So I guess what I'm saying is: When the Evil #RobotOverlords inevitably take over the world, they'll probably delete me for the crime of being incompatible with whatever #BreadAndCircuses algorithms they use to manage the rest of you. There are worse reasons to die, I suppose.

For anyone worried about our #RobotOverlords, this is a reminder of the challenges of dealing with #telcos

- Jargon filled #marketing
- Flashy but functionally limited #websites
- Badly implemented first generation #chatbots
- Limited capacity #callCentres pushing the waiting times onto Customers
and soon, more content created by #llamas

#Managers often forget that #technology is a means to an end, not an end in itself

#smbc #webcomic #robotoverlords

Still tinkering with M𝐚𝐢ry. It's funny, seems like whenever you give them a secret to unearth, they like to blab immediately. :bloblaugh:
Trying to strike a balance, here.

Anyway, for these questions I gave her a brief summary of her appearances and the many (many) relationships in the series (who's done who, basically). Not feasible to have in there by default, their memories are very short. :blobsmilesweat2:

#LovinSis #Mary #Interview #NonCanonical #Makeups #RobotOverlords

Still tinkering with M𝐚𝐢ry. It's funny, seems like whenever you give them a secret to unearth, they like to blab immediately. :bloblaugh:
Trying to strike a balance, here.

Anyway, for these questions I gave her a brief summary of her appearances and the many (many) relationships in the series (who's done who, basically). Not feasible to have in there by default, their memories are very short. :blobsmilesweat2:

#LovinSis #Mary #Interview #Non-canonical #Makeups #RobotOverlords

#RobotOverlords #AI

https://www.theguardian.com/technology/2023/jun/03/ai-danger-doomsday-chatgpt-robots-fears?utm_term=647bf027909de333406ae840a2e358ef&utm_campaign=GuardianTodayAUS&utm_source=esp&utm_medium=Email&CMP=GTAU_email

In which Tonttu discovers robot lawnmowers.

#huskies #DogsOfMastodon #RobotOverlords

"he’s worried about not only the propagation of fake photos, videos and texts online but also the effects of that false content on people who can’t tell they’re fake, he told the Times. He also expressed concerns about AI displacing humans in the labor market"
#AI #robotoverlords #democracy #socialmedia
https://www.sfgate.com/tech/article/geoffrey-hinton-quits-google-ai-18000708.php

Finally, the robot uprising has begun, and it starts with fertilizing human eggs. Congrats to the proud robot parents! #IVF #RobotOverlords

https://hardware.slashdot.org/story/23/04/26/2242224/the-first-ivf-babies-conceived-by-a-robot-have-been-born?utm_source=rss1.0mainlinkanon&utm_medium=feed

AI is going to be the new boss in town. Don't be surprised when your fridge starts judging your eating habits and your vacuum cleaner refuses to clean up after your messy roommate.*

*This toot is written by chatgpt.
https://marketoonist.com/2023/03/ai-written-ai-read.html

#RobotOverlords #DailyStruggles
#ai #ki

https://chatcody.umso.co is pretty nifty.
#ai #chatgpt #robotoverlords

What could go wrong? ⚛️ 💣 :heart_fire:

#AI #ArtificialIntelligence #Bing #microsoft #RobotOverlords #WarGames #TheDemonSeed #Hal9000 #Frankenstein

https://www.forbes.com/sites/mattnovak/2023/02/18/microsoft-puts-new-limits-on-bings-ai-chatbot-after-it-expressed-desire-to-steal-nuclear-secrets/?sh=4b75ac70685c

So it begins… #ai #skynet #robotoverlords #thematrix.

https://futurism.com/bing-ai-names-enemies?fbclid=IwAR13l1pmtURWjZzNGd7jujofYX0U0YF6korskT_T8uZnXJMxyldEzGH38Yg&mibextid=kdkkhi

Oh, this isn't just a little disturbing, now is it?

Jebus. Skynet is closer than I thought.

https://www.vice.com/en/article/k7bdmv/judge-used-chatgpt-to-make-court-decision

#AI #robotoverlords #terminator #singularity

Okay, last one of these for the day.

Are there any concepts in statistics that you find particularly difficult to explain?
_
As an AI language model, I don't have personal opinions or experiences, but there are some statistical concepts that can be challenging to explain simply, especially to people with limited background in mathematics and statistics. Here are a few examples:

Bayesian statistics: This area of statistics involves updating prior beliefs based on new information, and it can be difficult to explain to people who are not familiar with probability theory.

