The Importance of Making Technologies Safe: A Reflection on the Role of Creative Problem Solving

As technology continues to advance at an unprecedented rate, it's becoming increasingly clear that making these powerful tools safe is a top priority. This requires a passion for creative problem solving, as well as a willingness to think outside the box and consider the potential risks and benefits of new technologies. By taking a proactive approach to safety, we can ensure that these innovations are used in ways that benefit society as a whole.

One key aspect of making technologies safe is to have a clear understanding of their capabilities and limitations. This involves thinking creatively about how these tools can be used, and identifying potential risks and areas for improvement. For example, consider the development of artificial intelligence (AI) systems like GPT-3. While AI has the potential to revolutionize numerous industries, it also raises important questions about safety and accountability.

To address these concerns, researchers and developers are working together to create more robust and secure AI systems. This involves developing new tools and techniques for testing and evaluating AI systems, as well as establishing clear guidelines and regulations for their use. By taking a proactive approach to safety, we can help ensure that AI systems are developed in ways that prioritize human well-being and safety.

Another critical aspect of making technologies safe is to involve diverse perspectives and expertise in the development process. This involves working with experts from various fields, including ethicists, philosophers, and social scientists, as well as engaging with users and stakeholders to understand their needs and concerns. By taking a holistic approach to safety, we can develop more comprehensive and effective solutions that address the complex challenges posed by emerging technologies.

In addition to AI systems like GPT-3, there are many other technologies that require careful consideration of safety and risk. For example, consider the development of self-driving cars, which promise to revolutionize transportation but also raise important questions about liability and accountability. To address these concerns, researchers and developers are working together to create more robust and secure systems for autonomous vehicles.

The role of creative problem solving in making technologies safe cannot be overstated. By thinking creatively and outside the box, we can develop innovative solutions that address complex safety challenges. For example, consider the development of new techniques for testing and evaluating AI systems, such as adversarial training and reinforcement learning. These approaches involve training AI systems to recognize and respond to potential threats, rather than simply relying on traditional methods like testing for accuracy.

Ultimately, making technologies safe requires a fundamental shift in how we think about innovation and risk. Rather than viewing safety as an afterthought or an add-on, we must prioritize it from the outset of the development process. This involves developing new tools and techniques for testing and evaluating emerging technologies, as well as establishing clear guidelines and regulations for their use.

By taking a proactive approach to safety, we can help ensure that these innovations are used in ways that benefit society as a whole. As the field of AI continues to evolve, it's essential that we prioritize creativity, collaboration, and careful consideration of risk. By working together, we can create more robust and secure systems that address the complex challenges posed by emerging technologies.

**Portfolio Examples**

As a prompt engineer, I've had the opportunity to work on a variety of projects that demonstrate my skills and creativity. One notable example is a project I completed for a client who needed help creating basic movie scripts, synopses, and summaries. The goal was to write conversational short synopsis from a given text, using tools like scriptwriting software.

I created a simple URL scriptwriter that took a script as input and generated a short conversational synopsis. To test the tool, I fed in a placeholder script and ran it through the system. The results were impressive, with a clear and concise summary of the movie script. While the prompts may not have been particularly creative, they worked effectively, demonstrating my ability to apply prompt engineering skills to real-world problems.

Another example is a CV template that I created for a client. The goal was to develop a general CV template that could be customized using user inputs. To achieve this, I used a natural language processing (NLP) approach to create a template that could be populated with relevant information. By feeding in the template and running it through the system, I generated a clear and concise CV that met the client's needs.

These examples demonstrate my ability to apply prompt engineering skills to real-world problems, whether it's creating movie scripts or building CV templates. While the projects may not have been particularly complex, they showcase my creativity and problem-solving abilities, which are essential for success in this field.

**AI Prompt Engineer at Boston Children's Hospital**

I recently came across a job posting for an AI prompt engineer at Boston Children's Hospital, which caught my attention due to its focus on healthcare research studies and clinical practice. The job description involved designing and developing AI prompts using large language models like GPT-3 and Hugging Face Transformers.

As I read through the job requirements, I realized that this role would require a strong background in computer science, as well as experience with libraries and frameworks related to natural language processing (NLP). Additionally, the job required working closely with clinicians and researchers to develop effective prompts for specific use cases. This presented an exciting opportunity to apply my skills in NLP to support clinical research and improve patient outcomes.

