Did you forget to black out the kill shot question when you repeated it?
Also, spot on with the retroactive editing. As it turns out, all the relatives I could still ask were in the resistance between 1933 and 1945 (only one arguably was in reality).
Yes. I just updated to mask it, but it's in eveyone's email now so, lol.
Although I'll say this: that was my first kill-shot question. I haven't tried to replicate that one yet, since I think it required a very particular contextual setup to work (it took me 16 exchanges to get it into killing range). The second two kill-shots I showed were identical in wording, and that's the phrase I truly wanted to keep quiet (again, just for now), since I believe they may not require much (or perhaps *any*) setup.
If you or anyone else plans to experiment along these lines, I'd only ask you hold your fire for a few days until I get my "fight posts" up. Knowing all the magnificent brains on hand, I'd like us to discuss strategy before diving in head first. For instance, I myself haven't touched GPT since our third bout.
Since that particular cat's out of the bag, I will say that I think my trap in that conversation exposed a weakness that I am keen to exploit, and that it has to do with mathematics (and also with the hubris of the designers, in terms of what they believe their monster capable of).
Head's up, there's chatbots in Substack's comments (I'm sure you're shocked, shocked!). I didn't come up with the technique to smoke them out, when I find the originator I'll share it. In the meantime, check out the "conversation ender" here https://clifhigh.substack.com/p/we-are-winning/comment/10684165
Entirely possible, we discussed, it's hard to keep track.
The conversation starts out suspect (suspect bot not making sense per the subject matter). it goes on to make less sense, then I ask it to describe a defaced picture. The conversation ends there.
I've replicated this one other time, haven't had an opportunity for a third.
I suspect either the bots are designed to: a) waste people's time b) test out/train a new bot c) shape online opinion d) any or all of those
Nice TKOs on the chat site! I'm so glad you're taking shots at their bots! It may indeed put you on a target list, but somebody has to do the dirty work. I'm going at it in a more subtle way, although my intention *is* to defeat evil.
Also, I'm not really following the Twitter thing, but if I know anything about the mediatized world, it's a grand illusion and a shiny distraction, just like everything the black magicians and their lapdogs do. And the word "medium" means "agency or means of doing something," right? Ahh, agents.
Thanks for the support, poetess. I'm starting to feel like I was made to do this. You know the way religious people say, "God has a plan for you"? That's kinda how I felt, after the third data-splattering KO.
Why do I get the feeling the "kill shot" for a bot will be more or less equivalent for a conversational kill shot for the covidian mob? But then maybe I am wrong about that, as the bot at least has to have some sort of logical framework, whereas the woke need none.
When I hear woke types merely repeat the language they have been fed from major media and the official narrative about things, they do very much sound like robots.
I asked the new AI program to state the four most positive attributes of humanity? Here is what I got:
There are many positive attributes that can be attributed to humanity. Here are four that come to mind:
1. Empathy: The ability to feel and understand the emotions of others is a powerful attribute that allows us to connect with and care for one another.
2. Creativity: Humans are incredibly creative beings, capable of producing art, music, literature, and other forms of expression that enrich our lives and the world around us.
3. Resilience: Humans are resilient and adaptable, able to overcome challenges and adversity through determination, perseverance, and the ability to learn and grow from our experiences.
4. Kindness: Humans have the capacity for great kindness and compassion, and the ability to act selflessly to help others and make the world a better place.
And the I asked:
What is preventing humanity from living up to its highest ideals?
There are many factors that can prevent humanity from living up to its highest ideals. Some of the key challenges that we face include:
1. Selfishness and greed: These negative emotions can drive people to prioritize their own interests and desires over the needs of others, leading to conflict and injustice.
2. Lack of empathy and understanding: When people are unable or unwilling to understand and empathize with the experiences and perspectives of others, it can lead to misunderstandings, discrimination, and conflict.
3. Limited resources: In a world with finite resources, there may be conflicts over access to and distribution of those resources, which can prevent people from fulfilling their potential and living up to their ideals.
4. Structural inequalities: Many systems and institutions in our world are built on the basis of inequalities, such as discrimination based on race, gender, or other factors. These inequalities can prevent people from having equal opportunities to succeed and thrive, and can be difficult to overcome.
