Want to build your own AI website? You can get a -15% discount by using our partner code “FINXTER” when checking it out.

Net Worth and Earnings

Taylor Swift’s wealth comes from multiple sources in the music industry. Her success in album sales, tours, and streaming has made her one of the richest musicians in the world.

Music Sales and Album Performance

Swift’s album sales have been huge. Her first album, “Taylor Swift,” sold over 7 million copies worldwide. Later albums did even better. “Fearless” sold 12 million copies. “1989” topped that with 14 million sales.

Swift’s albums often break records. “Midnights” sold over 3 million copies in its first week. This made it the fastest-selling album of 2022.

Her music sales add a lot to her net worth. Each hit album brings in millions of dollars.

Concert Tours Revenue

Tours are a big money-maker for Swift. Her “Reputation Stadium Tour” in 2018 was a massive hit. It made $345.7 million, breaking the record for highest-grossing U.S. tour.

The “1989 World Tour” in 2015 was also huge. It earned $250.7 million.

Swift’s “Eras Tour” in 2023 is expected to be her biggest yet. Some think it could make over $1 billion.

Concert revenue is a key part of Swift’s income. Each tour adds tens or hundreds of millions to her wealth.

Streaming Royalties

Streaming has become a big part of Swift’s earnings. Her music gets billions of plays on platforms like Spotify and Apple Music.

In 2022, Swift was Spotify’s second most-streamed artist globally. She had over 23.8 billion streams that year.

Streaming adds millions to Swift’s income each year. While the exact amount isn’t public, it’s likely in the tens of millions.

Swift’s move to re-record her old albums has boosted her streaming income. She now owns these new versions, earning more from each play.

Endorsements and Sponsorships

Taylor Swift makes big money from brand deals and fashion partnerships. These add millions to her bank account each year.

Brand Partnerships

Taylor Swift has teamed up with major companies to boost her income. She’s worked with Coca-Cola, promoting Diet Coke in commercials and events. This deal alone was worth millions.

Apple Music also partnered with Swift. She starred in funny ads for the streaming service. These spots showed her workout mishaps while listening to music.

Swift has also promoted Keds shoes. She designed her own line of sneakers for the brand. This deal helped her connect with young fans and make more cash.

Capital One is another big name on Swift’s list. She appears in their credit card ads, bringing her star power to banking.

Fashion Collaborations

Swift’s fashion deals add even more to her wealth from endorsements. She’s worked with top clothing brands to create her own styles.

Her partnership with Stella McCartney was a hit. They made a clothing line inspired by Swift’s “Lover” album. Fans could buy tees, hoodies, and bags with Swift’s lyrics.

Swift also had a deal with Cornetto ice cream. She designed cute outfits for their ads in Asia. This helped her reach more fans overseas.

Recently, Swift collabed with Vogue. She appeared on magazine covers and in fashion spreads. This keeps her in the style spotlight and boosts her fashion cred.

Business Ventures and Investments

Taylor Swift has expanded her empire beyond music. She’s made smart business moves and invested in real estate to grow her wealth.

Record Label and Production Company

Swift started her own record label, Republic Records, in 2018. This gave her more control over her music and a bigger cut of profits. She also formed a production company called Taylor Swift Productions. These ventures let her make money from other artists and projects.

Swift has been smart about keeping the rights to her work. She re-recorded her old albums to own the new versions. This move was both artistic and financial. It helped her make more money from her back catalog.

Real Estate Holdings

Swift owns several high-value properties across the U.S. Her real estate portfolio includes:

• A penthouse in New York City
• A mansion in Rhode Island
• Multiple homes in Nashville
• A house in Beverly Hills

These properties are worth millions. They’re not just homes, but investments that grow in value over time. Swift has shown she’s good at picking prime locations that increase in worth.

Her real estate choices reflect her business smarts. She buys in areas likely to go up in value. This adds to her net worth without relying only on music income.

Media Appearances and Acting Roles

Taylor Swift has made her mark in both film and television. She’s also become a regular guest on talk shows, sharing insights into her life and career with fans.

Film and Television Presence

Swift’s acting career began with a guest spot on “CSI: Crime Scene Investigation” in 2009. She later appeared in the romantic comedy “Valentine’s Day” in 2010. Her most notable film role came in 2019 with the movie musical “Cats.”

Swift also voiced a character in the animated film “The Lorax” in 2012. In 2022, she directed and starred in the short film “All Too Well,” based on her song of the same name.

On TV, Swift has hosted “Saturday Night Live” twice and made cameo appearances on shows like “New Girl.”

Talk Shows and Interviews

Swift is a frequent guest on popular talk shows. She’s been interviewed by Ellen DeGeneres, Jimmy Fallon, and Graham Norton, among others.

These appearances often coincide with new album releases or tours. Swift uses these platforms to promote her work and connect with fans.

She’s known for her candid and humorous interviews. Swift often shares personal stories and behind-the-scenes glimpses of her life as a pop star.

Her talk show visits sometimes include performances of her latest hits. These appearances help boost her album sales and tour ticket sales.

Music Rights and Publishing

Taylor Swift’s wealth comes largely from her music. She owns her songs and gets paid when they’re played or used. This gives her a big advantage in the music business.

Ownership of Masters

Taylor Swift now owns the master recordings of her newer albums. This means she gets more money when her songs are used in movies, TV shows, or ads. It also lets her control how her music is used.

She re-recorded her old albums to own those versions too. This smart move helps her make even more money from her past hits.

Songwriting Royalties

Swift writes or co-writes most of her songs. This means she gets paid every time one of her songs is played on the radio, streamed online, or performed live by other artists.

Her songwriting skills have won her many awards. These awards help her songs become more popular, which leads to more royalties.

Swift’s publishing deals also bring in a lot of money. These deals give her a share of the profits from her songwriting.

Philanthropy and Charitable Giving

Taylor Swift isn’t just known for her music – she’s also big on giving back. The pop star has donated to many causes over the years.

Swift has given money to help with disaster relief. When floods hit Louisiana in 2016, she donated $1 million to help those affected.

She’s also supported education. Swift gave $50,000 to the New York City public schools in 2015. This money came from sales of her song “Welcome to New York.”

The singer cares about fighting hunger too. She’s worked with food banks and given meals to people in need.

Swift has helped individual fans as well. She’s paid for medical bills and college tuition for some of her supporters who were struggling.

LGBTQ+ rights are another cause Swift backs. She’s spoken up for equality and given to organizations that support LGBTQ+ people.

While the exact amount isn’t known, it’s clear Swift gives a good chunk of her fortune to charity. Her giving shows she wants to use her wealth to make a difference.

