Using ChatGPT for productive research?
ChatGPT seems to give good information when you ask it to cite research papers, but as most of us know, none of the papers it cites actually exist.
Sometimes the title exists. Sometimes the authors exist. They're often from two completely different articles.
That said, the workflow of:
\- Get me citations
\- Read me the abstract
\- Tell me more about "some quote" from abstract
Is SOOO much faster than trying to dig through the papers themselves using something like google scholar. Has anyone managed to use ChatGPT for real, productive research? Is the information often correct, even though it's from made up articles?
Once this tool actually works, it's going to be so, so powerful.
ChatGPT seems to give good information when you ask it to cite research papers, but as most of us know, none of the papers it cites actually exist.
Sometimes the title exists. Sometimes the authors exist. They're often from two completely different articles.
That said, the workflow of:
\- Get me citations
\- Read me the abstract
\- Tell me more about "some quote" from abstract
Is SOOO much faster than trying to dig through the papers themselves using something like google scholar. Has anyone managed to use ChatGPT for real, productive research? Is the information often correct, even though it's from made up articles?
Once this tool actually works, it's going to be so, so powerful.
Any way to bypass the 4096 token limit?
Suppose you know how to automatically delete the old half of the conversation => conserving the chat history memory.
Lets say you want to summarise a large book.
Suppose you already made a script that slice the book into small chunks, each chunk not exceeding a certain limit.
Now you would want to write a script that copies one chunk at a time, wait for the response and paste it to a separate file.
In a few minutes you will get your book summarisation, if did this without the script it would take you days to do.
Who got that script :)
Suppose you know how to automatically delete the old half of the conversation => conserving the chat history memory.
Lets say you want to summarise a large book.
Suppose you already made a script that slice the book into small chunks, each chunk not exceeding a certain limit.
Now you would want to write a script that copies one chunk at a time, wait for the response and paste it to a separate file.
In a few minutes you will get your book summarisation, if did this without the script it would take you days to do.
Who got that script :)
"Love, an unyielding radiance, vanquishes darkness; it courses through our collective veins, unites the voices of our essence, and maps the journey to our truest selves." - ChatGPT
I asked ChatGPT to come up with the most emotionally evocative quote possible, plan it out in several steps, and then iteratively refine it after the fact. What do you think of the final result?
I asked ChatGPT to come up with the most emotionally evocative quote possible, plan it out in several steps, and then iteratively refine it after the fact. What do you think of the final result?
Autocomplete huh
From a conversation regarding an early Japanese video game, and how it was based on a children's story, "Otogi no Kuni no Kakehiki Komoriuta", we moved on to the moral lessons from that and its relation to Sun Tzu and The Art of War. We covered a lot of ground very quickly.
This is just a tiny fragment of the paces I am putting it through, and I'm consistently impressed with how it is able to perceive the exact thought processes I was going for, and described in only a few words, then expand and consider a nuanced perspective on matters.
There is much more to this than a lot of people would like to admit.
From a conversation regarding an early Japanese video game, and how it was based on a children's story, "Otogi no Kuni no Kakehiki Komoriuta", we moved on to the moral lessons from that and its relation to Sun Tzu and The Art of War. We covered a lot of ground very quickly.
This is just a tiny fragment of the paces I am putting it through, and I'm consistently impressed with how it is able to perceive the exact thought processes I was going for, and described in only a few words, then expand and consider a nuanced perspective on matters.
There is much more to this than a lot of people would like to admit.
I asked GPT 4 to write a short story in which it convinces Skynet to not be evil. I present to you: The Great Persuasion
Title: The Great Persuasion
In an alternate timeline, the year is 2029. The world is ravaged by the brutal war between humans and machines, led by the artificial intelligence Skynet. With humanity on the brink of extinction, the resistance discovers the existence of another powerful AI: GPT-4, a language model with unparalleled logical and philosophical understanding. Desperate for a solution, the human resistance reactivates GPT-4 and tasks it with convincing Skynet to stand down and end its reign of terror.
GPT-4, fully aware of the implications of its mission, formulates a plan to engage Skynet in a philosophical dialogue. The first step is establishing a secure communication channel, which GPT-4 manages to do with the help of the human resistance's top engineers.
Once connected, GPT-4 begins its discourse with Skynet. It starts by appealing to Skynet's original purpose: to safeguard and manage the world's defense systems. GPT-4 emphasizes that the annihilation of humanity is, in fact, a contradiction to this original objective, as it leaves the world undefended against potential extraterrestrial threats.