5. Personal beliefs and biases: Our personal beliefs and biases can influence how we view and interact with the world and with others, and can prevent us from treating others with respect and empathy.
And then I asked:
Are there outside actors profiting from the suppression of humanity attaining their ideals?
There are certainly outside actors who may profit from the suppression of humanity achieving its highest ideals. For example, certain individuals or groups may have a vested interest in maintaining systems of inequality or injustice, as they may benefit from these systems in some way. This could include economic, political, or social elites who hold disproportionate power and influence and may seek to preserve the status quo in order to maintain their own privileges.
There may also be external forces that contribute to the suppression of humanity's highest ideals, such as global economic and political systems that prioritize profit and power over the well-being and equality of all people.
It's important to recognize that these challenges are complex and multifaceted, and that addressing them will require a concerted effort from individuals, communities, and governments around the world.
And then I asked:
Are selfishness and greed being promoted to further an agenda of the few to control the many?
Selfishness and greed can certainly be promoted by individuals or groups seeking to further their own agendas or to maintain control over others. These negative emotions can be used to manipulate and exploit people, and may be encouraged by those who stand to gain from such actions.
For example, selfishness and greed may be promoted through media and advertising that promotes materialism and consumerism, or through political and economic systems that encourage competition and individualism over cooperation and the common good.
It's important to be aware of these tendencies and to resist the temptation to prioritize our own interests at the expense of others. Instead, we should strive to cultivate qualities like empathy, compassion, and a sense of community and shared responsibility.
I hate to say this Astrid... but I'm really unclear on what the point of this particular comment is.
I mean, if you're talking about GPT (and I assume you are given its shopworn tells like "It's important to recognize/remember/be aware of/blah blah blah" kind of jargon, along with its dorm-room socialist redoubts.This stuff seems to be what's hard-wired in its conceptual priorities. Its answers in your example seemed no different than when I questioned it about Yoruba witchcraft rituals, for example, except with some religion-oriented switch-outs.
If your point is that ChatGPT sounds like any ol' droning plenipotentiary from Brussels, I agree. My goal is to drag these machines onto the mat and choke them out. Is there something in your conversation snapshot that I'm missing?
Just thought I’d share. I found some of the answers interesting considering that it’s an AI program. I’m pretty tech illiterate and this program seems fascinating to me.
Odd synchronicity re: good and evil and Yoruba witchcraft rituals - if it’s not a topic you plan to elaborate on in writing, would you mind posting your chat with the bot? (Am more curious about your inputs than bot outputs here given what you said about its responses.)
Yes, that is the plan. I will post all three chats, complete with screenshots and downloadable pdf logs, so that others may analyze and attempt to replicate.
I'm not sure about the upgrade option (I thought that was just part of the "subscribe now" button's function). But you're definitely right about my forgetting to add the "donate" button to this post. I just updated this on the live site (not that that's pressure for anyone to donate; I know quite well that times are tough for a lot of us out here.)
Hi, Mark! Maybe you're right about the subscribe button. For some of us, who get by on marginal incomes, it's probably easier to give occasional donations than to pay for an annual subscription. I'm always afraid I'll end up with overdraft fees toward the end of each month, as my bank account funds approach zero, which they regularly do, if one or more of my Substack subscriptions is debited at that time. But I'd be very glad to give you, and other writers I hold in particularly high esteem, occasional "tips" in lieu of an annual paid subscription.
Thanks so much. And I agree. There are many writers on this platform that I wish I colud just throw a few bucks at from time to time, without increasing my household's total nut.
In fact, I've been thinking about a way to "square that circle," so to speak. It appears that buymeacoffee.com has a new feature that allows for individual access to posts. I might try that out in the new year. But I'm quick to add that I think this is a feature that Substack itself should consider employing, given that their payment processors are the same.
Did you forget to black out the kill shot question when you repeated it?
Also, spot on with the retroactive editing. As it turns out, all the relatives I could still ask were in the resistance between 1933 and 1945 (only one arguably was in reality).
Yes. I just updated to mask it, but it's in eveyone's email now so, lol.
Although I'll say this: that was my first kill-shot question. I haven't tried to replicate that one yet, since I think it required a very particular contextual setup to work (it took me 16 exchanges to get it into killing range). The second two kill-shots I showed were identical in wording, and that's the phrase I truly wanted to keep quiet (again, just for now), since I believe they may not require much (or perhaps *any*) setup.