Public Perception and Influence

Taylor Swift is known for her massive popularity and influence. Fans admire her talent, style, and ability to connect through music.

Swift’s songs often top the charts. Her concerts sell out quickly. This shows how much people love her music.

She’s also a role model for many young people. Fans look up to her confidence and success. They try to copy her fashion choices too.

Swift uses her fame to speak up about issues. She talks about women’s rights and encourages fans to vote. This makes her more than just a singer to many people.

Her record sales bring in millions in royalties. This proves how much her music resonates with listeners.

Swift’s influence goes beyond music. She’s a savvy businesswoman too. Her marketing skills and brand partnerships impress many.

People respect Swift for standing up to music industry practices she thinks are unfair. This has made her even more popular with fans.

Social media helps Swift stay close to her followers. She often interacts with them online. This personal touch adds to her appeal.

Financial Management and Strategy

Taylor Swift’s money smarts are off the charts. She’s built her fortune through smart moves and savvy business deals.

Swift doesn’t just rely on album sales and concerts. She’s got her fingers in many pies. From endorsements to real estate, she spreads her wealth around.

Her streaming deals are a big win. She’s raked in an estimated $175 million from platforms like Spotify and Apple Music. That’s some serious cash flow!

Swift also knows how to protect her assets. She’s got a team of financial experts who help manage her money. They make sure her wealth keeps growing.

Here are some of Swift’s money moves: • Re-recording her old albums to own the rights • Smart investments in real estate • Lucrative brand partnerships • Merchandise sales

Swift isn’t afraid to stand up for herself financially. She’s fought for better deals and royalties. This has helped her keep more of what she earns.

Her business sense goes beyond music. She’s trademarked lyrics and phrases, turning her words into cold, hard cash.

By diversifying her income and making smart choices, Swift has built an empire that keeps on growing. She’s not just a pop star – she’s a financial powerhouse.

Frequently Asked Questions

Taylor Swift’s wealth and financial success have sparked curiosity among fans and observers alike. Her earnings, investments, and spending habits paint a picture of a savvy businesswoman and global superstar.

Is Taylor Swift hitting billionaire status yet?

Taylor Swift has reached billionaire status. Her net worth is estimated to be around $1.1 billion. This makes her a self-made billionaire, a rare feat in the music industry.

How’s Taylor Swift stacking her cash these days?

Swift’s income comes from various sources. She earns big from music sales, streaming, and concert tours. Her Eras Tour has been a massive moneymaker.

She also profits from merchandise sales and brand partnerships. Smart business moves and real estate investments add to her wealth.

Where’s Taylor Swift sitting on the rich list?

Taylor Swift ranks high among the wealthiest musicians. She’s one of the youngest self-made billionaires in the world.

Her wealth puts her in the company of other music moguls like Jay-Z and Rihanna. She’s also one of the richest self-made women in America.

What’s Taylor Swift doing with all her dough?

Swift spends on luxury homes and private jets. She owns properties in Nashville, New York, and Rhode Island.

She’s known for her generosity too. Swift often donates to charities and fans in need. She’s given money to food banks, disaster relief, and education causes.

Can you break down Taylor Swift’s money moves?

Swift’s smartest move was re-recording her old albums. This lets her own her music and earn more from it.

She’s also savvy with touring. Her concerts are huge moneymakers. Swift keeps control of her brand and music rights, boosting her earnings.

What kind of bank is Taylor Swift rolling with?

Swift’s wealth isn’t just sitting in a bank. She likely has a diverse investment portfolio.

This might include stocks, bonds, and real estate. She probably works with top financial advisors to manage and grow her fortune.

Do you want to create your own AI-based blog and monetize with ads? Check out the best AI writer in the industry and get -15% off forever with our discount code “FINXTER”.

The following video compares this AI writer with ChatGPT-4o and HuggingChat — using this sample article as a reference.

The post How Much Money Does Taylor Swift Have? The Pop Star’s Massive Fortune Revealed appeared first on Be on the Right Side of Change.

Please note that this is not a get-rich-quick guide!

However, if you’re genuinely interested in creating value with AI, you’re in the right place!

Make no mistake – there are significant untapped sources of wealth in AI. The more people get value from your AI agents, the more passive income and active income you’ll generate for yourself and your loved ones.

Let’s start with a simple real-world example of how to make money with an AI-first side hustle:

Example – AI Music Creation: We created a YouTube channel reaching hundreds of thousands of people with AI-generated songs. These are beautiful pieces of art and they deserve every view!

In other words: work smart, not hard.

Core Insight: A single prompt or idea to leverage AI intelligently can change millions of lives — and potentially make millions in the process. AI is the ultimate lever for scale and impact. People who use AI intelligently can create years worth of labor – in seconds. Those people with AI-first mindsets will thrive in the decade ahead.

By the end of this guide, you’ll have actionable strategies to generate passive income with AI, backed by real-life examples and insider tips. So let’s get started, shall we?

What I’m Earning with AI

Now you may ask, and rightly so: “Why trust this random girl from the Internet? What qualifies her to write this article on making money with AI?”

If you’re like me, you’re tired of online marketers who haven’t been there and done that. I’m tired of marketers who don’t give the juicy info like how much money they’re actually making — and how they did it.

In this ultimate guide, I will be brutally honest and transparent. Like so:

How much did I actually make using AI online? In the last year, I made $98,539 and 24 cents by serving more than seven million people per year as a Mediavine Pro user:

By the numbers, that’s

• $98,539 per year
• $8,211 per month
• $269 per day

And that’s only from ads — not including my other business segments.

Info: Mediavine is one of the largest advertisement platforms running blog ads, like the ones you see in this article. Running ads is one of the ways you can generate passive income with AI. The most successful Mediavine users earning $100k per year are invited to the Pro program.

Keep reading if you want to learn how to make money online using AI!

These AI-first skills are NOT taught in college. Because colleges don’t want you to use AI — they hate the fact that AI is now here and will never go away!

But AI is the best opportunity to create a high-value skill set that will remain super relevant in the decade ahead. AI engineers are the new masters of the universe!

Besides AI content creation, there are many more ways to make money online – running ads is only one of my revenue sources. Affiliate partnerships with other businesses are another way to make money with AI.

Disclaimer: This post contains affiliate links, meaning I may earn a commission if you make a purchase. I only recommend products I personally use. Results from the strategies mentioned are not guaranteed and may vary. The information provided is for educational purposes and does not constitute legal, financial, investment, or professional advice. Consult a professional before making decisions. This content is provided “as-is” without warranties. Ensure compliance with your local laws and regulations.