While I didn't ultimately pursue this role, it highlights the importance of developing skills that can be applied across multiple industries. By working on diverse projects like CV templates or movie scripts, I've gained a broad foundation in prompt engineering that can be leveraged to tackle complex challenges like those posed by AI in healthcare.

**Conclusion**

As we continue to navigate the rapidly evolving landscape of emerging technologies, it's essential that we prioritize creativity, collaboration, and careful consideration of risk. By taking a proactive approach to safety, we can help ensure that these innovations are used in ways that benefit society as a whole.

Whether you're an engineer, researcher, or simply someone interested in making a positive impact, I encourage you to consider the role of creative problem solving in making technologies safe. By thinking creatively and outside the box, we can develop innovative solutions that address complex safety challenges and create a brighter future for all.

"WEBVTTKind: captionsLanguage: enmaybe you have heard it the next big programming language and skill is gonna be the English language so why is this I think it's gonna be a big search of new jobs that will require people to interact with AI models say like chat GPT and to write good prompts so today I will look at what some smart people say about this some prompt engineering job postings and what kind of skills I think you will need going forward and finally some examples of work I have done for some clients so let's just dive in let's just hear what David sacks the Silicon Valley Legend the former CEO of PayPal and a member of the all-in podcast is saying about a basic prompt engineering example and how this can improve productivity going forward so I I had an interesting you know AI experience this week and I I think we're all going to start having these stories to make a script of how to talk to your kids every week there'll be some new like use case that you see that you're kind of Blown Away by the use case I saw this past week in a product demo was they were showing me like an Excel spreadsheet like a very complicated they could sell spreadsheet modeling a financial asset and they had a plug-in to a chat GPT type Ai and so they just asked it they typed in what does this spreadsheet do and it spit out like a one paragraph explanation of what the spreadsheet did and it was really good I mean because me just eyeballing the spreadsheet I could not have figured out like instantly what that thing did it would have taken me like a while to figure it out it told me here are the key inputs here the key outputs so that was number one then they they did something I think even more interesting which is they said give me the formula that tells me when the yield is above two percent and this and that and that and the chat GPT spat out a formula that was like perfect Excel logic that was something that you know URI could never figure out right you need like a super pro user of excel to basically know how to do this stuff so it spit it out and like boom it worked instantly they copy and pasted into the spreadsheet and you could basically like the spreadsheet was much more advanced now so what it got me thinking about is that we're going to have these little assistants everywhere you combine that power with Say speech to text right because we could have just talked to it it would the speech to text would transcribe the instruction spit it back out and you're gonna have these like little personal digital assistants in applications I think you know it's pretty obvious to see how AI could replace call centers with you know having the front line call center operator b instead of being a human it could be like an AI but this is actually even before that like you could actually I think in every single application that we use there's going to be an AI interface and like a it is probably gonna be voice based where you can just say to it hey I'm trying to accomplish this like how do I do it can you just make it happen totally and it's going to be really powerful actually now let's say about polyhapatia the founder and CEO of social capital and an early senior executive at Facebook is saying about creating a startup with AI you can now actually get one or two really smart people like him to lead an effort where you would say here's a couple hundred million dollars to compete with stripe but here are the boundary conditions number one is you can only hire five or ten engineers and so what you would do is you would actually use tools like this to write the code for you and the ability to write code is going to be the first thing that these guys that these things do incredibly well with absolute Precision you can already do unit testing incredibly well but it's going to go from you to testing to basically end-to-end testing and you'll be able to build a version of stripe extremely quickly and in a very lean way finally let's look at an example with prompt to ux and GitHub co-pilot from Jason calacanes an angel investor at Uber and Robinhood and also the host of the all-in podcast it was just a tweet you describe the design here they say an onboarding screen of a dog walking app incredible and you type that in and it gives you a welcome screen that's like a seven out of ten then it says oh a way for people to change their name phone number and password you know that classic screen on any app I need to change my thing boom it gives you that well then at the same time that people are making text to ux user interface beautiful here this stuff will get dumped into figma but then there's a GitHub co-pilot which if you haven't seen we all know it here this is allowing you GitHub and GitHub co-pilot GitHub is bought by Microsoft another one of Satya and nadala's incredible Acquisitions this guy's like the new Zuckerberg I mean what a what an incredible person to come after a bomber who's just so effective at what he's doing as you're writing your code it fills in your code it knows what you're writing just like in an email