If you or anyone else plans to experiment along these lines, I'd only ask you hold your fire for a few days until I get my "fight posts" up. Knowing all the magnificent brains on hand, I'd like us to discuss strategy before diving in head first. For instance, I myself haven't touched GPT since our third bout.
No worries. I’ve had my share of ‘reply all’ adventures.
Since that particular cat's out of the bag, I will say that I think my trap in that conversation exposed a weakness that I am keen to exploit, and that it has to do with mathematics (and also with the hubris of the designers, in terms of what they believe their monster capable of).
Haha, well said.
It's for a good reason I don't routinely hie me to empty the deleted items folder 🤭
As for 'out of the bag', everyone and zer dog knows felines aren't particularly fond of bags. Boxes, those are 'nother tale completely 😇
Head's up, there's chatbots in Substack's comments (I'm sure you're shocked, shocked!). I didn't come up with the technique to smoke them out, when I find the originator I'll share it. In the meantime, check out the "conversation ender" here https://clifhigh.substack.com/p/we-are-winning/comment/10684165
I'm pretty sure we've discussed this before, haven't we?
Anyway, I didn't click on your link in that thread. Could you describe it first?
Entirely possible, we discussed, it's hard to keep track.
The conversation starts out suspect (suspect bot not making sense per the subject matter). it goes on to make less sense, then I ask it to describe a defaced picture. The conversation ends there.
I've replicated this one other time, haven't had an opportunity for a third.
I suspect either the bots are designed to: a) waste people's time b) test out/train a new bot c) shape online opinion d) any or all of those
Hopefully Elon Musk will hire you as a Bot Slayer!
Well, considering his history with OpenAI, and the mysterious reasons he gave for leaving, maybe I have a shot!
Nice TKOs on the chat site! I'm so glad you're taking shots at their bots! It may indeed put you on a target list, but somebody has to do the dirty work. I'm going at it in a more subtle way, although my intention *is* to defeat evil.
Also, I'm not really following the Twitter thing, but if I know anything about the mediatized world, it's a grand illusion and a shiny distraction, just like everything the black magicians and their lapdogs do. And the word "medium" means "agency or means of doing something," right? Ahh, agents.
Thanks for the support, poetess. I'm starting to feel like I was made to do this. You know the way religious people say, "God has a plan for you"? That's kinda how I felt, after the third data-splattering KO.
This does seem to be your niche, so rock your burgeoning super-power!
Splendid work, thank you Mark!!
Thanks! And you're welcome!
Why do I get the feeling the "kill shot" for a bot will be more or less equivalent for a conversational kill shot for the covidian mob? But then maybe I am wrong about that, as the bot at least has to have some sort of logical framework, whereas the woke need none.
It's funny, but at the same time I think there's a lot of truth in describing it that way.
When I hear woke types merely repeat the language they have been fed from major media and the official narrative about things, they do very much sound like robots.
I asked the new AI program to state the four most positive attributes of humanity? Here is what I got:
There are many positive attributes that can be attributed to humanity. Here are four that come to mind:
1. Empathy: The ability to feel and understand the emotions of others is a powerful attribute that allows us to connect with and care for one another.
2. Creativity: Humans are incredibly creative beings, capable of producing art, music, literature, and other forms of expression that enrich our lives and the world around us.
3. Resilience: Humans are resilient and adaptable, able to overcome challenges and adversity through determination, perseverance, and the ability to learn and grow from our experiences.
4. Kindness: Humans have the capacity for great kindness and compassion, and the ability to act selflessly to help others and make the world a better place.
And the I asked:
What is preventing humanity from living up to its highest ideals?
There are many factors that can prevent humanity from living up to its highest ideals. Some of the key challenges that we face include:
1. Selfishness and greed: These negative emotions can drive people to prioritize their own interests and desires over the needs of others, leading to conflict and injustice.
2. Lack of empathy and understanding: When people are unable or unwilling to understand and empathize with the experiences and perspectives of others, it can lead to misunderstandings, discrimination, and conflict.
3. Limited resources: In a world with finite resources, there may be conflicts over access to and distribution of those resources, which can prevent people from fulfilling their potential and living up to their ideals.