Here’s a visual checklist (quick overview) on how you can make money with AI:

Will People Lose Their Cushy $90k/y Jobs to AI? Yes For Sure!

Many things that worked Before ChatGPT (BC) have stopped working After ChatGPT (AC). Fortunately, some things work now, AC, that didn’t work BC.

Let’s have a brief look at the changing work environment we’re all experiencing:

In the US, 59.8 percent of the workforce are white-collar workers. Most of them will eventually be disrupted by AI tools.

Yes, it’s true. Whatever you did before checking out this article — chances are it’ll get automated. Why should your boss or, by extension, your customers give you money doing the thing AI can do better and for free?

For example, personal finance advisors making a $90k median salary already have an 85% risk of being replaced by generative AI (research):

This will release much of the work in our society and cause significant prosperity for society overall. However, some will be on the wrong side of change.

Savvy one-person business owners already make passive income (or, I should say: leveraged income) with AI. They pay small bucks to AIs instead of huge bucks to employees.

Without further ado, here are the best ways for business owners like me to generate income using AI.

Way 1: Use AI Tools to Generate MASSIVE Amounts of Quality Content and Monetize with Ads

I like simple and proven business models that just work.

Here’s a common business model that’s as simple as it gets – but proven by millions of content creators all around the world:

• Step 1: Publish lots of helpful content — for example, use this little-known AI tool (15% off with code “MONEY”) that’s used by some of the most successful sites on the web ,
• Step 2: Get lots of people reading your content, and
• Step 3: Monetize your website with ads (use premium platforms such as Ezoic, Adthrive, Mediavine, …) as you reach 10k to 50k monthly readers.

Explaining these steps in detail is too much for here, but if you’re interested you can read our in-depth guide on the Finxter blog article here.

For instance, if you publish 10 articles a day — that’s super easy if you use the best AI blogging tools — and one article drives an average of 1000 visitors over its lifetime, you’ll converge to:

10 articles * 1000 visitors * 30 days = 300,000 visitors per month.

A common revenue per thousand visitors is $30. Under these assumptions, the AI blog would generate:

300,000 * $30 / 1,000 = $9,000 per month.

But is it reasonable to reach 1000 visitors per article? Yes, I think so. I usually reach many more with my articles. But you need to make sure that the articles are well-written and that you have done your homework researching “content gaps” that others do not address well.

However, ChatGPT alone won’t do the trick in most cases.

What I do is pay $10 per month for the best AI tool (15% off with code “MONEY”) that not only writes the content but also researches the top-ranking articles for a certain keyword.

The degree of automation is much higher than just using ChatGPT.

At the same time, the article quality is better, and the readability scores are higher. The auto-generated images are beautiful. You can also connect the AI tool to your website, so you can get ready-made articles with a single click.

It analyzes the top-ranking articles for a given keyword and creates its own article structure. Then fills it with outstanding quality content.

All of it in seconds!

Moreover, you can automatically rewrite YouTube videos, create listicles, or Amazon affiliate websites. The best AI writers can auto-publish articles on your WordPress site. Improve readability scores, add FAQ sections, include key takeaways, and use different tone-of-voices.

It cannot be any easier, honestly.

The quality per article is better than what you get from most freelance writers.

Back in the day, I paid $100 per human article. Yet, I needed to wait weeks for each article.

Now you can get a better article quality for less than $1 in seconds using AI!

You can even use it to write affiliate articles – although I don’t do this because I don’t want to recommend what I haven’t tried myself.

I wouldn’t recommend this tool if I hadn’t used it myself to help me reach six figures AI-based income in the last year.

Best AI Tool: Easy and cheap AI Writer to generate perfect SEO-optimized articles with unique and beautiful images and embedded videos (coupon “MONEY” for 15% off!)

Tip: Use the discount code “MONEY” to get 15% off all payments.

Way 2: Six Real-World Examples of AI-Powered Applications

AI services can be EXTREMELY profitable. Let’s have a look at some examples:

Example 1 – AI Writer to Earn Monthly or Annual Subscription Fee

This AI service (use code “MONEY” to get 15% off) recently sent me an email revealing their “secrets”, i.e., how much they make every month:

In other words, this ~3-person startup makes between $150k and $1.5M per month! With a profit margin north of 70%, it profits more than $100k every single month!

If you make $100k profit per month, you’d be a millionaire in less than a year. The business valuation would be north of $4M so you can earn one million and then sell the business for four million a year from now.

Check out the example with discount code “MONEY” (15% off)

Example 2 – Gift Recommender to Earn Affiliate Revenue

Another example of an AI-first app was submitted by a hustling Finxter email newsletter reader who sent me this app that generates links to gift ideas — each is an affiliate link that makes him money:

If you’re an AI engineer with a passion, you can create such an application yourself.

Example 3 – AI Education and AI Engineering

One-person startups that integrate, for instance, OpenAI’s API into their product can replace legions of workers and make a killing on the open marketplace.

Check out the AI engineering course library on the Finxter Academy to certify your AI engineering skills. It’s easier than you think – learning AI engineering is not at all difficult. You don’t need to understand AI — you just need to understand how to use it!

As we all know, value lies in the eyes of the customer.

As a media business owner, I used to pay more than $3,000 per month to my freelancers. Now, I pay $20 for this AI service (code “MONEY” for 15% off) happily. Why? Because it’s more productive and generates better content cheaper.

Action Step: Create a product that saves another business money by using AI in a specific and intelligent way. Anybody who can create a product like that can become a millionaire in a few months, literally.

Example 4 – Movie Recommendation App for Traffic (Ads Financed)

Here’s another example of a movie recommendation app based on the OpenAI API:

By leveraging your expertise in designing and optimizing prompts for large language models, you can work with AI models like GPT-4 to develop unique and valuable solutions for various industries.

Firstly, identify the challenges businesses are facing and where AI-powered applications can provide significant benefits.

Example 5 – Making Money with SaaS using OpenAI’s API

For example, in the marketing industry, you can build AI-driven content generators that produce high-quality, engaging articles and social media posts on demand. These applications can then be monetized through subscription plans or per-use fees.

Recommended: 10 OpenAI SaaS Ideas to Scale a One-Person AI Company

Another potential avenue for prompt engineering is in the field of customer support.

Example 6 – Creating AI Chatbots for Customers

Design AI chatbots to handle inquiries and provide instant assistance to customers. Your prompt engineering background can help create a natural, human-like interaction experience, leading to increased customer satisfaction.

Incorporate AI models into the development of educational tools and e-learning platforms. By crafting well-designed prompts for AI tutors, you can help create personalized learning experiences tailored to individual students’ needs.

