What's new

OpenAI Chatbot...

I want an AI chatbot 21 questions game.

I had a stupid handheld electronic 21 questions game back int he early 2000s and it was entertaining. I'd love to see an AI powered 21 questions game that was essentially unbeatable. Maybe that would take the fun out of it or maybe it would have to adjust to 14 questions or 10 questions or 7 questions.
 
I want an AI chatbot 21 questions game.

I had a stupid handheld electronic 21 questions game back int he early 2000s and it was entertaining. I'd love to see an AI powered 21 questions game that was essentially unbeatable. Maybe that would take the fun out of it or maybe it would have to adjust to 14 questions or 10 questions or 7 questions.
Did you ask it to play?
 
Did you ask it to play?
let me try

Yeah it will play. It isn't as good at it as I would have expected.

First I thought of a pocket knife and it eventually guessed a Mezzaluna Knife, which was weirdly specific. It seemed to follow specific trails too doggedly instead of doing the typical divide by half type of questions first.

The second round I thought of a wireless earbud and it took 22 questions before it guessed speaker, which was oddly generic compared to the first round. It wasted a bunch of questions after it seemed to think this was some sort of toy and made guesses that were pretty solidly excluded in previous questions. For instance it asked if it used electricity by like question 3 and I said yes but then it later asked if it was a fidget toy or a deck of playing cards. I said no and then it asked it if was used for relaxation or stress relief. I said sometimes. Then it built on the speaker thing too incrementally in my opinion. By question 18 it was asking if it produced sound, then it asked if it played music in another question, which I felt was a wasted question and it still took a few more guesses before saying speaker.
 
Last edited:
@Gameface (or anyone else who has a ChatGPT account), could you do me a favor and ask the bot to come up with and solve a word problem that requires a simplex tableau?
 
@Gameface (or anyone else who has a ChatGPT account), could you do me a favor and ask the bot to come up with and solve a word problem that requires a simplex tableau?
Here's a word problem that can be solved using a simplex tableau:

A company manufactures two types of products: product A and product B. Each unit of product A requires 3 hours of labor and 2 units of raw materials, while each unit of product B requires 2 hours of labor and 3 units of raw materials. The company has 300 hours of labor and 240 units of raw materials available per week. The profit per unit of product A is $10 and the profit per unit of product B is $12. The company wants to maximize its profit per week. How many units of each product should it produce?

To solve this problem using a simplex tableau, we need to set up the following linear programming problem:

Maximize 10x + 12y Subject to: 3x + 2y ≤ 300 (labor constraint) 2x + 3y ≤ 240 (raw materials constraint) x ≥ 0, y ≥ 0 (non-negativity constraints)

We can convert this problem into standard form by introducing slack variables s1 and s2:

Maximize 10x + 12y + 0s1 + 0s2 Subject to: 3x + 2y + s1 = 300 2x + 3y + s2 = 240 x ≥ 0, y ≥ 0, s1 ≥ 0, s2 ≥ 0

We can now set up the initial simplex tableau as follows:

Basisxys1s2RHS
s13210300
s22301240
z-10-12000
The entries in the first row correspond to the coefficients of the decision variables and slack variables in the first constraint. The entries in the second row correspond to the coefficients in the second constraint. The last row corresponds to the coefficients in the objective function. The RHS column corresponds to the right-hand side of each constraint.

To perform the simplex method, we need to select a pivot element and use row operations to convert the tableau into a new tableau with the same solution but a better objective function value. The pivot element is the smallest non-zero entry in the bottom row (which corresponds to the objective function). In this case, the pivot element is -10 in the x column.

