what should happen in each, and had it run each scene three or four times. Copied and kept each version. Absolutely have the skeleton and a lot of detail that could be cleaned up, fleshed out, and used as a single chapter.

 

[Maurile Tremblay]

I'm far from an expert, but I read a while back that one of the breakthroughs was when they stopped trying to teach computers and just let them learn through self-play. The best AI chess programs are only given the very basic game rules, then instructed to play themselves over and over.

There's a good documentary on this in the context of Go -- AlphaGo, available on YouToube.

 

[Maurile Tremblay]

Talking with my son about this today - his school has blocked ChatGPT and are probably going to follow the trend of banning it - it is apparent that this goes way beyond writing and will be useful in many other disciplines as several have noted in this discussion. He's putting physics questions into it and the program is creating perfect answers with narrative analysis, finding and using the right formulas and doing the math. There are surely many other programs in the works that will address specific uses and areas in the arts and sciences.

He also asked it to describe a 1 v 1 basketball game to eleven between Ang (Avatar) and Anakin Skywalker. Ang won 13-11 using his airbending powers to sky a 3 pointer. Anakin had taken an 11-10 lead using the force to clear Ang out of the way for an easy dunk. The program assumed the 'win by 2' rule applied even though that was not specified.

With some of the responses I've seen it give I'd be wary of it giving "perfect" answers for a physics test or homework. But I do agree with the overall sentiment that a lot of students will try to get out of writing using this thing.

Yes, it can be humorously bad at arithmetic. It's funny because a very basic calculator from the 1970s can kick any human's butt at multiplying four-digit numbers together, but something as sophisticated as ChatGPT can struggle to add four plus eleven. I suspect good AI is going to have multiple modules that each do different things. ChatGPT can be the language model and a calculator can do arithmetic and some other module can process visual information, and then some other module can be in charge of figuring out which module to use for which task as it puts everything together. I think this is roughly how mammalian brains work as well. Different neural networks (if not exactly different parts of the brain) used for different tasks, and they can operate somewhat independently of each other.

 

[Big League Chew]

I added ChatGPT expert to my resume and got 25 hits in a week

 

[Nathan R. Jessep]

I posted this in another thread but thought I'd drop it here too.
Dr. Jeremy Faust has a substack and he posted recently that he tried it out. It came up with some diagnoses and also wrote up an accurate patient chart, once it had the right input. Pretty impressive.

link: https://insidemedicine.substack.com/p/fun-with-openai-medical-charting

 

[jhib]

I'm sure I'm late to the game on this, but was just playing around with DALL-E 2, one of those art AI that creates an image based on your description. Can use your same login as for openai that you use for ChatGPT, but you only get a limited free usage each month. Pretty neat.

DALL·E

Experiment with DALL·E, an AI system by OpenAI

labs.openai.com

 

[Maurile Tremblay]

A College Student Created an App to Detect If Essays Were Written by ChatGPT

This seemed immediately obvious.

1) Created an article using GPT-3.
2) Used Originality AI to detect whether the content was AI or human generated. Score: 100% AI
3) Prompted GPT-3 to rewrite the article "to avoid detection by AI content detection tools".
4) New Score: 68% Human

Fairly confident with a couple more passes I could get it to score 100% human.