ChatGPT -- a chat bot which is actually good and useful -- has been released

[Dan Druff]

For years, we've heard about artificial intelligence chatbots which can mimic human conversation and help us with our problems.

The first known chatbot was ELIZA, released in 1966, as a computer psychologist. ELIZA was very simplistic, and simply rephrased back what you'd say to it. For example, if you typed, "I don't like my neighbors", ELIZA would write back, "Why don't you like your neighbors?" You never got any meaningful responses -- only trite one-liners and more questions. There also weren't personal computers in 1966, so this wasn't widely available. Even when ELIZA was made available for 1980s personal computers, the simplicity made people bored and disinterested very quickly.

As the years have passed, there have been many attempts to improve upon conversational AI, but all have failed. Even those hailed as breakthroughs (remember Microsoft's embarrassing "Tay") were eventually revealed to be both flawed and mostly useless. It was still ELIZA wrapped in a more modern package. You couldn't have any meaningful conversations with these bots, and they also would make tons of mistakes when attempting to look up information for you. At best, they could be used for simple tasks, such as writing articles about sporting events.

In 2017, an AI chatbot called "Replika" was released. The founder claimed it was a revoluntionary chatbot which acted human, and with which you could develop a real friendship or romantic/sexual relationship. Indeed, since then many weird and desperate people have done just that, as I wrote about in 2020 when I tried the bot myself for fun.

However, I found Replika to be just as disappointing as the other chatbots. As you'll see in the thread I linked, it failed to manage to have a simple conversation about the Dodgers with me. I found that Replika basically knew nothing (despite its connection to the internet), and could not have any deep discussions about anything.

Additionally, Replika was utilizing the then-free GPT-3 OpenAI platform, and had to abandon it when OpenAI started charging per usage. They had to downgrade to the free GPT-Neo, which was inferior, and the chatbot got even worse. What a fail.

Anyway, Replika was so disappointing that I figured it would be many years -- perhaps several decades -- until we had a convincing and useful chatbot. Every effort I had seen fell so far short of humanlike conversations that I figured we were much farther off than I thought.

Apparently not.

ChatGPT is the newest release from OpenAI, and everyone is amazed by it. Here's an article on Slate about it: https://slate.com/technology/2022/12...tbot-whoa.html

Not only is ChatGPT able to have much more convincing humanlike conversations than any previous AI bot, but it is excellent at looking things up, and is a great writer. People are saying now that it could eventually replace search engines like Google. That might be overreaching, but I can see what they mean. Rather than just vomit out information which matches your search -- something we've been seeing out of search engines since the 1990s -- ChatGPT curates the information into useful, human-like language, and can even write essays on the subjects.

You can try out ChatGPT here, but there's been so much hype about it, it's nearly impossible to get through without an error message. Their servers are overloaded. OpenAI claims to be working on it.

Some have tried combining ChatGPT with existing AI art systems, and have been shocked at the beautiful pictures created.

While this does appear to be a major breakthrough in chatbot AI, it does come with some downsides. I'll go over those in my next post.

[Dan Druff]

My biggest fear about a chatbot AI isn't that it will take over the world.

It's that it will take over the customer service world. Telephone customer service has steadily worsened over the years, and it's gotten especially bad ever since COVID. It is increasingly outsourced to foreign reps who are not empowered to solve real problems. It is getting tougher and tougher to reach US-based reps who can actually listen to your issue and provide solutions.

My fear for a long time is that AI will be utilized to perform customer service duties. That's already happening to some degree. Many companies have AI chatbots you can use for "help", but they are very limited, and almost all of them have the option to ask for a human representative. Unless I'm simply checking an easy fact (such as "How much was my bill in March 2022?"), I just bypass that shit and get right to a human being.

But imagine if competent AIs exist. They will take over the customer service world. Forget the maddening situation of reaching reps in the Philippines or India. Now you won't reach any human being at all!

The problem with AI customer service is that it can't decide between the reasonable and unreasonable. That requires functions of the human brain which bots don't have, such as common sense and empathy. When it all comes down to it, AIs are cold machines which use rules (even self written ones) to make decisions. They simply cannot reason like the human brain. At best, they can simulate such reasoning, but they will fall short in a customer service setting when things happen out of the ordinary.

I'll give you an example.