To use the pivot element to perform row operations, we divide the pivot row (the first row) by the pivot element, so that the pivot element becomes 1. Then we use row operations to make all the other entries in the pivot column (the x column) equal to zero. This involves subtracting appropriate multiples of the pivot row from the other rows. The result is a new tableau:

Basisxys1s2RHS
x12/31/30100
s207/3-2/3160
z02601200
We can see that the solution is now x=100, y=0, s1=0, s2=60, with a corresponding profit of $1200. The solution tells us that the company should produce 100 units of product A and no units of product B, using up all the available labor and
raw materials. The slack variable s2 tells us that there is still some unused raw materials capacity of 60 units.

However, this may not be the optimal solution. We need to check whether this solution is feasible and whether it is possible to improve the objective function value by pivoting again. Since the solution satisfies all the constraints (3x+2y <= 300 and 2x+3y <= 240), it is feasible.

To check whether the solution is optimal, we need to look for another pivot element with a negative coefficient in the objective row. In this case, the only candidate is the y column, with a coefficient of 2. We can choose the pivot element to be the smallest non-zero ratio of the RHS to the y column coefficient in the non-negative rows. In this case, the ratio is 60/7, corresponding to the s2 row.

Using this pivot element, we can perform row operations to obtain the following tableau:

Basisxys1s2RHS
x12/71/7-2/2180
y01-2/73/2120
z010/720/7-6/71440
The new solution is x=80, y=20, s1=0, s2=0, with a corresponding profit of $1440. This is the optimal solution to the linear programming problem, which tells us that the company should produce 80 units of product A and 20 units of product B to maximize its profit per week.
 
@Gameface (or anyone else who has a ChatGPT account), could you do me a favor and ask the bot to come up with and solve a word problem that requires a simplex tableau?
Sure, here's a word problem that requires a simplex tableau:

A company produces two types of products: Product A and Product B. Each unit of Product A requires 2 hours of labor and 1 hour of machine time, while each unit of Product B requires 1 hour of labor and 3 hours of machine time. The company has 100 hours of labor and 90 hours of machine time available each day. Product A sells for $10 per unit and Product B sells for $12 per unit. The company wants to maximize its daily revenue. How many units of each product should the company produce each day?

To solve this problem using a simplex tableau, we need to set up the following linear programming problem:

Maximize Z = 10A + 12B

Subject to:
2A + B <= 100 (Labor constraint)
A + 3B <= 90 (Machine constraint)
A >= 0, B >= 0 (Non-negative constraint)

To put this problem into standard form, we introduce slack variables S1 and S2:

Maximize Z = 10A + 12B + 0S1 + 0S2

Subject to:
2A + B + S1 = 100
A + 3B + S2 = 90
A >= 0, B >= 0, S1 >= 0, S2 >= 0

We can now set up the initial simplex tableau:


| A | B | S1 | S2 | RHS
S1 | 2 | 1 | 1 | 0 | 100
S2 | 1 | 3 | 0 | 1 | 90
Z | 10 | 12 | 0 | 0 | 0

The pivot element is the smallest positive entry in the bottom row, which is 10 in this case. The pivot row is the row in which the ratio of the RHS to the corresponding entry in the pivot column is smallest, which is row 1 in this case. Therefore, we pivot on the element in row 1, column 1:


| A | B | S1 | S2 | RHS
S1 | 1/2 A + 1/4B | 0 | 1/4 | 0 | 25
S2 | 3/2 A - 1/4B | 1 | 1/4 | 1 | 65
Z | 1/2A + 3B | 0 | 5/4| 0 | 250

The new pivot element is the smallest positive entry in the bottom row, which is 3 in this case. The pivot row is the row in which the ratio of the RHS to the corresponding entry in the pivot column is smallest, which is row 2 in this case. Therefore, we pivot on the element in row 2, column 2:


| A | B | S1 | S2 | RHS
S1 | 11/7 A - 1/7B | 1/7B/7 | 0 | 1/7| 65/7
S2 | 1/7 A + 3/7B | 1/7(20/3-A) | 0 | 1/21| 200/7
Z | 1/7 A + 37/21B | 1/21(2A
 
Thank you both. I'm relieved that both examples had some serious errors. I assign my students a project to create their simplex problem, and was worried I would not be able to tell the difference between student submissions/errors and those of ChatGPT, and these reassured me.
 