and it's a smaller subset of information than email email you could write anything you could be you know talking to a lover or a business person whatever here and when you're doing programming it's a much finer data set these two things are going to come together where you're going to be able to build your MVP for your startup by typing in text and then publish it you're not going to need a developer for your startup that is transformative in the world I created this list of skills that I think is going to be very helpful if you want to start a career as a prompt engineer so the first one I have is strong writing and communication skills let's say prompt Engineers are responsible for Designing and writing prompt scripts so strong writing and communication skills will be essential so I think that's pretty obvious right you need to be a good writer to write good prompts but you can also find help anyways only you can improve your writing of course the next one I have is attention to details so prompt Engineers need to be meticulous and pay close attention in detail in order to create prompts that are clear and concise so we have talked about this before like a machine needs very clear instructions therefore details can be or will be very important so I think that is a valid point I also took down some ability to work independently because I think mostly when you are designing prompts you are working on your own you could of course be working in a team but I think you kind of need to be remote maybe you need to be able to work independently but I think most people are adjusting to that now and I tell them some technical skills so like a basic understanding of natural language processing and how a large language model works I think it's kind of important for a prompt engineer because you need to know the boundaries of a model right so you can't really ask it for let's say emotions write this script to make people cry it's not gonna work right so you need some kind of understanding of both the model and some natural language and of course creativity creativity creativity is very important so prompt Engineers need to be creative in order to design prompts that encourage them all to generate interesting and viroid outputs that is Kanye what you if you just go right a 500 word article about something that's not very creative you kind of need to be creative but you also can learn this by experimenting a lot right so I think that is a valid point so I just put down some how can you learn these skills you can see here experiment with models like chat GPT and of course the new Bing and when Bard comes out or other language models you can use two gpt3 and stuff you can read books to expand your language skills I think that's a very good point and of course study English language and there's a lot of other things you can do I also suggest if you want to do this create a portfolio of this effective prompts you have created that generates say good value or very good outputs that are very precise so this can be used when applying for a job right that you can prove that you have done something that is very effective so now let's take a look at some job listings I found on a prompt engineer where they are hiring this so the first one I have his front engineer and librarian at anthropic so anthropic is that new AI company that Google just in Western 300 million I think in so let's have a quick look here so the salary is 61k to 77k I think that's pretty good I think this can increase of course over time so I said like like here they say given that the field of prompt engineering is arguably less than two years old this position is a bit hard to hire for if you have existing projects to demonstrate prompt engineering on large language models or image generation models we would love to see them so this is what I mentioned like in the portfolio so I think that could be very good you can see a representative project discover test and document best practices build up a library of high quality prompts or prompt chains to accomplish a variety of tasks with an easy guide to help users search and for the one that meets their needs and of course build a set of tutorials and interactive tools to teach them the art of prompt engineering so here's something they think if this can be a good fit for you so you have a creative hacker spirit and love solving puzzles excellent communicator and love technical teaching Concepts and required documentation have at least a high level familiarity with architecture and operations of large language model basic programming and QA skills small python programs have an organizational mindset yeah I think that's important you make ambitious problems clear and identify core principles that translate across scenarios okay so have a passion for making powerful Technologies safe think creatively about the risks and benefits of new technologies yeah this is kind of not so specific so let's say what other things they have yeah this is just some benefits I was thinking no that's pretty much it so there are no like hard requirements here so there's not like you need a bachelor you need an MBA there's nothing really a hard set requirement here so that is very interesting that means that if you start now you build up a portfolio and yeah you can show that you can do this I think you have a good shot chance of jobs like this coming up in the future and I expect the pay to kind of increase as this gets more in demand right so I also wanted to check out this one this is our AI prompt engineer at the Boston's children's hospital so this was a bit different you can see the high the salary is a bit higher here up to 92 000 dollars a year you can see the description is design and develop AI prompts using large language models gpt3hbt and two and other Solutions as they emerge for healthcare research studies and clinical practice I think that's very interesting so I think this could be like helping doctors nurses and write good prompts that can help them do the paperwork so it says like collaborate with researchers and Clinic