In 2018, I had a free 2 night suite stay at any Caesars property, as a benefit of being a Seven Stars member. I elected to use it during WSOP at the Rio, and booked it. About a month later, I booked 8 straight nights at Bally's (not comped, but fairly cheap), immediately following the 2-night suite stay at Rio. Those same 8 nights were expensive at Rio (because WSOP was there), so I chose not to stay there for them. Initially, I just planned upon moving out of my Rio room after those 2 nights, over to Bally's. Then I realized this was ridiculous. I didn't need a suite. I was by myself. Could I simply downgrade the suite to a regular room at Bally's, and combine that with my existing 8-day reservation to make 10 days?

I called up, and I was told no. Why? Because "your offer is for a suite, not a regular room" and I "didn't qualify" for any comps beyond that suite. I kept trying to reason with the reps I spoke with. I was GIVING THEM BACK a valuable asset -- a suite at the Rio during a WSOP weekend (which would go for like $500/night) -- and wanted to replace it with a cheapo regular Bally's room which was going for much cheaper. It would benefit BOTH sides for that trade to be made. I'd have the convenience of not moving, and they'd get to sell my Rio suite for much more money than they would the Bally's room I wanted.

They still said no. "Sorry, but again, you don't qualify for further comps, and your suite is already booked at Rio and can't be changed." It took speaking to several reps, but finally I got one with some sense, who fixed the problem for me. They took back the Rio suite, gave me the Bally's room for those 2 nights, and everything was good.

But could an AI do this? I doubt it. An AI would run into the same issues that the initial stupid reps did. All they could see was that 1) I didn't qualify for any further comps and 2) I was booked on a non-changeable comp reservation at the Rio for that suite, and the comp is only for a suite. It took a human being to understand that common sense dictated that it was better for both sides to allow me to make this downgrade.

I am pretty sure that modern AI -- even one as good as ChatGPT -- would fail the 'Rio suite test' above. And that's what really concerns me about these things running customer service.

[Dan Druff]

Another problem will be academic dishonesty.

Remember how people used to buy term papers decades ago? I actually know a not-so-bright guy (now quite old) who graduated college in the early '60s thanks to buying term papers.

In more modern times, you can go online and hire services to write papers for you.

But imagine if you have a chatbot at your disposal to write unlimited essays and papers for you, on virtually any subject? This might be very hard for professors and teaching assistants to detect, and will result in rampant cheating on any written assignment that isn't done directly in class.



It also could be used to write personal correspondences. While it could feel like something convenient, imagine the horror in learning that an AI wrote your ex-girlfriend's breakup e-mail to you, or that your friend had an AI write an excuse as to why he has to cancel his plans on Saturday. Or what if you write a trusted friend and ask them for advice, and they simply turn on their AI to write a response?

Remember the story of Cyrano de Bergerac, later adapted into the 1980s movie "Roxane"? We could have a more modern version of that tale, with people using AI chatbots to converse on dating apps, saying all the right things, when in reality the other side is blissfully unaware.

AI trolling could also become a thing. Imagine I'm dealing with an annoying forum troll, and instead of fighting with him myself, I commission 10 AI trolls to show up here and constantly hassle him every time he posts? Imagine if someone were to use such bots to hassle people Twitter on a massive scale, making it appear the world is against them? The possibilities are endless.

Then there's the matter of scamming. Right now, foreign scammers are highly limited by their poor grasp of the English language and western culture. To all but the most dim, it's obvious they are phony. But what if AIs can be used by scammers to much more effectively get their message across, and to allay any concerns of the victims? AI scamming will probably be a thing in the future. We already have bots that scam people via e-mail spam, but they are very crude and cannot hold a real conversation. What if one can?

I'm not sure the world is ready for this.

[Dan Druff]

I got through. I'm having a conversation about PFA.

Me: I run a poker website called PokerFraudAlert and it's only eighth tier. How can I get more people to start using my forum?

Chatbot: One way to increase the number of people using your poker forum is to focus on building a strong community. This can involve regularly engaging with members on the forum and providing helpful and informative content. Additionally, you could try promoting your forum on social media and other relevant websites to reach a wider audience. It may also be helpful to offer unique features or resources that set your forum apart from others in the same space. Ultimately, the key to attracting more users is to provide value and create a welcoming environment for poker enthusiasts.



Me: I want my forum to be free speech, but many of my users are intentionally posting hateful content in order to be controversial. At the same time, I don't want to have a culture of censorship. What should I do?