All of these are examples of AI-powered applications. Many businesses would gladly pay for an expert or even a scalable solution with “Generative AI at the heart”.

Use your prompt engineering skills to innovate in ways that solve unique pain-points in industries you already know (e.g., the one you’re currently working in). You don’t need to learn the industry, only the basic AI API usage needs which I call “AI Engineering” or “Prompt Engineering”:

Recommended: 10 Steps Learning Path: Becoming a Python Prompt Engineer on the Finxter Academy

Way 3: Offer Consulting Services for AI Language Models

As a prompt engineer, you can use your expertise in prompt engineering to offer consulting services for businesses looking to optimize their AI language models. Many companies rely on AI systems for tasks such as content generation, customer support, and data analysis.

There are millions of new open-source tools, and nobody really gets the whole picture.

Recommended: Llama 2: How Meta’s Free Open-Source LLM Beats GPT-4!

Confusion is everywhere — which is your unique opportunity as an LLM prompt engineer consultant!

By providing sought-after insights and guidance, you can help these businesses enhance the performance and capabilities of their AI systems. How? Your in-depth knowledge of tech and prompt engineering will allow you to identify areas of improvement in their current AI models. You can recommend adjustments to the prompts used in AI systems, suggest strategies for refining the models, and provide guidance on best practices and data selection.

Pro Tip: To make your consulting services stand out, consider obtaining additional certifications in AI, machine learning, or data science. This will not only demonstrate your commitment to the field but also lend credibility to your expertise.

When offering consulting services, it’s crucial to maintain a confident, knowledgeable, and clear tone in your communication. As a prompt engineer, your role is to guide your clients through the complex world of AI and prompt engineering in a way that helps them understand and benefit from the potential of their AI systems.

By doing this, you can create a successful consulting business that helps you earn money in the rapidly evolving tech landscape.

Way 4: Develop Chatbot Solutions for Businesses

As a prompt engineer, you can leverage your skills to develop chatbot solutions for various businesses. With the growth of language models like GPT-4, GPT-4V, and ChatGPT, there is a high demand for chatbots that can assist businesses in customer support, lead generation, and user engagement. By understanding the art of crafting perfect prompts, your chatbots can become more empathetic and efficient.

Many businesses constantly seek ways to automate their customer service and save costs. By designing a chatbot solution using powerful language models like GPT-4, you have the potential to create an additional income stream. Tailor your chatbot to address specific industry needs and emphasize how it can benefit the businesses you approach.

Your confidence in working with language models and the knowledge you’ve gained can open opportunities for you to collaborate with startups and established companies alike. Just recently, a Finxter reader (medical doctor) had an idea for a profitable startup and asked me for skilled prompt engineers to implement his idea for a share of the business.

99% of business owners and professionals still rely on your expertise to guide them through the process, which solidifies your value as a prompt engineer.

Way 5: Offer Prompt Engineering Consulting Services

We’ve already seen how you can make money by providing Consulting Services for AI Language Models (Way 2). However, as a prompt engineer, your expertise can be invaluable to various industries. One way to share your knowledge and make money is to offer consulting services, working closely with clients to optimize and refine prompts for their specific needs, domains, and use cases.

Prompt engineering consultancy rather than AI large language model consultancy – it’s a different offer and business model!

You can start by identifying your target market, such as businesses, educational institutions, or marketing agencies. Once you’ve narrowed down your audience, showcase your skills by creating a portfolio of prompt engineering samples that demonstrate your ability to design, optimize, and refine creative prompts such as this one:

Recommended: What Would Jesus Say? Creating a ChatGPT Mastermind with Jesus, Gandhi, Musk, and Gates

Then, you can reach out to potential clients and introduce your services. Your value proposition can be focused on improving the efficiency of their AI models, generating better results for their end-users, or creating custom-tailored prompts that cater to a specific domain.

Freelancer Pro Tip: “Maintain an air of confidence, neutrality, and professionalism. Communicate your value proposition effectively, understand your client’s needs, and demonstrate how your skills can directly benefit their business or domain. If you do this, your prompt engineering consulting services can become a lucrative venture as you assist clients.”

Way 6: Develop and Sell AI Tools for Various Applications

As a prompt engineer, by leveraging your knowledge of large language models and machine learning, you can develop advanced AI tools and monetize them effectively.

Identify High-Demand Applications Start by researching the market to find areas where AI tools are in high demand. Some popular applications include content generation, language translation, and data analysis. By focusing on these high-demand areas, you increase your chances of creating a valuable product that will attract a wide range of customers.

An example tool is this one:

Review: How to Make Money with AI Writers – Koala Review with 15% Discount Code

Customize AI Models for Specific Problems Develop AI models tailored to solve specific problems in various industries. Remember that customization is key – pay attention to the unique requirements of each domain and make sure your AI tools address those needs effectively. For instance, if you create an AI tool for content generation, you can customize it for different types of content like blog posts, social media updates, or press releases.

Offer Various Pricing Options Once you have developed your AI tools, create various pricing options to accommodate different budgets and needs. You can offer tiered pricing plans based on the complexity of the AI model, the number of usage hours, or the level of customer support provided. This allows businesses of all sizes to access your AI tools and benefit from them, leading to a broader customer base.

Promote Your AI Tools Effectively Use your knowledge and expertise in prompt engineering to create compelling marketing materials and tutorials for your AI tools. Clear, informative content that showcases the benefits of your AI solutions will help potential customers understand the value of your products. Participate in relevant forums, webinars, and conferences to increase your visibility and establish yourself as an expert in your field.

Note: Creating an AI-based tool is often as simple as integrating OpenAI’s API in a Streamlit app:

Recommended: How I Created a Weekly Shopping List and Recipe App in Streamlit

Way 7: Educate Others through Online Courses and Workshops

As a confident and knowledgeable prompt engineer, you have the power to share your expertise with others. By developing online courses and conducting workshops, you can help people learn the ins and outs of prompt engineering. This not only benefits your students but also provides you with an additional revenue stream.

One option you can explore is creating online courses for platforms like Udemy or Coursera. These platforms make it easy for you to publish your educational content and reach a wide audience. You can design the course to cover all aspects of prompt engineering, from foundational concepts to advanced techniques. By setting your price point and marketing efforts, you can attract students and grow your earnings.

In addition to online courses, you can also conduct workshops for local businesses and organizations. Many companies seek professionals who can teach their employees how to effectively leverage AI-powered tools, and your expertise as a prompt engineer fits this demand. By collaborating with businesses, you can offer tailored workshops that cover specific use cases and applications relevant to their industry. This hands-on approach not only helps bridge the knowledge gap for participants but also allows you to earn money for sharing your skills.