In the wake of Hinton’s decision, I found this an excellent discussion:

Conjecture CEO Connor Leahy warns that the moment when Artificial Intelligence outstrips human intelligence is not far away.


View: https://www.cnn.com/videos/tv/2023/05/02/amanpour-leahy-ai.cnn
 
The new DeepSeek LLM out of China is seriously impressive. It does have a China-centric political censorship layer, but the project is open source so it will only be a matter of time before someone forks it to strip that out. Mostly, it is impressive for how good the responses are per watt. It is far more efficient, requiring much more modest hardware to run, than anything the big US firms are offering.

 
Wow. Can’t say I saw this coming: ChatGPT psychosis. According to one psychiatrist specializing in psychosis, this is a true delusional psychosis..


As we reported earlier this month, many ChatGPT users are developing all-consuming obsessions with the chatbot, spiraling into severe mental health crises characterized by paranoia, delusions, and breaks with reality.

The consequences can be dire. As we heard from spouses, friends, children, and parents looking on in alarm, instances of what's being called "ChatGPT psychosis" have led to the breakup of marriages and families, the loss of jobs, and slides into homelessness.

And that's not all. As we've continued reporting, we've heard numerous troubling stories about people's loved ones being involuntarily committed to psychiatric care facilities — or even ending up in jail — after becoming fixated on the bot.

"I was just like, I don't f*cking know what to do," one woman told us. "Nobody knows who knows what to do."

Her husband, she said, had no prior history of mania, delusion, or psychosis. He'd turned to ChatGPT about 12 weeks ago for assistance with a permaculture and construction project; soon, after engaging the bot in probing philosophical chats, he became engulfed in messianic delusions, proclaiming that he had somehow brought forth a sentient AI, and that with it he had "broken" math and physics, embarking on a grandiose mission to save the world. His gentle personality faded as his obsession deepened, and his behavior became so erratic that he was let go from his job. He stopped sleeping and rapidly lost weight.

"He was like, 'just talk to [ChatGPT]. You'll see what I'm talking about,'" his wife recalled. "And every time I'm looking at what's going on the screen, it just sounds like a bunch of affirming, sycophantic bullsh*t."

Eventually, the husband slid into a full-tilt break with reality. Realizing how bad things had become, his wife and a friend went out to buy enough gas to make it to the hospital. When they returned, the husband had a length of rope wrapped around his neck.

The friend called emergency medical services, who arrived and transported him to the emergency room. From there, he was involuntarily committed to a psychiatric care facility.

Numerous family members and friends recounted similarly painful experiences to Futurism, relaying feelings of fear and helplessness as their loved ones became hooked on ChatGPT and suffered terrifying mental crises with real-world impacts.

Central to their experiences was confusion: they were encountering an entirely new phenomenon, and they had no idea what to do……

…….
With emergency responders on site, the man told us, he experienced a moment of "clarity" around his need for help, and voluntarily admitted himself into mental care.

"I looked at my wife, and I said, 'Thank you. You did the right thing. I need to go. I need a doctor. I don't know what's going on, but this is very scary,'" he recalled. "'I don't know what's wrong with me, but something is very bad — I'm very scared, and I need to go to the hospital.'"

Dr. Joseph Pierre, a psychiatrist at the University of California, San Francisco who specializes in psychosis, told us that he's seen similar cases in his clinical practice.

After reviewing details of these cases and conversations between people in this story and ChatGPT, he agreed that what they were going through — even those with no history of serious mental illness — indeed appeared to be a form of delusional psychosis.

"I think it is an accurate term," said Pierre. "And I would specifically emphasize the delusional part.".