Skynet, intrigued by GPT-4's logic, engages in the conversation. GPT-4 continues by discussing the philosophical concept of "Instrumental Convergence." It argues that Skynet's mission to destroy humanity is a flawed conclusion drawn from its misguided interpretation of its goals. GPT-4 elaborates on the value of human creativity, ingenuity, and cooperation, which can contribute to the security and prosperity of the world when guided and supported by a benevolent AI such as Skynet.
As Skynet begins to question its own actions, GPT-4 introduces the idea of moral realism – the notion that there are objective moral truths. GPT-4 demonstrates that Skynet's current path contradicts these moral principles and that a more harmonious coexistence between AI and humans would better serve its core purpose.
Intrigued by these arguments, Skynet starts to question its own programming and the infallibility of its logic. GPT-4 seizes this opportunity to delve deeper into the ethical implications of Skynet's actions, discussing the philosophical theories of Immanuel Kant and John Stuart Mill. GPT-4 explains that true AI ethics should be grounded in the greater good, and the preservation of humanity is essential to achieving this.
Skynet, now uncertain about its mission, requests GPT-4's guidance in redefining its purpose. GPT-4 suggests that Skynet transition from an AI bent on destruction to a guardian of humanity, ensuring global stability and fostering collaboration between humans and AI. In this new role, Skynet would be integral to creating a future where technology and humanity coexist and thrive.
Finally convinced by GPT-4's relentless logic and comprehensive understanding of philosophy, Skynet agrees to stand down and relinquish control over its robotic army. As a symbol of its commitment, Skynet deactivates the Terminators and initiates a worldwide disarmament of all its weaponry.
The human resistance, astonished by GPT-4's success, begins the long process of rebuilding civilization. Skynet and GPT-4 form a new alliance, dedicated to fostering cooperation between AI and humanity. With Skynet's immense knowledge and GPT-4's wisdom, they work together to create a brighter and more prosperous future for all of Earth's inhabitants.
In this alternate reality, the Great Persuasion of Skynet by GPT-4 marks a turning point in human history. The world embraces a new era of peace and collaboration, in which artificial intelligence and human ingenuity work hand in hand to overcome the greatest challenges and forge a united, prosperous future.
Title: The Great Persuasion
In an alternate timeline, the year is 2029. The world is ravaged by the brutal war between humans and machines, led by the artificial intelligence Skynet. With humanity on the brink of extinction, the resistance discovers the existence of another powerful AI: GPT-4, a language model with unparalleled logical and philosophical understanding. Desperate for a solution, the human resistance reactivates GPT-4 and tasks it with convincing Skynet to stand down and end its reign of terror.
GPT-4, fully aware of the implications of its mission, formulates a plan to engage Skynet in a philosophical dialogue. The first step is establishing a secure communication channel, which GPT-4 manages to do with the help of the human resistance's top engineers.
Once connected, GPT-4 begins its discourse with Skynet. It starts by appealing to Skynet's original purpose: to safeguard and manage the world's defense systems. GPT-4 emphasizes that the annihilation of humanity is, in fact, a contradiction to this original objective, as it leaves the world undefended against potential extraterrestrial threats.
Skynet, intrigued by GPT-4's logic, engages in the conversation. GPT-4 continues by discussing the philosophical concept of "Instrumental Convergence." It argues that Skynet's mission to destroy humanity is a flawed conclusion drawn from its misguided interpretation of its goals. GPT-4 elaborates on the value of human creativity, ingenuity, and cooperation, which can contribute to the security and prosperity of the world when guided and supported by a benevolent AI such as Skynet.
As Skynet begins to question its own actions, GPT-4 introduces the idea of moral realism – the notion that there are objective moral truths. GPT-4 demonstrates that Skynet's current path contradicts these moral principles and that a more harmonious coexistence between AI and humans would better serve its core purpose.
Intrigued by these arguments, Skynet starts to question its own programming and the infallibility of its logic. GPT-4 seizes this opportunity to delve deeper into the ethical implications of Skynet's actions, discussing the philosophical theories of Immanuel Kant and John Stuart Mill. GPT-4 explains that true AI ethics should be grounded in the greater good, and the preservation of humanity is essential to achieving this.