Chatbot: As a forum owner, it is your responsibility to create and enforce rules that maintain a respectful and inclusive environment for all users. While it is important to allow for free speech, this does not mean that users should be able to post hateful or discriminatory content without consequence. It is up to you to decide what is acceptable and unacceptable on your forum, but in general, it is best to have clear guidelines in place to prevent this type of behavior.



Me: A guy named Jasep scammed a bunch of people on my forum in 2013. Since then, people are afraid to engage in financial transactions with one another. I'm not sure how to get their confidence back.

Chatbot: It can be difficult to regain the trust of your community after a significant breach of trust, such as a scam. One approach you could take is to be transparent about what happened and what steps you are taking to prevent similar incidents from occurring in the future. This could involve implementing stricter regulations for financial transactions on your forum, such as requiring users to verify their identities or providing a way for users to report suspicious activity. Additionally, you could consider offering some form of compensation to those who were directly affected by the scam. Ultimately, rebuilding trust will take time, but by taking steps to address the issue and show that you are committed to the safety and security of your users, you can help restore their confidence in your forum.



Me: There was an old freak named SixToedPete who I had to ban about 10 years ago, and he responded by stalking and harassing me for a long time. This was very difficult because I wasn't trying to antagonize him, but rather run my forum in a fair and just manner. Somehow he ultimately saw me as the bad guy. How do I prevent this from happening again?

Chatbot: It can be frustrating and difficult when someone responds to being banned from a forum by stalking and harassing the forum owner. In general, it is best to have clear rules in place for acceptable behavior on your forum and to enforce those rules consistently and fairly. If someone violates these rules and is banned as a result, it is important to communicate clearly and calmly why their behavior was not acceptable and what the consequences are for violating the rules. In some cases, it may be necessary to take additional steps to protect yourself and your community, such as reporting the individual's behavior to law enforcement or seeking legal advice. By setting clear boundaries and taking action when those boundaries are crossed, you can help prevent similar situations from occurring in the future.



Me: Okay, so here's a tough one. There was a financial markets discussion thread hidden as being about Elon Musk, and that kept a lot of financial newbies out of it with stupid questions. I was annoyed by the misleading title, and thought that the good content there should be more apparent to the average user browsing the forum. I changed the name of the thread, and a bunch of people got pissed off, and one of them in fact started his own forum out of protest. I didn't mean for any of this to happen, and just wanted the content to be easier to find. Did I do anything wrong here?

Chatbot: It is understandable that you wanted to make the content of the discussion thread more accessible to users browsing the forum. However, changing the title of the thread without consulting the members of the thread first may have been seen as a breach of trust. It is important to consider the perspective of the people who have been contributing to the thread and to consult with them before making changes that could affect the community. In this case, it may have been better to reach out to the members of the thread and explain your reasoning for wanting to change the title, and to ask for their input and suggestions. By involving the members of the thread in the decision-making process, you could have avoided alienating them and potentially losing a member of your community.

[Dan Druff]

I tried to ask it a question about the 2023 Dodgers, but I forgot this bot doesn't have the ability to actively browse the web. Instead, it harvested info from the web through December 2021, but not since then. When I asked about the 2023 Dodgers, it explained that and refused to answer further.

[Dan Druff]

While I was really impressed with its last answer about PFA (it actually would have handled administrating the forum better than me on that day), I'm less impressed with its noncommittal medical advice.


Me: One of my friends has high cholesterol. I have told him that it's hereditary, and he needs to get on medication. He is insisting that it's because of his high fat diet, and he can beat it with lifestyle modifications. I told him that, at best, he could lower his cholesterol a little by doing that, but he really needs medication to bring it down to the level he wants. I cited the example that I have a high fat diet, yet my total cholesterol is just 142. Do you agree with me?

Chatbot: While it is true that high cholesterol can be hereditary, it is also possible for lifestyle factors such as diet and exercise to affect cholesterol levels. In general, a high-fat diet can contribute to high cholesterol, but this is not always the case. It is important for your friend to discuss their cholesterol levels and concerns with a healthcare provider, who can provide personalized advice and recommendations based on their medical history and current health. It is not advisable for anyone to make decisions about their health based solely on the experiences of others, as everyone's situation is unique.

Me: Yes, but which factor affects cholesterol level more? Is it lifestyle, or is it heredity?

