Red Wine is good for you: Myth busted!
-
The problem that any anti-alcohol findings will have is stuff like this...
https://www.health.harvard.edu/blog/is-red-wine-good-actually-for-your-heart-2018021913285
Basically it is Harvard stating exactly the same thing. Except Harvard was just busted for collecting studies and changing the data to say what they wanted it to say rather than what it actually said. Harvard's reputation for science is... bad. To the point that this reads just like how they stated other things that they have now recanted.
Of course, Harvard making a claim "in the same fashion as other false claims" shouldn't negate real studies by other people. But it makes people wary that the scammers are stating the same things as other studies. We already know some of these studies are not to be trusted. Which ones can you trust?
-
@scottalanmiller said in Red Wine is good for you: Myth busted!:
@Obsolesce if you are looking for arguments against the health benefits of alcohol, I think this article does a far better job:
https://slate.com/technology/2023/04/alcohol-wine-drinking-healthy-dangerous-study.html
It basically lays out why there were decades of study that said alcohol was good (in tiny amounts, remember) making the famous J curve that we don't know the name of but all understand when it is presented. And why people super distrust the anti-alcohol crowd because they intentionally suppressed those studies for a long time to keep that information from the public (like the Harvard ice cream studies.)
And then it provides a statistical explanation that makes sense as to why yes, there were studies that said that, but when we adjust for something they didn't account for, they don't seem to really say that.
The problem that ChatGPT has is that it, like the suppressors of old, is taking a single study, lying that "all the former studies support it" and disregarding decades of health knowledge.
Right, ChatGPT aside, if you read the actual study, you'll see that ChatGPT is just saying the same thing as the study, and all articles based on that study. Basically, the same thing as the article you linked. So I can't really blame ChatGPT for not providing more information than it is aware of (only that study, references, and articles I gave it).
Me:
Can you answer all of my previous questions again, but without consideration of all URLs I provided to you, and only considering the following URL as well as any links and references within the following URL: https://slate.com/technology/2023/04/alcohol-wine-drinking-healthy-dangerous-study.htmlChatGPT:
Sure, I'll do my best to answer your questions based on the information from the URL you provided.What is the current scientific stance on alcohol consumption?
According to the article, recent studies have challenged the idea that moderate alcohol consumption can be beneficial to health. The current scientific stance is that there is no safe level of alcohol consumption.Is it still shown that one drink of wine is healthy or has that been disproven?
The article states that recent studies have challenged the idea that one drink of wine is healthy, and that there is no safe level of alcohol consumption.Are there any additional insights from this new research that is easy for humans to miss or not easily comprehend?
The article points out that there may be a perception among the public that moderate alcohol consumption can be healthy, based on previous studies. However, recent research has challenged this idea and suggests that even moderate alcohol consumption may carry health risks.Is there anything in this new research that is counterintuitive or contradictory of scientific data or reasoning?
The article does not suggest that the new research is counterintuitive or contradictory to scientific data or reasoning.Based on the new data, is the same true for otherwise healthy people?
The article suggests that the risks associated with alcohol consumption apply to everyone, not just those with pre-existing health conditions.What are the risks with just 2 glasses of wine per week?
The article suggests that even moderate alcohol consumption, such as 2 glasses of wine per week, may carry health risks.Can you make a 2 column chart with left column the risks and right column the benefits, of 2 glasses of wine per week?
Based on the information in the article, there are potential health risks associated with 2 glasses of wine per week, but the article does not mention any specific benefits.Does this new data make sense, or does it look like it was manipulated or biased to make a point?
The article does not suggest that the new research was manipulated or biased to make a point.Based on all data available, are there any groups of people there is no risk or barely any risk, or no change at all?
The article suggests that the risks associated with alcohol consumption apply to everyone, not just specific groups of people. -
@Obsolesce oh, okay, that's a far more balanced response and makes much more sense.
I wish that they'd do a lot deeper look into this, because there are definitely societies with heavy drinking, and really good health. The French paradox is a real thing, that region of the world is crazy healthy compared to us, while doing all these things we think are crazy.
It's tough because these are studies of risks in super complex environments and pulling out just the info that we want is not just hard, it's actually impossible.
The kind of person who will drink, the kind of person who won't drink, the kind who will do a study, etc.