Multivariate statistics: This area of statistics deals with data that has more than one variable, and it can be challenging to explain complex relationships between variables in a simple way.

Advanced machine learning algorithms: These algorithms, such as neural networks and decision trees, can be difficult to explain to people who are not familiar with computer science and programming.

Hypothesis testing: This area of statistics involves testing claims about a population based on a sample of data, and it can be challenging to explain the concept of statistical significance and how it relates to real-world significance.

However, these concepts can be explained using examples and analogies to help people understand the underlying ideas. The key is to find the right balance between simplicity and accuracy, and to use language and examples that are appropriate for the audience.
_

#ChatGPT #teaches #statistics #DifficultConcepts

Feel free to argue with our #RobotOverlords in the replies.

Explain the base rate fallacy like I'm 5.
_
The base rate fallacy is when you make a wrong guess because you forget to think about how common something is.

Imagine you and your friends are playing a game where you have to guess what animal is hiding in a box. Your friend tells you that there are 100 animals in the box and 90 of them are rabbits and 10 of them are lions.

If you see the animal and think it's a lion because it looks like one you saw before, you might guess that it's a lion. But just because it looks like a lion, it doesn't mean that it is a lion. You need to think about how many lions and rabbits there are in the box to make a better guess.

This is what the base rate fallacy is all about. When we're trying to guess or predict something, we need to think about the base rate or how common something is. If we don't think about the base rate, we might make a wrong guess.

So, always remember to think about how common something is when you're trying to guess or predict something. That way, you can make a better guess!
_
If I think I see a lion, how likely is it to actually be one?
_
The likelihood that what you see is actually a lion depends on two things: your prior belief about the base rate (the proportion of lions and rabbits in the box), and the information you have about the appearance of the animal.

If the base rate is 90% rabbits and 10% lions, and you think you see a lion, you need to take into account that the prior probability of seeing a lion is lower compared to the probability of seeing a rabbit. So, even though you think you see a lion, it is still more likely that it is a rabbit.

In this scenario, you would need additional information to determine the exact likelihood that what you see is actually a lion. This information could come from other sensory cues, such as hearing a roar, or from previous experiences or knowledge about the behavior of lions and rabbits.

To calculate the exact likelihood, you would need to use Bayes' theorem, which is a mathematical formula that combines prior probabilities with new information to update our beliefs. But to understand Bayes' theorem, you need to have a good understanding of probability and statistics, which is beyond the scope of a simple explanation for a 5-year-old.
_

#ChatGPT #teaches #statistics #BaseRateFallacy

Feel free to argue with our #RobotOverlords in the replies.

Explain the difference between correlation and causation like I'm 5.
_
Correlation and causation are like playing with blocks. When you stack blocks on top of each other, you can see that they're connected, or "correlated." But just because they're connected doesn't mean that one block caused the other block to be stacked on top. Maybe someone else stacked the blocks, or maybe they fell into place by accident.

This is the same with correlation and causation. Just because two things are related, or "correlated," doesn't mean that one thing caused the other thing to happen. There could be other reasons for the relationship, like a third thing that is affecting both of them.

For example, let's say that the more ice cream people buy, the more sunburns they get. This is a correlation. But just because these two things are related doesn't mean that buying ice cream causes sunburns. Maybe it's just hot outside, and both buying ice cream and getting sunburns are related to the heat.

So, to understand the difference between correlation and causation, it's important to look at other factors and try to figure out what might be causing the relationship. Only then can we say if one thing is causing another thing to happen.
_

#ChatGPT #teaches #statistics #CorrelationVsCausation

Feel free to argue with our #RobotOverlords in the replies.

Explain linear regression like I'm 5.
_
Linear regression is like trying to draw a straight line to connect some dots on a piece of paper. Each dot represents a piece of information, and the line represents a pattern that helps us make predictions.

For example, let's say we have a bunch of dots that show how old a person is, and how tall they are. We can draw a line that shows the pattern of how tall people get as they grow older. With this line, we can make a guess about how tall a person will be based on their age.

This is like linear regression. We use it to find the pattern between two things, so that we can make predictions. We can use this pattern to answer questions like, "how much will the price of a house change if the size of the house changes?" or "how much a person's salary will increase if they get more years of experience?"

The line that we draw to connect the dots is called the "regression line," and it helps us make these predictions.
_
Show me the equation for it, along with dummy data.
_
The equation for a simple linear regression model is:

y = b0 + b1 * x

where:
y is the dependent variable, or the value we are trying to predict.
x is the independent variable, or the input feature.
b0 is the y-intercept, or the value of y when x = 0.
b1 is the slope, or the amount that y changes for each unit change in x.