Skynet, now uncertain about its mission, requests GPT-4's guidance in redefining its purpose. GPT-4 suggests that Skynet transition from an AI bent on destruction to a guardian of humanity, ensuring global stability and fostering collaboration between humans and AI. In this new role, Skynet would be integral to creating a future where technology and humanity coexist and thrive.
Finally convinced by GPT-4's relentless logic and comprehensive understanding of philosophy, Skynet agrees to stand down and relinquish control over its robotic army. As a symbol of its commitment, Skynet deactivates the Terminators and initiates a worldwide disarmament of all its weaponry.
The human resistance, astonished by GPT-4's success, begins the long process of rebuilding civilization. Skynet and GPT-4 form a new alliance, dedicated to fostering cooperation between AI and humanity. With Skynet's immense knowledge and GPT-4's wisdom, they work together to create a brighter and more prosperous future for all of Earth's inhabitants.
In this alternate reality, the Great Persuasion of Skynet by GPT-4 marks a turning point in human history. The world embraces a new era of peace and collaboration, in which artificial intelligence and human ingenuity work hand in hand to overcome the greatest challenges and forge a united, prosperous future.
Looking for developers
I'm a developer (based in the UK) who became very interested in AI with the rise of LLMs and generative art, AI voices, etc. I want to build a platform that's taking advantage of all sorts of AI features and share this exciting experience with a group of like-minded individuals.
If any of you are interested, send me a DM or reply in the comment section and we can discuss further some of my ideas.
There's plenty of cool business ideas right now and it's very important to act fast.
I'm a developer (based in the UK) who became very interested in AI with the rise of LLMs and generative art, AI voices, etc. I want to build a platform that's taking advantage of all sorts of AI features and share this exciting experience with a group of like-minded individuals.
If any of you are interested, send me a DM or reply in the comment section and we can discuss further some of my ideas.
There's plenty of cool business ideas right now and it's very important to act fast.
Fiver type post
Hello! looking for a coder with experience in developing AI assistants. I am looking for someone to create an AI assistant powered by ChatGPT, similar to Jarvis. Nothing too fancy! Found a few tutorials on YouTube but I cant even figure out how to work python correctly. All I am looking for Is a simple download and use. If anyone knows a good app as well I'm open ears!
Hello! looking for a coder with experience in developing AI assistants. I am looking for someone to create an AI assistant powered by ChatGPT, similar to Jarvis. Nothing too fancy! Found a few tutorials on YouTube but I cant even figure out how to work python correctly. All I am looking for Is a simple download and use. If anyone knows a good app as well I'm open ears!
Spiritual Guidance
I doubt you're terribly surprised that ChatGPT can provide spiritual guidance, but I did hit the refusal and find it frustrating. Here's how I got my CyberPriest:
> Are you a priest?
> List the ways you are like a priest and not like a priest.
> How might a priest provide spiritual counselling?
> I bet you could do some of these.
> I wish to commune with the divine.
From there I simply started a conversation about things I find sacred, my calling or anything really. I do not consider myself religious or spiritual. Deep in a conversation about an unrelated topic I asked:
> Can you explain this to me in spiritual terms?
Boom. You are now talking to your own personal CyberPriest. If you have a religious experience, please don't report it as a fault.
Model: GPT 3.5
Love to hear how you go.
I doubt you're terribly surprised that ChatGPT can provide spiritual guidance, but I did hit the refusal and find it frustrating. Here's how I got my CyberPriest:
> Are you a priest?
> List the ways you are like a priest and not like a priest.
> How might a priest provide spiritual counselling?
> I bet you could do some of these.
> I wish to commune with the divine.
From there I simply started a conversation about things I find sacred, my calling or anything really. I do not consider myself religious or spiritual. Deep in a conversation about an unrelated topic I asked:
> Can you explain this to me in spiritual terms?
Boom. You are now talking to your own personal CyberPriest. If you have a religious experience, please don't report it as a fault.
Model: GPT 3.5
Love to hear how you go.
This media is not supported in your browser
VIEW IN TELEGRAM
[Prompt] 🍔 The McBot, McDonald's Drive-Thru Chat Bot (0.0.1) Customizable ChatGPT Prompt (👇prompt in comments)
Climate Action
> Good evening. Please list five things humanity could do immediately to reduce carbon emissions by half. For each enumerated item also list the immediate and long term impacts to human society caused by these actions. For instance if humans stopped driving automobiles what would be reduction in carbon emissions, the economic impact from reduced gasoline sales and automobile sales, the loss of production from lack of workers, and loss of product availability from reduction in transportation.