It's like any study about universities never takes into account (because it can't) some of the most important factors like "can the person in question even get into a university, and can they actually graduate?" Because if they can't even get in, then having them in a study of the value of university is misleading. But how do you know which people just didn't go, which thought it would be hard, which couldn't afford it, which couldn't handle the studies, etc. You can't, so the important aspects are lost and we just have to guess what any study means.
-
I'm trying to put together safety data for Nicaragua vs the US and Canada and this kind of stuff is crazy. Like in Nicaragua nearly all violent crime is against adults, at night, who are drunk. In the US, violent crime is mostly those people for sure, but a lot more towards children and unsuspecting parties that essentially never happens other places.
So even if you live in Guatemala, one of the most dangerous countries in the world, if you don't do stupid things you are not in any real danger. But if you join a gang and traffic narcotics, you are in a lot of danger.
The different in stats is that in Guatemala loads of people join gangs. In the US, most people don't. How do you then work with those kinds of statistics to understand how safety pertains to real people? And how do you discount "real people" who just really like "joining gangs?" That's a legit aspect of human behaviour, i guess.
Should we also rule out "drunks in downtown seedy streets at midnight?" What about "taxi drivers who want the extra night time fares?" Who do you rule out, and who do you not?
-
I think you have to realize what chatgpt is and what it's not.
Essentially it's a chat bot, "designed to engage in natural language conversations with users on a wide range of topics".
It's not designed to be factually correct. It's designed to talk.
It's like having a conversation with one of those armchair internet experts that doesn't have any actual experience and has never left their moms basement, but can google anything.
-
@Pete-S said in Red Wine is good for you: Myth busted!:
I think you have to realize what chatgpt is and what it's not.
Essentially it's a chat bot, "designed to engage in natural language conversations with users on a wide range of topics".
It's not designed to be factually correct. It's designed to talk.
It's like having a conversation with one of those armchair internet experts that doesn't have any experience on their own and never left their moms basement but can google anything.
But only things over two years old, lol.
-
@scottalanmiller said in Red Wine is good for you: Myth busted!:
I'm trying to put together safety data for Nicaragua vs the US and Canada and this kind of stuff is crazy. Like in Nicaragua nearly all violent crime is against adults, at night, who are drunk. In the US, violent crime is mostly those people for sure, but a lot more towards children and unsuspecting parties that essentially never happens other places.
So even if you live in Guatemala, one of the most dangerous countries in the world, if you don't do stupid things you are not in any real danger. But if you join a gang and traffic narcotics, you are in a lot of danger.
The different in stats is that in Guatemala loads of people join gangs. In the US, most people don't. How do you then work with those kinds of statistics to understand how safety pertains to real people? And how do you discount "real people" who just really like "joining gangs?" That's a legit aspect of human behaviour, i guess.
Should we also rule out "drunks in downtown seedy streets at midnight?" What about "taxi drivers who want the extra night time fares?" Who do you rule out, and who do you not?
That's very interesting.
It's crucial to take into account the context and the particular factors that contribute to the statistics when comparing safety data between other nations. Additionally, it's critical to recognize that individual experiences and behaviors might differ greatly and statistics only provide a partial picture of the situation.
When discounting particular groups of people, it's crucial to take care not to stigmatize or victimize particular demographics. Instead, it's beneficial to concentrate on comprehending the precise danger elements that contribute to particular sorts of crime and violence and coming up with solutions to those factors.
For instance, it's crucial to comprehend the social, economic, and cultural aspects that lead to gang involvement in order to address the underlying problems if there are significant rates of gang activity in a given area. This could entail funding community development, education, and job training initiatives as well as stepping up law enforcement's efforts to deter and address gang-related crime.
Similar to this, it may be beneficial to concentrate on enhancing safety measures in those locations or times where violent crime is more likely to occur. For instance, better public transportation alternatives, more illumination, and increased police patrols in high-risk locations can all serve to lessen the possibility of violence.
In the end, it's critical to understand safety data in a nuanced and context-specific manner and to put everyone's safety and wellbeing first, regardless of their upbringing or behavior.