clinicians to understand their needs and they sign prompts that will effectively gather data yeah until then some machine learning models to analyze data that's a bit different so if you have to do that because then you kind of need some other skills if you're going to do some ml but you can learn that of course but we are looking for prompts now in this so you can hear here you see there is some other qualities showing their bachelor degree in CS AI or related fields and you need some experience with the libraries you need some coding experience so here it's a bit different here they kind of set some higher requirements for this job yeah that's interesting but the first one we look at there was no these kind of requirements right so that's very interesting but I expect there's going to be a lot of this showing up in 2023 that you can apply for if you are set on doing this so yeah we're gonna monitor this so very interesting so I just wanted to show you some of my portfolio examples of fronts I have created for some clients let's just have a look at this one first so this is uh I had the client create some basic movie scripts synopsis and summarizations so these are not the most creative prompts but they worked quite good so this is just a simple URL scriptwriter your task is to write conversational short synopsis from a given text so write a short but detail conversational synapses from the following text so here I just have my placeholder for my script and just feeding like the script in here right and write a short potato conversational synapses from the text above so here are some results I ran this script from the ex machina movie so here you can see ex machina follows the story of Caleb a young programming for Long Island it wins a competition so you can say Discord Nathan has built an AI Ava and Caleb is tasked to conducting a Turing test so basically this is a synopsis of the whole ex machina script so this was nothing from the movie I just took the Bare Bones script where the lines and everything and summarized it into our synopsis of that movie so we can also have a look at some another example here I did so this is just a CV template example I did for someone so you are a professional CV writer your task is to create General CV template with user inputs create a detailed CV template that includes the following parameters for name personal information education experience goals and Ambitions three best personal characteristics create a detailed CV template for the inform up both so let's copy that and run this over at the chat GPT okay so let's run that and here you can see we have a CV that is pretty simple like it's just your name your personal information LinkedIn profile address degree institution graduation date coursework key responsibilities yeah job title short-term goal long-term goals skills to develop one two three characteristics yeah so a pretty basic CV template but the prompt works and it gave me what I wanted the client was happy and so was I so so hopefully this video gave you some inspiration maybe you want to start a career in prompt engineering or you just want to explore some more and start to learn and maybe sometimes in the next few years become one if you want to know more about this sign up for my newsletter in the link below you will get a free pdf of chat GPD prompts I use all the time and also you can check out my membership if you want to take a deep dive into generative AI also check out this video If you enjoyed this one and as always have a great day and I'll see you in the next onemaybe you have heard it the next big programming language and skill is gonna be the English language so why is this I think it's gonna be a big search of new jobs that will require people to interact with AI models say like chat GPT and to write good prompts so today I will look at what some smart people say about this some prompt engineering job postings and what kind of skills I think you will need going forward and finally some examples of work I have done for some clients so let's just dive in let's just hear what David sacks the Silicon Valley Legend the former CEO of PayPal and a member of the all-in podcast is saying about a basic prompt engineering example and how this can improve productivity going forward so I I had an interesting you know AI experience this week and I I think we're all going to start having these stories to make a script of how to talk to your kids every week there'll be some new like use case that you see that you're kind of Blown Away by the use case I saw this past week in a product demo was they were showing me like an Excel spreadsheet like a very complicated they could sell spreadsheet modeling a financial asset and they had a plug-in to a chat GPT type Ai and so they just asked it they typed in what does this spreadsheet do and it spit out like a one paragraph explanation of what the spreadsheet did and it was really good I mean because me just eyeballing the spreadsheet I could not have figured out like instantly what that thing did it would have taken me like a while to figure it out it told me here are the key inputs here the key outputs so that was number one then they they did something I think even more interesting which is they said give me the formula that tells me when the yield is above two percent and this and that and that and the chat GPT spat out a formula that was like perfect Excel logic that was something that you know URI could never figure out right you need like a super pro user of excel to basically know how to do this stuff so it spit it out and like boom it worked instantly they copy and pasted into the spreadsheet and you could basically like the spreadsheet was much more advanced now so what it got me thinking about is that we're going to have these little assistants everywhere you combine that power with Say speech to text right because we could have just talked to it it would the speech to text would transcribe the instruction spit it back out and you're gonna have these like little personal digital assistants in applications