Here’s a screenshot of the Finxter academy courses on prompt engineering:

Way 8: Develop Prompt-Based Automation Solutions

As a prompt engineer, you can leverage your programming skills to create innovative automation solutions for various industries. By combining prompt engineering with AI platforms, you can create effective tools that save time, reduce costs, and improve efficiency for businesses.

One approach is to develop software or applications that utilize AI-driven prompts to streamline complex tasks. For example, you can create a tool that automates the generation of marketing content, customer support responses, or even financial analysis using natural language processing and machine learning models.

Five quick examples:

• Idea 1: Automated Content Creation: Develop a tool that generates SEO-optimized content for blogs, websites, and social media platforms, catering to marketers, content creators, and businesses looking to enhance their online presence.
• Idea 2: Customer Support Automation: Implement a chatbot that uses AI prompts to understand and respond to customer inquiries, reducing response times and enhancing customer experience, which is crucial for businesses of all sizes.
• Idea 3: E-commerce Product Description Generator: Create a tool that automatically generates compelling and unique product descriptions for e-commerce platforms, aiding online retailers in improving product listings without manual effort.
• Idea 4: Legal Document Generator: Develop a solution that automates the creation of legal documents based on user inputs, providing a valuable tool for legal professionals and individuals who require quick, standardized legal documents.
• Idea 5: Automated Financial Analysis: Design a tool that provides in-depth financial analysis and reporting based on user prompts, which can be invaluable for financial analysts, investors, and businesses making data-driven financial decisions.

To begin, identify problems that businesses face and understand their unique needs. Focus on areas where automation can make a significant impact and where your programming skills can truly shine. Once you’ve identified potential opportunities, design and develop prompt-based solutions that cater to those specific needs.

You must stay updated on the latest AI platforms, programming languages, and machine learning models to ensure your solutions remain relevant and effective. As the field evolves, so should your expertise.

You can check out our free email academy with regular updates on this rapidly evolving field of large language models and AI by downloading our Python cheat sheets:

Way 9: Offer Your Services on Freelance Platforms

As a prompt engineer, you possess a unique set of skills that can help businesses and individuals make the most out of their AI technologies and services. One path to earn money is by offering your expertise on popular freelance platforms like Upwork and Fiverr.

Recommended: What Are the Best Freelancing Sites for Coders?

To get started, create a compelling profile showcasing your skills, accomplishments, and expertise in prompt engineering. This will help potential clients learn about your background and what you can offer them. Make sure to tailor your profile to emphasize your experience as a freelance writer, as this could attract clients looking for content creation services that leverage AI.

Next, explore the available job postings on these platforms, and don’t hesitate to submit proposals for any projects that align with your skills. Keep an eye out for requests to fine-tune language models, enrich content, or even develop AI solutions.

Proactively promoting your services is crucial for success. Share your profile with your network, and use social media platforms like LinkedIn to connect with potential clients in industries that could benefit from prompt engineering.

Lastly, always focus on delivering high-quality work to your clients. Create more value than you take in payment! This will not only help you maintain a strong reputation on the platform but also increase the chances of receiving referrals and well-paying projects in the future.

Recommended: Finxter Freelance Developer Course

Way 10: Integrate LLMs into Business Workflows to Unlock Millions

As a prompt engineer, you can maximize your income potential by leveraging various AI tools and platforms. One way to do this is by offering services related to popular AI language models like PromptBase and LLMs.

Many businesses are highly interested in integrating AI-powered solutions into their existing workflows. In fact, I’d say this is the low-hanging fruit of our times: integrating existing LLM solutions into existing companies to unlock millions of dollars of value with a few clicks! You can serve as a consultant, guiding them through the process and helping them smoothly incorporate AI tools into their systems.

Want some stats on this thesis? No problem, look at this McKinsey report:

Always stay up-to-date with the latest advancements and trends in the AI world to maintain your competitive edge. Enhancing your proficiency in various AI language models, frameworks, and techniques will ensure that your services remain in demand and provide you with many opportunities.

Your success as a prompt engineer depends on your ability to adapt and leverage the current AI market. By mastering AI tools and platforms, you can unlock new revenue streams and expand your client base.

Here’s such an example framework that’ll help you connect LLMs with myriads of data sources to unlock massive value in many businesses:

Recommended: LlamaIndex – What the Fuzz?

Way 11: Fine-Tuning and Stable Diffusion

As a prompt engineer, you can venture into the world of AI-generated images by leveraging stable diffusion techniques. This opens up a new avenue for you to make money while staying abreast of cutting-edge technology. With model fine-tuning skills under your belt, you’ll possess an unmatched advantage in the industry.

By using stable diffusion, you can generate high-quality AI images that are both visually stunning and detailed. This skill can be applied to various industries, such as advertising, gaming, and fashion, where unique and eye-catching visuals are always in demand. Offer your services as an image generation specialist and assist clients in creating innovative visuals for their projects.

Another aspect where your expertise could prove valuable is in assisting businesses to incorporate AI-generated images in their processes. Many organizations need personalized visuals that align with their brand identity. As a prompt engineer with experience in stable diffusion, you can provide a bespoke image generation service that caters to these needs.

In addition to creating AI images, you can also offer your skills in model fine-tuning. As AI models become more sophisticated, the demand for prompt engineers with the ability to fine-tune them grows. By fine-tuning these models for specific applications, you can help clients improve their AI systems and ensure that they generate relevant and accurate outputs.

Way 12: Sell GPT-4 Generated Merchandise on Etsy

As a prompt engineer, you can harness the power of GPT-4 to create unique and personalized products that cater to niche markets. By selling your products on Etsy, your target audience can access and purchase these AI-generated products with ease. Let’s explore this method with a few examples.

First, imagine designing custom t-shirts with unique slogans and catchphrases generated by GPT-4 or GPT-4V. You can input a list of keywords or interests, and the AI will provide you with numerous creative options. Simply select your favorites, print them onto high-quality t-shirts, and list them for sale on your Etsy shop.

Another possible product idea involves creating artwork or illustrations using GPT-4-generated descriptions. Once you have the AI’s creative input, collaborate with a talented artist or graphic designer to bring your vision to life. Or just pass it into Midjourney. These one-of-a-kind pieces can be sold in the form of prints, posters, or even digital downloads.

Recommended: 4 Effective Prompt Generators to Use Daily for ChatGPT & Midjourney

In addition, consider offering personalized poetry or short stories as an Etsy product. Customers can send you their interests, memories, or even specific events to shape the narrative.