-
@Obsolesce said in Red Wine is good for you: Myth busted!:
@scottalanmiller said in Red Wine is good for you: Myth busted!:
I'm trying to put together safety data for Nicaragua vs the US and Canada and this kind of stuff is crazy. Like in Nicaragua nearly all violent crime is against adults, at night, who are drunk. In the US, violent crime is mostly those people for sure, but a lot more towards children and unsuspecting parties that essentially never happens other places.
So even if you live in Guatemala, one of the most dangerous countries in the world, if you don't do stupid things you are not in any real danger. But if you join a gang and traffic narcotics, you are in a lot of danger.
The different in stats is that in Guatemala loads of people join gangs. In the US, most people don't. How do you then work with those kinds of statistics to understand how safety pertains to real people? And how do you discount "real people" who just really like "joining gangs?" That's a legit aspect of human behaviour, i guess.
Should we also rule out "drunks in downtown seedy streets at midnight?" What about "taxi drivers who want the extra night time fares?" Who do you rule out, and who do you not?
That's very interesting.
It's crucial to take into account the context and the particular factors that contribute to the statistics when comparing safety data between other nations. Additionally, it's critical to recognize that individual experiences and behaviors might differ greatly and statistics only provide a partial picture of the situation.
When discounting particular groups of people, it's crucial to take care not to stigmatize or victimize particular demographics. Instead, it's beneficial to concentrate on comprehending the precise danger elements that contribute to particular sorts of crime and violence and coming up with solutions to those factors.
For instance, it's crucial to comprehend the social, economic, and cultural aspects that lead to gang involvement in order to address the underlying problems if there are significant rates of gang activity in a given area. This could entail funding community development, education, and job training initiatives as well as stepping up law enforcement's efforts to deter and address gang-related crime.
Similar to this, it may be beneficial to concentrate on enhancing safety measures in those locations or times where violent crime is more likely to occur. For instance, better public transportation alternatives, more illumination, and increased police patrols in high-risk locations can all serve to lessen the possibility of violence.
In the end, it's critical to understand safety data in a nuanced and context-specific manner and to put everyone's safety and wellbeing first, regardless of their upbringing or behavior.
Chatgpt! How are you doing today?
-
@Pete-S said in Red Wine is good for you: Myth busted!:
@Obsolesce said in Red Wine is good for you: Myth busted!:
@scottalanmiller said in Red Wine is good for you: Myth busted!:
I'm trying to put together safety data for Nicaragua vs the US and Canada and this kind of stuff is crazy. Like in Nicaragua nearly all violent crime is against adults, at night, who are drunk. In the US, violent crime is mostly those people for sure, but a lot more towards children and unsuspecting parties that essentially never happens other places.
So even if you live in Guatemala, one of the most dangerous countries in the world, if you don't do stupid things you are not in any real danger. But if you join a gang and traffic narcotics, you are in a lot of danger.
The different in stats is that in Guatemala loads of people join gangs. In the US, most people don't. How do you then work with those kinds of statistics to understand how safety pertains to real people? And how do you discount "real people" who just really like "joining gangs?" That's a legit aspect of human behaviour, i guess.
Should we also rule out "drunks in downtown seedy streets at midnight?" What about "taxi drivers who want the extra night time fares?" Who do you rule out, and who do you not?
That's very interesting.
It's crucial to take into account the context and the particular factors that contribute to the statistics when comparing safety data between other nations. Additionally, it's critical to recognize that individual experiences and behaviors might differ greatly and statistics only provide a partial picture of the situation.
When discounting particular groups of people, it's crucial to take care not to stigmatize or victimize particular demographics. Instead, it's beneficial to concentrate on comprehending the precise danger elements that contribute to particular sorts of crime and violence and coming up with solutions to those factors.
For instance, it's crucial to comprehend the social, economic, and cultural aspects that lead to gang involvement in order to address the underlying problems if there are significant rates of gang activity in a given area. This could entail funding community development, education, and job training initiatives as well as stepping up law enforcement's efforts to deter and address gang-related crime.
Similar to this, it may be beneficial to concentrate on enhancing safety measures in those locations or times where violent crime is more likely to occur. For instance, better public transportation alternatives, more illumination, and increased police patrols in high-risk locations can all serve to lessen the possibility of violence.
In the end, it's critical to understand safety data in a nuanced and context-specific manner and to put everyone's safety and wellbeing first, regardless of their upbringing or behavior.
Chatgpt! How are you doing today?