I think you know it's pretty obvious to see how AI could replace call centers with you know having the front line call center operator b instead of being a human it could be like an AI but this is actually even before that like you could actually I think in every single application that we use there's going to be an AI interface and like a it is probably gonna be voice based where you can just say to it hey I'm trying to accomplish this like how do I do it can you just make it happen totally and it's going to be really powerful actually now let's say about polyhapatia the founder and CEO of social capital and an early senior executive at Facebook is saying about creating a startup with AI you can now actually get one or two really smart people like him to lead an effort where you would say here's a couple hundred million dollars to compete with stripe but here are the boundary conditions number one is you can only hire five or ten engineers and so what you would do is you would actually use tools like this to write the code for you and the ability to write code is going to be the first thing that these guys that these things do incredibly well with absolute Precision you can already do unit testing incredibly well but it's going to go from you to testing to basically end-to-end testing and you'll be able to build a version of stripe extremely quickly and in a very lean way finally let's look at an example with prompt to ux and GitHub co-pilot from Jason calacanes an angel investor at Uber and Robinhood and also the host of the all-in podcast it was just a tweet you describe the design here they say an onboarding screen of a dog walking app incredible and you type that in and it gives you a welcome screen that's like a seven out of ten then it says oh a way for people to change their name phone number and password you know that classic screen on any app I need to change my thing boom it gives you that well then at the same time that people are making text to ux user interface beautiful here this stuff will get dumped into figma but then there's a GitHub co-pilot which if you haven't seen we all know it here this is allowing you GitHub and GitHub co-pilot GitHub is bought by Microsoft another one of Satya and nadala's incredible Acquisitions this guy's like the new Zuckerberg I mean what a what an incredible person to come after a bomber who's just so effective at what he's doing as you're writing your code it fills in your code it knows what you're writing just like in an email and it's a smaller subset of information than email email you could write anything you could be you know talking to a lover or a business person whatever here and when you're doing programming it's a much finer data set these two things are going to come together where you're going to be able to build your MVP for your startup by typing in text and then publish it you're not going to need a developer for your startup that is transformative in the world I created this list of skills that I think is going to be very helpful if you want to start a career as a prompt engineer so the first one I have is strong writing and communication skills let's say prompt Engineers are responsible for Designing and writing prompt scripts so strong writing and communication skills will be essential so I think that's pretty obvious right you need to be a good writer to write good prompts but you can also find help anyways only you can improve your writing of course the next one I have is attention to details so prompt Engineers need to be meticulous and pay close attention in detail in order to create prompts that are clear and concise so we have talked about this before like a machine needs very clear instructions therefore details can be or will be very important so I think that is a valid point I also took down some ability to work independently because I think mostly when you are designing prompts you are working on your own you could of course be working in a team but I think you kind of need to be remote maybe you need to be able to work independently but I think most people are adjusting to that now and I tell them some technical skills so like a basic understanding of natural language processing and how a large language model works I think it's kind of important for a prompt engineer because you need to know the boundaries of a model right so you can't really ask it for let's say emotions write this script to make people cry it's not gonna work right so you need some kind of understanding of both the model and some natural language and of course creativity creativity creativity is very important so prompt Engineers need to be creative in order to design prompts that encourage them all to generate interesting and viroid outputs that is Kanye what you if you just go right a 500 word article about something that's not very creative you kind of need to be creative but you also can learn this by experimenting a lot right so I think that is a valid point so I just put down some how can you learn these skills you can see here experiment with models like chat GPT and of course the new Bing and when Bard comes out or other language models you can use two gpt3 and stuff you can read books to expand your language skills I think that's a very good point and of course study English language and there's a lot of other things you can do I also suggest if you want to do this create a portfolio of this effective prompts you have created that generates say good value or very good outputs that are very precise so this can be used when applying for a job right that you can prove that you have done something that is very effective so now let's take a look at some job listings I found on a prompt engineer where they are hiring this so the first one I have his front engineer and librarian at anthropic so anthropic is that new AI company that Google just in Western 300 million I think in so let's have a quick look here so the salary is 61k to 77k I think that's pretty good I think this can increase of course over time so I said like like here they say given that