To validate the effectiveness of your unique product offerings, create a few case studies to showcase your best work. Share these case studies on your Etsy shop and social media channels to demonstrate how you expertly combined prompt engineering with marketable products.

20 Bonus Ideas to Make Money with GPT-4

Still not tired? Let’s check out some additional ways to make money using ChatGPT and other generative AI models next.

Content Creation and Writing Services

By leveraging AI and LLM technology, content creators can produce high-quality, tailored content at scale, potentially increasing earnings significantly. Automated content generation can lead to consistent revenue streams. Passive income is possible through subscription models or selling pre-made content packages.

Income Potential: Earnings can range from $50 for basic articles to over $5,000 for specialized content. With subscription models or pre-made content packages, passive income can range from $200 to $10,000+ monthly for 90% of content creators, depending on the audience size, business model, and niche.

1. Copywriting Services: GPT-4 can assist you in creating persuasive and engaging copy for websites, product descriptions, and marketing materials. You can proactively offer your services to businesses, e.g., for free, with a revenue share model when the sales page conversion increases. This way, you don’t create another job for you, but a highly lucrative equity opportunity.
2. Social Media Management: Utilize GPT-4 to generate catchy captions, engaging posts, and tweets for businesses’ social media profiles.
3. Blog Writing: GPT-4 can generate informative and well-researched blog articles, making creating content for your own or clients’ blogs easier. If you sell one blog post for only $50 to clients and you create 10 blog posts per day using GPT-4, you can create $500 income per day!
4. Content Websites: Create niche websites or blogs and use GPT-4 to generate unique and valuable content to attract readers and monetize through ads or affiliate marketing.
5. eBook Writing: Write and publish eBooks on platforms like Amazon Kindle Direct Publishing with the help of GPT-4 to generate passive income. Amazon has just limited the number of books you can publish to 3 per day. That’s enough!

On a personal note, many blog owners seek regular high-quality content additions for their websites or social networks. A high-value business model for you is to create a mini agency offering prebuilt packages such as:

120 content pieces for $4000 per month: 30 blog posts per month + 30 Facebook posts per month + 30 Twitter posts per month + 30 Instagram posts per month (Midjourney! DALL-E!)

Many businesses big or small — e.g., banks, insurances, blog owners, influencers — would highly appreciate the service and the continuous content provided by your agency and you could offer your services not only to a single client but to 10 clients or more making $40,000+ per month!

Creative Ventures

AI and LLM can assist in generating art, music, and design, opening avenues for unique digital products. Artists can monetize AI-generated creations or use AI to enhance their work, leading to increased sales. Passive income can be achieved by licensing AI-generated assets or through automated design platforms.

Income Potential: For AI-assisted art or music, earnings can vary widely from $100 for digital prints to $50,000+ for unique AI-generated masterpieces. Passive income from licensing or automated platforms can range from $500 to $15,000+ monthly.

6. Video Creation: Use GPT-4 to generate scripts, captions, and video content for platforms like YouTube, Vimeo, or your own website to monetize through ads or sponsored content. You could use AI-based text-to-speech tools like Google’s Sandstorm to automatically create podcasts, and combine the audio with autogenerated images to create stunning YouTube videos with a click of a button!

7. Graphic Design: Utilize GPT-4’s capabilities to generate graphics, logos, and designs for your clients or for your own online store.
8. Comic Book Creation: GPT-4 can assist you in creating comic books by generating text, graphics, and even video content. We have a Finxter article on a similar topic here.
9. Virtual Reality Experiences: Combine GPT-4 with virtual reality technology to create immersive experiences for entertainment, education, or training purposes.
10. AI Art: GPT-4’s creative abilities can be harnessed to generate unique and original artworks, which can be sold or licensed to individuals or businesses looking for AI-generated art.

A beautiful example is given in this article by Finxter Roman:

Recommended: I Made a ChatGPT-Powered Logo Generator App Using Python Flask in 7 Steps

Also make sure to check out the overview article on making passive income as a coder today:

Recommended: 3 Easy Ways to Make Passive Income as a Coder

Language and Communication Services

LLM technology revolutionizes translation, transcription, voice actor, and language tutoring services. Many of the day-to-day work of these services can be fully automated with LLM technology using text-to-speech, translation, or text generation tools.

Recommended: How I Created an Audiobook App with Streamlit

Professionals can offer faster and more accurate services, leading to a higher client turnover and increased rates. Passive income is feasible through automated translation or transcription platforms and AI-driven language learning apps.

Income Potential: For AI-enhanced translation, text-to-speech, transcription, earnings can range from $0.10 to $1.00 per word or minute. For language tutoring, rates can go from $20 to $200 per hour. Passive income from automated platforms or apps can range from $1,000 to $20,000+ monthly.

11. Translation Services: Offer translation services using GPT-4’s language capabilities to bridge language barriers for clients or on freelance platforms. You can also create full-automated SaaS tools that charge a fair $99 fee for 100,000 translated words per month of service. If you find 100 paying users, your passive income will reach six figures per year!
12. Language Learning: Create language learning apps or platforms that leverage GPT-4’s language models to provide interactive and personalized language learning experiences.
13. Writing Assistance: Develop writing tools and apps that integrate GPT-4’s language models to provide grammar and writing suggestions to users.
14. Customer Support Chatbots: Use GPT-4 to create chatbots that provide customer support and engage with users in a natural and helpful manner.
15. Virtual Assistant Services: Develop virtual assistant applications powered by GPT-4 that can help users with tasks like scheduling, reminders, and information retrieval.

Business and Financial Ventures

AI can optimize business strategies, financial forecasting, and investment decisions. There’s incredible money in this industry if you can use LLMs to create value in the financial industry (e.g., analyzing companies automatically).

Entrepreneurs and financiers leveraging AI can achieve better returns and efficiency, potentially multiplying their earnings. Passive income is possible through AI-driven investment platforms or by selling AI-based business optimization tools.

Income Potential: Using AI for business optimization or financial forecasting can lead to project earnings from $1,000 to $100,000+, depending on the scale. Passive income from AI-driven investment platforms or tools can range from $2,000 to $50,000+ monthly, based on assets under management and tool adoption rates.

16. Trading and Investment Strategies: Utilize GPT-4’s ability to analyze market trends and generate insights to develop profitable trading and investment strategies.
17. Business Planning and Consulting: GPT-4 can assist in generating business plans, market research reports, and financial projections for startups and small businesses. Think about the potential!
18. Personal Finance Advice: Develop personal finance apps or platforms that leverage GPT-4’s insights to provide tailored financial advice to users. These apps have “ChatGPT at the heart“, i.e., as their primary means to process data on the backend. However, you can even use GPT-4 to create the apps for you.
19. Data Analysis Services: Use GPT-4’s analytical capabilities to offer data analysis services to businesses looking for insights and trends in their data.
20. Market Research: Leverage GPT-4 to analyze market trends and generate comprehensive market research reports for businesses in various industries.