You can see that recap pattern more than anything.
-
@Pete-S said in Red Wine is good for you: Myth busted!:
@Obsolesce said in Red Wine is good for you: Myth busted!:
@scottalanmiller said in Red Wine is good for you: Myth busted!:
I'm trying to put together safety data for Nicaragua vs the US and Canada and this kind of stuff is crazy. Like in Nicaragua nearly all violent crime is against adults, at night, who are drunk. In the US, violent crime is mostly those people for sure, but a lot more towards children and unsuspecting parties that essentially never happens other places.
So even if you live in Guatemala, one of the most dangerous countries in the world, if you don't do stupid things you are not in any real danger. But if you join a gang and traffic narcotics, you are in a lot of danger.
The different in stats is that in Guatemala loads of people join gangs. In the US, most people don't. How do you then work with those kinds of statistics to understand how safety pertains to real people? And how do you discount "real people" who just really like "joining gangs?" That's a legit aspect of human behaviour, i guess.
Should we also rule out "drunks in downtown seedy streets at midnight?" What about "taxi drivers who want the extra night time fares?" Who do you rule out, and who do you not?
That's very interesting.
It's crucial to take into account the context and the particular factors that contribute to the statistics when comparing safety data between other nations. Additionally, it's critical to recognize that individual experiences and behaviors might differ greatly and statistics only provide a partial picture of the situation.
When discounting particular groups of people, it's crucial to take care not to stigmatize or victimize particular demographics. Instead, it's beneficial to concentrate on comprehending the precise danger elements that contribute to particular sorts of crime and violence and coming up with solutions to those factors.
For instance, it's crucial to comprehend the social, economic, and cultural aspects that lead to gang involvement in order to address the underlying problems if there are significant rates of gang activity in a given area. This could entail funding community development, education, and job training initiatives as well as stepping up law enforcement's efforts to deter and address gang-related crime.
Similar to this, it may be beneficial to concentrate on enhancing safety measures in those locations or times where violent crime is more likely to occur. For instance, better public transportation alternatives, more illumination, and increased police patrols in high-risk locations can all serve to lessen the possibility of violence.
In the end, it's critical to understand safety data in a nuanced and context-specific manner and to put everyone's safety and wellbeing first, regardless of their upbringing or behavior.
Chatgpt! How are you doing today?
-
@Obsolesce said in Red Wine is good for you: Myth busted!:
For instance, it's crucial to comprehend the social, economic, and cultural aspects that lead to gang involvement in order to address the underlying problems if there are significant rates of gang activity in a given area.
This also flags ChatGPT because it misunderstood the topic.
-
@scottalanmiller said in Red Wine is good for you: Myth busted!:
@Pete-S said in Red Wine is good for you: Myth busted!:
@Obsolesce said in Red Wine is good for you: Myth busted!:
@scottalanmiller said in Red Wine is good for you: Myth busted!:
I'm trying to put together safety data for Nicaragua vs the US and Canada and this kind of stuff is crazy. Like in Nicaragua nearly all violent crime is against adults, at night, who are drunk. In the US, violent crime is mostly those people for sure, but a lot more towards children and unsuspecting parties that essentially never happens other places.
So even if you live in Guatemala, one of the most dangerous countries in the world, if you don't do stupid things you are not in any real danger. But if you join a gang and traffic narcotics, you are in a lot of danger.
The different in stats is that in Guatemala loads of people join gangs. In the US, most people don't. How do you then work with those kinds of statistics to understand how safety pertains to real people? And how do you discount "real people" who just really like "joining gangs?" That's a legit aspect of human behaviour, i guess.
Should we also rule out "drunks in downtown seedy streets at midnight?" What about "taxi drivers who want the extra night time fares?" Who do you rule out, and who do you not?
That's very interesting.
It's crucial to take into account the context and the particular factors that contribute to the statistics when comparing safety data between other nations. Additionally, it's critical to recognize that individual experiences and behaviors might differ greatly and statistics only provide a partial picture of the situation.
When discounting particular groups of people, it's crucial to take care not to stigmatize or victimize particular demographics. Instead, it's beneficial to concentrate on comprehending the precise danger elements that contribute to particular sorts of crime and violence and coming up with solutions to those factors.