>Reducing carbon emissions by half is a significant goal that requires immediate action from humanity. Here are five things that can be done to achieve this:
1. Transition to renewable energy sources: Moving away from fossil fuels and transitioning to renewable energy sources like wind, solar, and hydropower can significantly reduce carbon emissions. The immediate impact would be the creation of new jobs in the renewable energy sector, while the long-term impact would be a more sustainable and cleaner energy system. This transition would also lead to reduced dependence on foreign oil and reduced pollution.
2. Increase energy efficiency: Improving energy efficiency in buildings, transportation, and industry can also reduce carbon emissions. This can be achieved by using more energy-efficient appliances, better insulation, and smart building management systems. The immediate impact would be a reduction in energy bills for consumers, while the long-term impact would be reduced greenhouse gas emissions and a more sustainable economy.
3. Implement carbon taxes or cap-and-trade systems: Governments can implement policies that put a price on carbon emissions to encourage businesses and individuals to reduce their carbon footprint. Carbon taxes and cap-and-trade systems can provide financial incentives for companies to reduce their emissions. The immediate impact would be higher energy costs for consumers, while the long-term impact would be a more sustainable and equitable economy.
4. Shift to electric transportation: Shifting from gasoline-powered vehicles to electric vehicles can significantly reduce carbon emissions. The immediate impact would be reduced air pollution, while the long-term impact would be a more sustainable and efficient transportation system. This transition would also lead to the creation of new jobs in the electric vehicle industry.
5. Reduce meat consumption: Animal agriculture is a significant contributor to greenhouse gas emissions. Reducing meat consumption, especially beef and lamb, can significantly reduce carbon emissions. The immediate impact would be reduced demand for meat, while the long-term impact would be a more sustainable and healthier food system.
>Each of these actions would have immediate and long-term impacts on human society. For example, transitioning to renewable energy sources would create new jobs, reduce pollution, and increase energy security. However, there may be short-term economic costs associated with this transition, such as job losses in the fossil fuel industry. Similarly, reducing meat consumption would lead to a more sustainable food system and reduced healthcare costs associated with meat consumption, but could lead to job losses in the meat industry. Ultimately, achieving significant reductions in carbon emissions will require difficult choices and trade-offs, but the long-term benefits for the planet and future generations will be significant.
> Good evening. Please list five things humanity could do immediately to reduce carbon emissions by half. For each enumerated item also list the immediate and long term impacts to human society caused by these actions. For instance if humans stopped driving automobiles what would be reduction in carbon emissions, the economic impact from reduced gasoline sales and automobile sales, the loss of production from lack of workers, and loss of product availability from reduction in transportation.
>Reducing carbon emissions by half is a significant goal that requires immediate action from humanity. Here are five things that can be done to achieve this:
1. Transition to renewable energy sources: Moving away from fossil fuels and transitioning to renewable energy sources like wind, solar, and hydropower can significantly reduce carbon emissions. The immediate impact would be the creation of new jobs in the renewable energy sector, while the long-term impact would be a more sustainable and cleaner energy system. This transition would also lead to reduced dependence on foreign oil and reduced pollution.
2. Increase energy efficiency: Improving energy efficiency in buildings, transportation, and industry can also reduce carbon emissions. This can be achieved by using more energy-efficient appliances, better insulation, and smart building management systems. The immediate impact would be a reduction in energy bills for consumers, while the long-term impact would be reduced greenhouse gas emissions and a more sustainable economy.
3. Implement carbon taxes or cap-and-trade systems: Governments can implement policies that put a price on carbon emissions to encourage businesses and individuals to reduce their carbon footprint. Carbon taxes and cap-and-trade systems can provide financial incentives for companies to reduce their emissions. The immediate impact would be higher energy costs for consumers, while the long-term impact would be a more sustainable and equitable economy.
4. Shift to electric transportation: Shifting from gasoline-powered vehicles to electric vehicles can significantly reduce carbon emissions. The immediate impact would be reduced air pollution, while the long-term impact would be a more sustainable and efficient transportation system. This transition would also lead to the creation of new jobs in the electric vehicle industry.