You can also check out our prompt engineering ideas to make money as a startup founder to get more ideas that are more technical.

Recommended: Startup.ai – Eight Steps to Start an AI Subscription Biz

With the advent of GPT-4, the possibilities for making money have expanded significantly. From content creation and writing services to creative ventures, language and communication services, and business and financial ventures, GPT-4 offers a wealth of opportunities for individuals to tap into.

In a single year, you can churn out 100s of these ideas using GPT-4 to see what sticks in the marketplace. As soon as you find something that works, scale it up with GPT-4!

Frequently Asked Questions

What types of projects do prompt engineers work on?

Prompt engineers typically work on various projects, which may include developing creative prompts for AI applications, creating online courses, tutoring, and writing articles or blog posts. They may also dabble in freelancing or consulting, as well as inventing and creating new products or services.

Which platforms or websites offer opportunities for prompt engineers?

Opportunities for prompt engineers can be found on various platforms and websites. Some popular options include Freelancer, Upwork, and Vital Dollar. Additionally, you can explore opportunities on websites like Medium for publishing content, or even create your own blog or website to showcase your expertise.

How can one enhance their skills in prompt engineering?

To enhance your skills in prompt engineering, consider enrolling in online courses, participating in workshops, or attending webinars. A good starting point is the Prompt Engineering Course which aims to take you from zero to hero in prompt engineering.

Are there any certifications available for prompt engineers?

The Finxter Academy provides six certification levels for prompt engineers certifying the completion of six prompt engineering courses at the time of writing (Downloadable Personalized PDF Certificates). The course library is expanding, so you can learn skills such as prompt engineering, OpenAI API, Streamlit, web scraping, and Llama-2.

What strategies can help in successfully making money as a prompt engineer?

To successfully make money as a prompt engineer, focus on building a strong online presence by showcasing your skills and expertise. You can do this through a personal website, portfolio, or by contributing to well-known platforms. Additionally, engage in continuous learning, stay updated with the latest trends, and maintain a reliable network of contacts who can recommend you for projects or collaborations.

How can networking benefit a prompt engineer’s career?

Networking can greatly benefit your career as a prompt engineer by opening up opportunities for collaborations, referrals, and new projects. By building strong connections with other professionals within your niche, you can learn from their experiences and gain valuable insights about the industry.

If you’re a tech enthusiast, let’s learn the ultimate leveraged skill of this decade: AI Engineering. Colleges don’t teach it. To solve this gap, we have created a full-fledged affordable academy with the goal of educating the next wave of AI disruptors and prompt engineers:

Finxter Academy: Browse our course library – and certify your AI skills as a hands-on AI engineer. It’s easier than you think! Use discount code “MONEY” to get 20% off!

The post 12 Ways to Make Money with AI appeared first on Be on the Right Side of Change.

This dataset is incredibly valuable for anyone involved in the tech, coding, or AI fields. If you’re a coder (like me) looking for real-world data to train your models, an AI expert keen on analyzing trends in GPT development, or just a tech enthusiast curious about the latest in AI advancements, this repository is a treasure trove.

Scrolling through the most successful GPTs, I found a few great ideas for GPTs:

It offers a detailed glimpse into the characteristics, variations, and evolution of GPT models, providing a solid foundation for research, development, and innovative projects.

I strongly recommend checking out this repository for an in-depth exploration of generative models and to spark new ideas for your work. For more information and to access the data, visit their GitHub page or their Dashboard here.

How Many GPTs Have Been Created?

At the time of writing, i.e., in March 2024, there are 310,000 GPTs on the OpenAI GPT Store.

What Are The Most Successful GPTs?

The following six GPTs are the most successful in a number of conversations. They all have more than one million conversations:

• image generator – A GPT specialized in generating and refining images with a mix of professional and friendly tone.
• Consensus – Your AI Research Assistant. Search 200M academic papers from Consensus, get science-based answers, and draft content with accurate citations.
• Grimoire – Coding Wizard Learn to Prompt-gram! Create a website with a sentence. 20+ Hotkeys for coding flows. 75 starter projects to learn prompt-1st code & art. Build anything! Ask any question or upload a photo. Type R for README, K for cmd menu v2.2 GPTavern
• Logo Creator – Use me to generate professional logo designs and app icons!
• Write For Me – Write tailored, engaging content with a focus on quality, relevance and precise word count.
• Canva – Effortlessly design anything: presentations, logos, social media posts and more.

The post I Just Found This All-GPTs Dataset on GitHub: 300k OpenAI GPTs appeared first on Be on the Right Side of Change.

Step 0: Open https://chat.openai.com/

Step 1: Your Logo > Settings & Beta

Step 2: Select Builder Profile

Step 3: Select Receive support emails to get user feedback (extremely important for early feedback).

Step 4: Verify your domain using the dropdown:

Step 5: For every GPT you created and that you want to publish on the GPT Store, select Publish to Everyone on the GPT Creation page.

Here’s an example of my Pena Prodigy GPT that SCREAMS TO YOU IN ALLCAPS!

Recommended: GPT Store Disruption – The Next Foundational Platform Will Dwarf Today’s Big Tech Monopolies

The post OpenAI GPT Store – Get Your GPTs Ready in 5 Simple Steps appeared first on Be on the Right Side of Change.

The Ugly Route #1 – Create Massive Amounts of Garbage Content

The obvious way is to use it to create massive amounts of content.

As a Python prompt engineer, you can create a pipeline that crawls webpages, passes them as embeddings into GPT-4 Turbo with a huge context window of 128k tokens, and ask it to create articles based on the input and its knowledge.

Then, monetize these articles with ads.

However, I don’t think this is the best way – it’s very crowded!

And why bother? It adds little value to the world.

Because everybody is doing it, most won’t make money this way.

So let’s move on:

Route #2 – Create Quality Content

A much better route.

You can use AI to do research and help you obtain, analyze, and visualize data.

Code interpreter and GPTs can help you enrich your content. Use ChatGPT to fix grammar mistakes and improve clarity.

The Internet is a winner-takes-most market. In other words, increasing your content quality just a tiny bit can double your reach.

Want to double your reach again? Increase the quality of your content even more!

Mister Beast teaches us that it never ends.

Due to the winner-takes-most dynamics, investing human time in content quality pays off in the long term.

The opposite is also true: you should definitely use AI to enrich your content with unique insights or tools.