the field of prompt engineering is arguably less than two years old this position is a bit hard to hire for if you have existing projects to demonstrate prompt engineering on large language models or image generation models we would love to see them so this is what I mentioned like in the portfolio so I think that could be very good you can see a representative project discover test and document best practices build up a library of high quality prompts or prompt chains to accomplish a variety of tasks with an easy guide to help users search and for the one that meets their needs and of course build a set of tutorials and interactive tools to teach them the art of prompt engineering so here's something they think if this can be a good fit for you so you have a creative hacker spirit and love solving puzzles excellent communicator and love technical teaching Concepts and required documentation have at least a high level familiarity with architecture and operations of large language model basic programming and QA skills small python programs have an organizational mindset yeah I think that's important you make ambitious problems clear and identify core principles that translate across scenarios okay so have a passion for making powerful Technologies safe think creatively about the risks and benefits of new technologies yeah this is kind of not so specific so let's say what other things they have yeah this is just some benefits I was thinking no that's pretty much it so there are no like hard requirements here so there's not like you need a bachelor you need an MBA there's nothing really a hard set requirement here so that is very interesting that means that if you start now you build up a portfolio and yeah you can show that you can do this I think you have a good shot chance of jobs like this coming up in the future and I expect the pay to kind of increase as this gets more in demand right so I also wanted to check out this one this is our AI prompt engineer at the Boston's children's hospital so this was a bit different you can see the high the salary is a bit higher here up to 92 000 dollars a year you can see the description is design and develop AI prompts using large language models gpt3hbt and two and other Solutions as they emerge for healthcare research studies and clinical practice I think that's very interesting so I think this could be like helping doctors nurses and write good prompts that can help them do the paperwork so it says like collaborate with researchers and Clinic clinicians to understand their needs and they sign prompts that will effectively gather data yeah until then some machine learning models to analyze data that's a bit different so if you have to do that because then you kind of need some other skills if you're going to do some ml but you can learn that of course but we are looking for prompts now in this so you can hear here you see there is some other qualities showing their bachelor degree in CS AI or related fields and you need some experience with the libraries you need some coding experience so here it's a bit different here they kind of set some higher requirements for this job yeah that's interesting but the first one we look at there was no these kind of requirements right so that's very interesting but I expect there's going to be a lot of this showing up in 2023 that you can apply for if you are set on doing this so yeah we're gonna monitor this so very interesting so I just wanted to show you some of my portfolio examples of fronts I have created for some clients let's just have a look at this one first so this is uh I had the client create some basic movie scripts synopsis and summarizations so these are not the most creative prompts but they worked quite good so this is just a simple URL scriptwriter your task is to write conversational short synopsis from a given text so write a short but detail conversational synapses from the following text so here I just have my placeholder for my script and just feeding like the script in here right and write a short potato conversational synapses from the text above so here are some results I ran this script from the ex machina movie so here you can see ex machina follows the story of Caleb a young programming for Long Island it wins a competition so you can say Discord Nathan has built an AI Ava and Caleb is tasked to conducting a Turing test so basically this is a synopsis of the whole ex machina script so this was nothing from the movie I just took the Bare Bones script where the lines and everything and summarized it into our synopsis of that movie so we can also have a look at some another example here I did so this is just a CV template example I did for someone so you are a professional CV writer your task is to create General CV template with user inputs create a detailed CV template that includes the following parameters for name personal information education experience goals and Ambitions three best personal characteristics create a detailed CV template for the inform up both so let's copy that and run this over at the chat GPT okay so let's run that and here you can see we have a CV that is pretty simple like it's just your name your personal information LinkedIn profile address degree institution graduation date coursework key responsibilities yeah job title short-term goal long-term goals skills to develop one two three characteristics yeah so a pretty basic CV template but the prompt works and it gave me what I wanted the client was happy and so was I so so hopefully this video gave you some inspiration maybe you want to start a career in prompt engineering or you just want to explore some more and start to learn and maybe sometimes in the next few years become one if you want to know more about this sign up for my newsletter in the link below you will get a free pdf of chat GPD prompts I use all the time and also you can check out my membership if you want to take a deep dive into generative AI also check out this video If you enjoyed this one and as always have a great day and I'll see you in the next one\n"