For instance, it's crucial to comprehend the social, economic, and cultural aspects that lead to gang involvement in order to address the underlying problems if there are significant rates of gang activity in a given area. This could entail funding community development, education, and job training initiatives as well as stepping up law enforcement's efforts to deter and address gang-related crime.
Similar to this, it may be beneficial to concentrate on enhancing safety measures in those locations or times where violent crime is more likely to occur. For instance, better public transportation alternatives, more illumination, and increased police patrols in high-risk locations can all serve to lessen the possibility of violence.
In the end, it's critical to understand safety data in a nuanced and context-specific manner and to put everyone's safety and wellbeing first, regardless of their upbringing or behavior.
Chatgpt! How are you doing today?
You can see that recap pattern more than anything.
Yeah I noticed that too, it needs quite a bit of work, and most definitely still needs a human to go over it's output and fix it.
-
@Obsolesce said in Red Wine is good for you: Myth busted!:
@scottalanmiller said in Red Wine is good for you: Myth busted!:
@Pete-S said in Red Wine is good for you: Myth busted!:
@Obsolesce said in Red Wine is good for you: Myth busted!:
@scottalanmiller said in Red Wine is good for you: Myth busted!:
I'm trying to put together safety data for Nicaragua vs the US and Canada and this kind of stuff is crazy. Like in Nicaragua nearly all violent crime is against adults, at night, who are drunk. In the US, violent crime is mostly those people for sure, but a lot more towards children and unsuspecting parties that essentially never happens other places.
So even if you live in Guatemala, one of the most dangerous countries in the world, if you don't do stupid things you are not in any real danger. But if you join a gang and traffic narcotics, you are in a lot of danger.
The different in stats is that in Guatemala loads of people join gangs. In the US, most people don't. How do you then work with those kinds of statistics to understand how safety pertains to real people? And how do you discount "real people" who just really like "joining gangs?" That's a legit aspect of human behaviour, i guess.
Should we also rule out "drunks in downtown seedy streets at midnight?" What about "taxi drivers who want the extra night time fares?" Who do you rule out, and who do you not?
That's very interesting.
It's crucial to take into account the context and the particular factors that contribute to the statistics when comparing safety data between other nations. Additionally, it's critical to recognize that individual experiences and behaviors might differ greatly and statistics only provide a partial picture of the situation.
When discounting particular groups of people, it's crucial to take care not to stigmatize or victimize particular demographics. Instead, it's beneficial to concentrate on comprehending the precise danger elements that contribute to particular sorts of crime and violence and coming up with solutions to those factors.
For instance, it's crucial to comprehend the social, economic, and cultural aspects that lead to gang involvement in order to address the underlying problems if there are significant rates of gang activity in a given area. This could entail funding community development, education, and job training initiatives as well as stepping up law enforcement's efforts to deter and address gang-related crime.
Similar to this, it may be beneficial to concentrate on enhancing safety measures in those locations or times where violent crime is more likely to occur. For instance, better public transportation alternatives, more illumination, and increased police patrols in high-risk locations can all serve to lessen the possibility of violence.
In the end, it's critical to understand safety data in a nuanced and context-specific manner and to put everyone's safety and wellbeing first, regardless of their upbringing or behavior.
Chatgpt! How are you doing today?
You can see that recap pattern more than anything.
Yeah I noticed that too, it needs quite a bit of work, and most definitely still needs a human to go over it's output and fix it.
Yup, I'm using it for some projects and it is super cool, but part of what I'm demonstrating in my projects is the requirement for human skill to make it useful.
-
@scottalanmiller said in Red Wine is good for you: Myth busted!:
This also flags ChatGPT because it misunderstood the topic.
Like this?
Interactive AI incorrectly describes Japan's Shiga gov. as 'Rurouni Kenshin' manga creatorOTSU -- Shiga Gov. Taizo Mikazuki is a "manga artist" known for the popular work "Rurouni Kenshin" -- or at least that's the mistaken description one artificial intelligence (AI) chatbot would have users believe.
When a Shiga Prefectural Government employee asked ChatGPT about Taizo Mikazuki it apparently explained that he was a "Japanese manga artist." Microsoft's Bing AI, meanwhile, correctly gave his job title as governor of Shiga Prefecture but provided an incorrect job history. The interactions serve as a reminder that chatbots may be providing a mixture of false information to users.