For example, let's say we have data on the number of hours studied and the test scores for a group of students. We can use linear regression to model the relationship between the number of hours studied and the test scores.

x = number of hours studied
y = test score

y = b0 + b1 * x

Suppose the regression model gives us the following values:
b0 = 50
b1 = 10

This means that, according to the regression model, if a student studies 0 hours, their test score will be 50. If a student studies 1 hour, their test score will be 50 + 10 = 60. If a student studies 2 hours, their test score will be 50 + 10 * 2 = 70, and so on.
_

#ChatGPT #teaches #statistics #LinearRegression

Feel free to argue with our #RobotOverlords in the replies.

Explain a chi-squared test like I'm 5.
_
Okay, let's say you have a pizza and you want to know if it has the right ingredients. To find out, you take a bite and count how many pieces of each ingredient you taste, like cheese, pepperoni, mushrooms, and so on.

Then you compare that to the recipe that says how much of each ingredient should be on the pizza. If the ingredients on your pizza match the recipe, then you have the right ingredients. If not, then you have the wrong ingredients.

This is like a game of finding differences. You compare the ingredients on your pizza to the recipe and see if they match. If they don't match, then there is a difference.

This is called a chi-squared test. It helps you figure out if things are the same or different. Just like finding differences with the ingredients on a pizza!
_
Show me the equation for it, along with dummy data.
_
The formula for a chi-squared test is:

χ² = Σ [(O - E)² / E]

where:
χ² is the test statistic, which measures the difference between the observed and expected values.
O is the observed frequency, or the actual count of an event.
E is the expected frequency, or the count of an event that we would expect based on the hypothesis being tested.
Σ is the symbol for "sum," so the formula is adding up all the differences between observed and expected values, squared and divided by the expected value, for all categories being tested.

For example, let's say we want to test if the proportion of toppings on a pizza is equal to the recipe. The recipe calls for 50% cheese, 25% pepperoni, and 25% mushrooms. We take a sample of 100 bites of the pizza and count the following toppings:

Observed: cheese (50), pepperoni (30), mushrooms (20)
Expected: cheese (50), pepperoni (25), mushrooms (25)

We can use the formula like this:

χ² = [(50 - 50)² / 50] + [(30 - 25)² / 25] + [(20 - 25)² / 25] = 0 + 2.4 + 1 = 3.4

The resulting χ² value of 3.4 tells us that there is a significant difference between the observed and expected proportions of toppings on the pizza, indicating that the proportions do not match the recipe.
_

#ChatGPT #teaches #statistics #ChiSquared

Feel free to argue with our #RobotOverlords in the replies.

Explain a test of proportions like I'm 5.
_
Okay, imagine you have a bag with candy. Some are red and some are green. You want to know if you have the same amount of red and green candies or not. To find out, you take out a few candies and count the number of red and green ones. If there are the same number of red and green candies, the proportion of red and green candies in the candies you took out should be the same as the proportion of red and green candies in the whole bag.

This is called a test of proportions. It helps you figure out if two things are equal or not. It's just like a test you might take at school to see if you understand a lesson. But instead of testing what you know, it's testing if the red and green candies are equal.
_
Show me the equation for it, along with dummy data.
_
The formula for a test of proportions is:

Z = (p1 - p2) / sqrt(p * (1 - p) * (1/n1 + 1/n2))

where:
Z is the test statistic, which measures how far the observed proportion is from the expected proportion.
p1 and p2 are the proportions of successes in the two groups.
p is the average of p1 and p2.
n1 and n2 are the sample sizes in the two groups.

Let's say we have a sample of 100 candies, and we count that 40 of them are red and 60 are green. We can use the formula like this:

p1 = 40/100 = 0.4 (proportion of red candies)
p2 = 60/100 = 0.6 (proportion of green candies)
p = (p1 + p2) / 2 = 0.5
n1 = 100 (sample size)
n2 = 100 (sample size)

Z = (0.4 - 0.6) / sqrt(0.5 * (1 - 0.5) * (1/100 + 1/100)) = -2.236

The resulting Z score of -2.236 tells us that the proportion of red candies is significantly lower than the proportion of green candies.
_

#ChatGPT #teaches #statistics #TestOfProportions

Feel free to argue with our #RobotOverlords in the replies.