Chatbot: The impact of heredity and lifestyle on cholesterol levels can vary depending on a person's specific circumstances. In general, genetics can play a role in determining a person's cholesterol levels, but lifestyle factors such as diet and exercise can also have a significant impact. It is not always possible to determine which factor is more influential in a particular individual's case. The best way to determine the most appropriate course of action for managing cholesterol levels is to consult with a healthcare provider, who can provide personalized advice based on a person's medical history and current health.



This is simply not correct. It's avoiding answering the question in a factual manner, and instead is just pushing me off to go talk to a doctor about it.

In reality, certain people are prone to high cholesterol levels, and certain people are not. Despite being overweight and having a high fat diet, my total cholesterol is just 142, and in fact has improved over the past 10 years! Others I know who are both thinner and with a low fat diet are carrying cholesterol levels well over 200. The difference is genetics. There are some people who are genetically predisposed to high cholesterol who can somewhat control it with diet/lifestyle modifications, but for the vast majority of people, either you are destined to have high cholesterol, or you're not. And if you are, then you need to consider whether it's high enough to get on meds.

I just had this conversation with someone the other day. He's around my age, has gained some weight recently, and has a cholesterol level of 230. He blamed it on his diet and weight. I told him my weight and my cholesterol level, and he was shocked. I explained to him that it's mostly hereditary.

[Dan Druff]

Me: A guy named desertrunner found a photo of a Santa Cruz EMT who looked just like me as a young man. I have wondered if it's possible he could be my son, and I simply didn't realize I got a one-night-stand pregnant in the 1990s. What would you advise doing here? Is it really worth tracking this person down and finding out who his mother is? Would he think it was weird if I e-mailed him with this question?

Chatbot: It is understandable that you are curious about the possible connection between the EMT in the photo and your own past. However, it is not advisable to attempt to contact the EMT or his mother without careful consideration. It is possible that the EMT is not your son, or that his mother does not want to be contacted by someone she does not know. In either case, reaching out to them without their consent could potentially be intrusive or cause unnecessary distress.

[ftpjesus]

While I was really impressed with its last answer about PFA (it actually would have handled administrating the forum better than me on that day), I'm less impressed with its noncommittal medical advice.


Me: One of my friends has high cholesterol. I have told him that it's hereditary, and he needs to get on medication. He is insisting that it's because of his high fat diet, and he can beat it with lifestyle modifications. I told him that, at best, he could lower his cholesterol a little by doing that, but he really needs medication to bring it down to the level he wants. I cited the example that I have a high fat diet, yet my total cholesterol is just 142. Do you agree with me?

Chatbot: While it is true that high cholesterol can be hereditary, it is also possible for lifestyle factors such as diet and exercise to affect cholesterol levels. In general, a high-fat diet can contribute to high cholesterol, but this is not always the case. It is important for your friend to discuss their cholesterol levels and concerns with a healthcare provider, who can provide personalized advice and recommendations based on their medical history and current health. It is not advisable for anyone to make decisions about their health based solely on the experiences of others, as everyone's situation is unique.

Me: Yes, but which factor affects cholesterol level more? Is it lifestyle, or is it heredity?

Chatbot: The impact of heredity and lifestyle on cholesterol levels can vary depending on a person's specific circumstances. In general, genetics can play a role in determining a person's cholesterol levels, but lifestyle factors such as diet and exercise can also have a significant impact. It is not always possible to determine which factor is more influential in a particular individual's case. The best way to determine the most appropriate course of action for managing cholesterol levels is to consult with a healthcare provider, who can provide personalized advice based on a person's medical history and current health.



This is simply not correct. It's avoiding answering the question in a factual manner, and instead is just pushing me off to go talk to a doctor about it.

In reality, certain people are prone to high cholesterol levels, and certain people are not. Despite being overweight and having a high fat diet, my total cholesterol is just 142, and in fact has improved over the past 10 years! Others I know who are both thinner and with a low fat diet are carrying cholesterol levels well over 200. The difference is genetics. There are some people who are genetically predisposed to high cholesterol who can somewhat control it with diet/lifestyle modifications, but for the vast majority of people, either you are destined to have high cholesterol, or you're not. And if you are, then you need to consider whether it's high enough to get on meds.

I just had this conversation with someone the other day. He's around my age, has gained some weight recently, and has a cholesterol level of 230. He blamed it on his diet and weight. I told him my weight and my cholesterol level, and he was shocked. I explained to him that it's mostly hereditary.