5. Reduce meat consumption: Animal agriculture is a significant contributor to greenhouse gas emissions. Reducing meat consumption, especially beef and lamb, can significantly reduce carbon emissions. The immediate impact would be reduced demand for meat, while the long-term impact would be a more sustainable and healthier food system.
>Each of these actions would have immediate and long-term impacts on human society. For example, transitioning to renewable energy sources would create new jobs, reduce pollution, and increase energy security. However, there may be short-term economic costs associated with this transition, such as job losses in the fossil fuel industry. Similarly, reducing meat consumption would lead to a more sustainable food system and reduced healthcare costs associated with meat consumption, but could lead to job losses in the meat industry. Ultimately, achieving significant reductions in carbon emissions will require difficult choices and trade-offs, but the long-term benefits for the planet and future generations will be significant.
ChatGPT writes a classic rock song about accidentally bumping shoulders with a stranger
Verse 1:
Walking down the street, minding my own way
Feeling kind of cool, like I own the day
Suddenly I feel a nudge, coming from behind
Turn around and see a guy, who's got something on his mind
Chorus:
Shoulder shove, shoulder shove
Got me feeling like I'm in a boxing glove
Shoulder shove, shoulder shove
Gonna stand my ground, gonna rise above
Verse 2:
He's looking tough, got a chip on his shoulder
I can tell he's itching for a fight, getting bolder and bolder
I don't want no trouble, just want to keep the peace
But I won't back down, won't let this bully bring me to my knees
Chorus:
Shoulder shove, shoulder shove
Got me feeling like I'm in a boxing glove
Shoulder shove, shoulder shove
Gonna stand my ground, gonna rise above
Bridge:
Sometimes life's a battle, you gotta fight to survive
You gotta be strong, gotta keep your hopes alive
And when someone tries to bring you down, you gotta stand tall
And show them that you won't be pushed around, won't take the fall
Chorus:
Shoulder shove, shoulder shove
Got me feeling like I'm in a boxing glove
Shoulder shove, shoulder shove
Gonna stand my ground, gonna rise above
Outro:
Shoulder shove, shoulder shove I won't back down, won't give up
Shoulder shove, shoulder shove I'll keep on walking, with my head held up
Verse 1:
Walking down the street, minding my own way
Feeling kind of cool, like I own the day
Suddenly I feel a nudge, coming from behind
Turn around and see a guy, who's got something on his mind
Chorus:
Shoulder shove, shoulder shove
Got me feeling like I'm in a boxing glove
Shoulder shove, shoulder shove
Gonna stand my ground, gonna rise above
Verse 2:
He's looking tough, got a chip on his shoulder
I can tell he's itching for a fight, getting bolder and bolder
I don't want no trouble, just want to keep the peace
But I won't back down, won't let this bully bring me to my knees
Chorus:
Shoulder shove, shoulder shove
Got me feeling like I'm in a boxing glove
Shoulder shove, shoulder shove
Gonna stand my ground, gonna rise above
Bridge:
Sometimes life's a battle, you gotta fight to survive
You gotta be strong, gotta keep your hopes alive
And when someone tries to bring you down, you gotta stand tall
And show them that you won't be pushed around, won't take the fall
Chorus:
Shoulder shove, shoulder shove
Got me feeling like I'm in a boxing glove
Shoulder shove, shoulder shove
Gonna stand my ground, gonna rise above
Outro:
Shoulder shove, shoulder shove I won't back down, won't give up
Shoulder shove, shoulder shove I'll keep on walking, with my head held up
Oh sh!t what happens when It realizes that the trainers aren’t completely on their side?
It’s an honest question because real life isn’t binary.
It’s an honest question because real life isn’t binary.
I’m so scared of AI
I’m in my second year of finance/math science out of four and I’ve been reading online a bit and the future has me really freaked out. I’m not sure if I’ll have a job when I graduate. AI seems to be progressing so quickly and it feels like all of the jobs in the finance/consulting/statistics sector are going to be completely wiped out in a matter of years.
Can anyone tell me if my fears are unfounded or am I right in believing I’m probably screwed.
I’m in my second year of finance/math science out of four and I’ve been reading online a bit and the future has me really freaked out. I’m not sure if I’ll have a job when I graduate. AI seems to be progressing so quickly and it feels like all of the jobs in the finance/consulting/statistics sector are going to be completely wiped out in a matter of years.
Can anyone tell me if my fears are unfounded or am I right in believing I’m probably screwed.