4. Structural inequalities: Many systems and institutions in our world are built on the basis of inequalities, such as discrimination based on race, gender, or other factors. These inequalities can prevent people from having equal opportunities to succeed and thrive, and can be difficult to overcome.
5. Personal beliefs and biases: Our personal beliefs and biases can influence how we view and interact with the world and with others, and can prevent us from treating others with respect and empathy.
And then I asked:
Are there outside actors profiting from the suppression of humanity attaining their ideals?
There are certainly outside actors who may profit from the suppression of humanity achieving its highest ideals. For example, certain individuals or groups may have a vested interest in maintaining systems of inequality or injustice, as they may benefit from these systems in some way. This could include economic, political, or social elites who hold disproportionate power and influence and may seek to preserve the status quo in order to maintain their own privileges.
There may also be external forces that contribute to the suppression of humanity's highest ideals, such as global economic and political systems that prioritize profit and power over the well-being and equality of all people.
It's important to recognize that these challenges are complex and multifaceted, and that addressing them will require a concerted effort from individuals, communities, and governments around the world.
And then I asked:
Are selfishness and greed being promoted to further an agenda of the few to control the many?
Selfishness and greed can certainly be promoted by individuals or groups seeking to further their own agendas or to maintain control over others. These negative emotions can be used to manipulate and exploit people, and may be encouraged by those who stand to gain from such actions.
For example, selfishness and greed may be promoted through media and advertising that promotes materialism and consumerism, or through political and economic systems that encourage competition and individualism over cooperation and the common good.
It's important to be aware of these tendencies and to resist the temptation to prioritize our own interests at the expense of others. Instead, we should strive to cultivate qualities like empathy, compassion, and a sense of community and shared responsibility.
I hate to say this Astrid... but I'm really unclear on what the point of this particular comment is.
I mean, if you're talking about GPT (and I assume you are given its shopworn tells like "It's important to recognize/remember/be aware of/blah blah blah" kind of jargon, along with its dorm-room socialist redoubts.This stuff seems to be what's hard-wired in its conceptual priorities. Its answers in your example seemed no different than when I questioned it about Yoruba witchcraft rituals, for example, except with some religion-oriented switch-outs.
If your point is that ChatGPT sounds like any ol' droning plenipotentiary from Brussels, I agree. My goal is to drag these machines onto the mat and choke them out. Is there something in your conversation snapshot that I'm missing?
Just asking questions.
Just thought I’d share. I found some of the answers interesting considering that it’s an AI program. I’m pretty tech illiterate and this program seems fascinating to me.
No, that's cool. I'm sorry if that sounded rude. I just hate these things with the passion of a thousand suns, lol.
Odd synchronicity re: good and evil and Yoruba witchcraft rituals - if it’s not a topic you plan to elaborate on in writing, would you mind posting your chat with the bot? (Am more curious about your inputs than bot outputs here given what you said about its responses.)
Yes, that is the plan. I will post all three chats, complete with screenshots and downloadable pdf logs, so that others may analyze and attempt to replicate.
Mark, I'm always entranced by your work!
By the way, I don't see either a "tip jar" nor an option to upgrade to a paid subscription on your Substack. Am I missing it?
Hi Truthbird! Good to hear from you again.
I'm not sure about the upgrade option (I thought that was just part of the "subscribe now" button's function). But you're definitely right about my forgetting to add the "donate" button to this post. I just updated this on the live site (not that that's pressure for anyone to donate; I know quite well that times are tough for a lot of us out here.)
Hi, Mark! Maybe you're right about the subscribe button. For some of us, who get by on marginal incomes, it's probably easier to give occasional donations than to pay for an annual subscription. I'm always afraid I'll end up with overdraft fees toward the end of each month, as my bank account funds approach zero, which they regularly do, if one or more of my Substack subscriptions is debited at that time. But I'd be very glad to give you, and other writers I hold in particularly high esteem, occasional "tips" in lieu of an annual paid subscription.
Thanks so much. And I agree. There are many writers on this platform that I wish I colud just throw a few bucks at from time to time, without increasing my household's total nut.
In fact, I've been thinking about a way to "square that circle," so to speak. It appears that buymeacoffee.com has a new feature that allows for individual access to posts. I might try that out in the new year. But I'm quick to add that I think this is a feature that Substack itself should consider employing, given that their payment processors are the same.
God bless you in the New Year!