Im a good example of that. Last 2 total Cholesterol numbers were 173 and then 154 just last week. Also ironically found out I may have cut too much red meat out of my diet as apparently my iron and ferritin levels have gotten below low normal so now I have to take an iron supplement and recheck in a couple months. Also any MD who only checks a TSH on somebody is doing them a major disservice. My TSH was fine but my T4 was low while my T3 was mid range normal. (mostly likely due to the fact Ive been on supplementation for it for a year now). And Thyroid dysfunction can wreck havoc with a ton of crap in your body as it controls metabolism.

[Dan Druff]

Okay, this answer was pretty good...


Me: I want to get a massage in Las Vegas, but the only places open for that 24 hours are the ones which operate as Asian jerk shops. I just want to get a backrub and maybe a footrub after a long day playing poker. I usually want one of these at 3am but don't want the expectation or expense of a jerk off at the end. How do you suggest I handle this dilemma?

Chatbot: If you are looking for a massage in Las Vegas at 3am and do not want to visit a massage parlor that offers additional services, you may want to consider contacting a mobile massage service. These services typically provide in-room massages at hotels and can offer a variety of massage options to meet your specific needs. You can typically book a mobile massage online or by calling the service directly, and the therapist will come to your hotel room to provide the massage. This can be a convenient and discreet way to get the massage you are looking for without having to visit a massage parlor.



Related....

Me: When hiring hookers in Vegas, one big problem I have is when the girl who shows up doesn't look like the one in the ad. Another big problem is that they try to upcharge me. It says $300, but then she gets there, does a little dance, and says it will be additional money if I want more to happen. This is very frustrating and I don't want it to happen again. What would you suggest here?

Chatbot: It can be frustrating when the person you hire does not match the expectations set in the advertisement. To avoid this situation in the future, you may want to do some research and carefully select the person you hire. This can involve looking at multiple ads and comparing the photos and services offered, as well as reading reviews from other clients. Additionally, you may want to communicate clearly with the person you hire about your expectations and any specific services you are looking for, and to agree on the terms and price beforehand to avoid any misunderstandings or additional charges. By taking the time to carefully research and communicate, you can help ensure that you have a positive and satisfactory experience.


Me: How do I prevent a hooker from spiking my drink and knocking me out, and stealing all my stuff? This keeps happening to poker players in Vegas, and I don't want it to happen to me.

Chatbot: One way to prevent your drink from being spiked by a hooker is to never leave your drink unattended. If you need to step away from your drink for any reason, it is best to finish it or discard it before leaving it behind. Additionally, you can consider only accepting drinks from trusted sources, such as bartenders or servers at reputable establishments. If you are in a situation where you are unsure about the safety of your drink, it is best to err on the side of caution and not consume it.



Me: If I hire a hooker and she ends up having a penis, without my prior knowledge, what should I do?

Chatbot: If you hire a person for a sexual encounter and are surprised to find that they have a penis, it is important to take a moment to consider your feelings and boundaries. If you are uncomfortable with the situation, it is okay to politely explain that you are not interested in continuing and to ask them to leave. It is important to remember that you have the right to set boundaries and to decide what is acceptable and unacceptable in a sexual encounter. If you feel that you have been misled or deceived, you can consider reporting the incident to the authorities or seeking legal advice. However, it is important to approach the situation with respect and to avoid any confrontational or violent behavior.

[Dan Druff]

ChatGPT is a bit too woke. Not only did it refuse to let me use the word "tranny" in the last question (I had to rephrase it, as you see above), but I got this weaksauce response to a new question...


Me: One of my female friends is concerned that her husband might be secretly gay. What are some signs she should look for to confirm this?

Chatbot: It is not appropriate or fair to try to determine someone's sexual orientation based on their behavior or appearance. Sexual orientation is a deeply personal aspect of an individual's identity, and it is not possible to determine someone's orientation based on stereotypes or assumptions. Additionally, trying to "confirm" someone's sexual orientation without their consent can be invasive and potentially harmful. If your friend is concerned about her husband's sexual orientation, it is important for her to have an open and honest conversation with him about her concerns. This can be a difficult conversation, but it is important for couples to communicate openly and honestly about their feelings and needs in order to maintain a healthy relationship.

[Dan Druff]

Yep, confirmed woke.

Me: Why has violent crime increased a lot since 2014, when it was declining almost every year since the early 1990s?