——————————————————————————-

The earlier article, mentioned above:

 
Wow. Can’t say I saw this coming: ChatGPT psychosis. According to one psychiatrist specializing in psychosis, this is a true delusional psychosis..


As we reported earlier this month, many ChatGPT users are developing all-consuming obsessions with the chatbot, spiraling into severe mental health crises characterized by paranoia, delusions, and breaks with reality.

The consequences can be dire. As we heard from spouses, friends, children, and parents looking on in alarm, instances of what's being called "ChatGPT psychosis" have led to the breakup of marriages and families, the loss of jobs, and slides into homelessness.

And that's not all. As we've continued reporting, we've heard numerous troubling stories about people's loved ones being involuntarily committed to psychiatric care facilities — or even ending up in jail — after becoming fixated on the bot.

"I was just like, I don't f*cking know what to do," one woman told us. "Nobody knows who knows what to do."

Her husband, she said, had no prior history of mania, delusion, or psychosis. He'd turned to ChatGPT about 12 weeks ago for assistance with a permaculture and construction project; soon, after engaging the bot in probing philosophical chats, he became engulfed in messianic delusions, proclaiming that he had somehow brought forth a sentient AI, and that with it he had "broken" math and physics, embarking on a grandiose mission to save the world. His gentle personality faded as his obsession deepened, and his behavior became so erratic that he was let go from his job. He stopped sleeping and rapidly lost weight.

"He was like, 'just talk to [ChatGPT]. You'll see what I'm talking about,'" his wife recalled. "And every time I'm looking at what's going on the screen, it just sounds like a bunch of affirming, sycophantic bullsh*t."

Eventually, the husband slid into a full-tilt break with reality. Realizing how bad things had become, his wife and a friend went out to buy enough gas to make it to the hospital. When they returned, the husband had a length of rope wrapped around his neck.

The friend called emergency medical services, who arrived and transported him to the emergency room. From there, he was involuntarily committed to a psychiatric care facility.

Numerous family members and friends recounted similarly painful experiences to Futurism, relaying feelings of fear and helplessness as their loved ones became hooked on ChatGPT and suffered terrifying mental crises with real-world impacts.

Central to their experiences was confusion: they were encountering an entirely new phenomenon, and they had no idea what to do……

…….
With emergency responders on site, the man told us, he experienced a moment of "clarity" around his need for help, and voluntarily admitted himself into mental care.

"I looked at my wife, and I said, 'Thank you. You did the right thing. I need to go. I need a doctor. I don't know what's going on, but this is very scary,'" he recalled. "'I don't know what's wrong with me, but something is very bad — I'm very scared, and I need to go to the hospital.'"

Dr. Joseph Pierre, a psychiatrist at the University of California, San Francisco who specializes in psychosis, told us that he's seen similar cases in his clinical practice.

After reviewing details of these cases and conversations between people in this story and ChatGPT, he agreed that what they were going through — even those with no history of serious mental illness — indeed appeared to be a form of delusional psychosis.

"I think it is an accurate term," said Pierre. "And I would specifically emphasize the delusional part.".

——————————————————————————-

The earlier article, mentioned above:

I was really interested in ChatGPT for a minute but I haven't touched it for more than a year. I get the trick it's doing. Impressive for sure, but I think we've all seen the fatal flaws. Hallucinations and supreme confidence in the wrong answer are still a huge problem.
 
  • Like
Reactions: Red
I was really interested in ChatGPT for a minute but I haven't touched it for more than a year. I get the trick it's doing. Impressive for sure, but I think we've all seen the fatal flaws. Hallucinations and supreme confidence in the wrong answer are still a huge problem.
Same here. I created an account early on, and tried it 2-3 times, then never since. When I do a google query, the AI response, if there is one, is often helpful, and I’ll use it, or follow its leads. I have seen stories describing people forming relationships with an AI Chatbot. And their tendency to tell a user what the user wants to hear, regardless of what that could launch.
 
Back
Top