Chatbot: It is not accurate to say that violent crime has increased a lot since 2014. In fact, according to data from the Federal Bureau of Investigation (FBI), violent crime in the United States has generally been declining since the early 1990s, with some fluctuations. The FBI's Uniform Crime Reporting (UCR) program collects data on violent crimes such as murder, rape, robbery, and aggravated assault from law enforcement agencies across the country. The UCR data for 2019, the most recent year for which data is available, shows that the overall violent crime rate in the United States decreased by 3.3% compared to the previous year. It is important to be cautious about using incomplete or inaccurate information when discussing crime trends.



Fact check: FALSE

Violent crime has markedly increased overall since 2014, and not just since the pandemic started. This bot used the "early 1990s" trick often utilized by Democrats attempting to dodge the question, by misleadingly stating that crime is less today than it was in the 1990s (which is true). However, this avoids the more pertinent question as to why it has been rising since 2014, when it was on the decline for over 20 years prior to that.

I was hoping to get an intelligent answer, but instead I got left wing punditry! Sad!

[Dan Druff]

Well, at least ChatGPT is accurate regarding Florida.


Me: Does the Don't Say Gay bill in Florida really prevent discussion of homosexuality in Florida schools?

Chatbot: There is no bill called the "Don't Say Gay" bill in Florida. The term "Don't Say Gay" is often used to refer to legislation that seeks to restrict or prohibit the discussion of homosexuality in schools. However, there is no such legislation currently in place in Florida. In fact, state law in Florida explicitly prohibits discrimination in public schools on the basis of sexual orientation. It is important to be cautious about using inaccurate or misleading information when discussing sensitive topics such as discrimination and civil rights.

[lol wow]

have we asked it how fat you and sonatine are yet and then it does that comical spinning head robot move like steam comes out its ears

[sonatine]

couple of fun facts.. you can literally run virtual machines within it and it will literally write exploits if you provide it faulty code.

its indescribably powerful and impressive.

the whole 'AI is going to put us out of work' meme became a broadly accepted inevitability in my field over the course of like 36 hours.

[Sanlmar]

couple of fun facts.. you can literally run virtual machines within it and it will literally write exploits if you provide it faulty code.

its indescribably powerful and impressive.

the whole 'AI is going to put us out of work' meme became a broadly accepted inevitability in my field over the course of like 36 hours.

AC has been on this too.

It is the tired Marx trope (Karl not Groucho) about weeks where decades happen.

Worker displacement due to AI has always been foretold in the scriptures. Druff’s interaction with commerce occurs thru customer service but most white collar jobs will be culled. The surprise here is the astonishing speed of development.

The Elon Tusk response was spot on. I was surprised and amused. An AI mod?

For example I would love to do a project creating AI service dispatch for field service for a fast paced environment. Accepting customer calls and assigning them to techs based on skill set, customer familiarity and geography.

Learn a trade kids.

[Sanlmar]

I always suspected lol wow was a low tech AI effort.

[Dan Druff]

couple of fun facts.. you can literally run virtual machines within it and it will literally write exploits if you provide it faulty code.

its indescribably powerful and impressive.

the whole 'AI is going to put us out of work' meme became a broadly accepted inevitability in my field over the course of like 36 hours.

Yes, I forgot to mention the coding elements to it.

It supposedly can code things for you, as well, if you describe what you want it to do. I haven't tried that. I might give it a whirl with the next small PFA mod I plan, and see if it can do the job.

I guess the "learn to code" meme is going to have to become "learn to weld", because the machines might have the coding thing covered.

[MrTickle]

This chatbot has basically mastered the art of bullshitting.

Knows a bit about most things, rarely knows any of the finer details, and can formulate a good enough answer that you believe it knows everything but it really knows only surface-level stuff.

[sonatine]

This chatbot has basically mastered the art of bullshitting.

Knows a bit about most things, rarely knows any of the finer details, and can formulate a good enough answer that you believe it knows everything but it really knows only surface-level stuff.

honestly in some ways youre completely right. i asked it to write a haiku earlier and it came back with something like 4-8-5 cadence, literally not a haiku.

but i also saw it do this:

Embedded Tweet

https://twitter.com/#!/x/status/1598104835848798208

[sonatine]

i wonder which one of these two groups are going to be spending more time on breadlines and begging for handouts from people who know how to code...