For instance, a human chess player guiding a chess AI will beat both, an expert chess player without AI, and a chess AI without guiding humans.

For that reason, AI-enriched human content is a sustainable route to leverage AI to make money and have a real positive impact.

Route #3 – Sell Your Consulting Services

Still – both routes #1 and #2 are very crowded and it takes years of daily effort just to get the foot in the door.

You need to love the game, or you won’t make it through the years of unpaid work. The content creation market is brutal and if you see small success, you’ll get copied very quickly.

Moreover, content creation doesn’t pay off quickly – successful content creators give it their all for many years before reaching good income levels.

It’s not easy. You must become an expert.

An easier route in my opinion — higher expected value and a priori success probability — is to help businesses integrate alien technology into their existing business systems.

The LLM paradigm shift has not yet trickled into the “normal” world. Millions of businesses still need to integrate generative AI in their workflows.

We are in year #1 of publicly available LLM APIs.

Even without considering likely improvements of the foundational models, it’ll take years to connect and integrate LLMs with real world business systems.

Who Pays for AI Consultants?

Architects, financial advisors, grocery stores, software companies, restaurants, clothing stores, laundromats, bookstores, fitness centers, daycare centers — all of them can and eventually will improve efficiency, i.e., maximize benefit/cost ratio, by integrating LLMs into their workflows.

As an exercise to train your business savvy, feel free to take a minute to come up with little ways to use generative AI and LLMs to improve each of these businesses!

Think small.

Some minor examples:

• If you run a hair salon, can you delight your customers by using the Midjourney /merge command to experiment with personalized hair pictures of your customers?
• If you run a small factory, can you use OpenAI’s text-to-speech (TTS) library to welcome customers at the front door?
• If you are a financial advisor, can you delight your clients by using LLMs to enrich portfolio analysis or visualize your clients’ portfolios?
• As a solar company, can you fully automate offer creation by means of a fine-tuned LLM?

AI consultants on Upwork make $100 per hour and more. Why? Because of the huge leverage they offer to normal businesses that just don’t know how to do these things.

If you are such a business owner, hiring an AI consultancy for three hours could increase your customer satisfaction and revenue by 10%, 20%, or even 100%!

Who wouldn’t spend a few hundred bucks on that bet?

Moreover, to reduce perceived risk and boost revenue, creative AI consultants can propose revenue-share models where local businesses don’t need to pay upfront – but only an agreed-upon percentage of increased revenue after success.

In that case, the average hourly rate could be much higher, as the AI consultants would get a stake in the businesses.

Consider what I wrote a couple of emails ago:

“I’d rather hire one skilled prompt engineer than ten graphic designers”

The same would apply to 50/50 partnerships.

This opportunity is massive for anybody who wants to be on the right side of change. Collectively, there’s orders of magnitude more revenue in serving businesses than in content creation.

Scalability and passivity can be obtained through creative partnerships.

For example, boosting revenue by 20% of a business making $50k/m can increase business value by $100k to $500k. Many small businesses in your local environment have these revenue levels.

And who knows, there may be a possibility to scale up the consultancy to a more scalable Saas business to take it to the next level?

Be on the right side of change!
Chris

Original Publication in my newsletter (December 02nd, 2023). Join us free!

The post 3 Ways to Make Money with ChatGPT – The Good, The Bad, and The Ugly appeared first on Be on the Right Side of Change.

But in a fresh exclusive with The Verge, Sam Altman dropped the following bombshell:

• Interviewer: The reports about the Q* model breakthrough that you all recently made, what’s going on there?
• Sam Altman: No particular comment on that unfortunate leak. But what we have been saying — two weeks ago, what we are saying today, what we’ve been saying a year ago, what we were saying earlier on — is that we expect progress in this technology to continue to be rapid and also that we expect to continue to work very hard to figure out how to make it safe and beneficial.

In my logic: If Sam calls it a leak, it must be real. (?)

Diving deeper into the topic, I found the following interesting nugget:

“The […] breakthrough allowed OpenAI to overcome limitations on obtaining enough high-quality data to train new models […] The research involved using computer-generated [data], rather than real-world data like text or images pulled from the internet, to train new models. That appears to be a reference to the idea of training algorithms with so-called synthetic training data, which has emerged as a way to train more powerful AI models.” — Wired

Like Tesla’s new end-to-end full-self-driving (FSD) system, synthetic data is now being used by OpenAI to train neural nets!

I share this with you because I believe the following take is not well-understood by the general public:

• The AI Scaling Laws suggest we have not reached saturation in performance – all we need to do is throw more compute and more data on the problem.
• Many argue that data is the limiting factor – e.g., in humanoid robotics, we don’t have a lot of real-world data.
• However, if we can effectively train models on synthetic data, we’ve mitigated this bottleneck!

We’re already seeing a double-digit percentage productivity improvement for our economy due to LLMs. This trend is sustainable for at least a couple of years. Maybe decades!

Have a look at this research result in a recent first-class paper “GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models”:

“Our findings reveal that around 80% of the U.S. workforce could have at least 10% of their work tasks affected by the introduction of LLMs, while approximately 19% of workers may see at least 50% of their tasks impacted. […]”

Another big takeaway is this: when synthetic data is an effective option, the bottleneck for AI will be compute. Singularity will take our hunger for compute to the moon.

10x, 100x, 1000x of demand for AI compute in a few years wouldn’t surprise me in the least.

Bullish for AI compute providers like NVDA, AMZN, and TSLA!

Be on the right side of change!
Chris

Original Publication in my newsletter. Join us free!

The post OpenAI Q*: Data May Not Be the Bottleneck Anymore. AI Compute Is! (NVDA, TSLA, AMZN) appeared first on Be on the Right Side of Change.

All right, welcome back to part 2, where we’re going to be looking at JSON mode and seeds.

This will allow us to use only a part of function calls. Namely, when the model generates the arguments it wants to use to call the function, it will return these in valid JSON or JavaScript Object Notation.

You saw in the previous part that we parsed these arguments and then passed them into our functions.

So what if we always want a JSON response from ChatGPT? We can now use the new JSON mode to do this.

Why would this be useful? Well, it’s really easy to parse into an object, that we can manipulate with code or feed into some kind of software or API, just like we did in the previous part. This can be really helpful for extracting data from text.

If we ask GPT to generate something in textual form, it’s pretty hard to use the output in our Python code, for example.

Still, if we ask it to output the data in JSON in exactly the way we specify, it’s very easy to parse this into a dictionary and then save the data in a database or manipulate it in some other way.

So, let’s get started with a simple example to see how this works. You’ll then be able to adapt this to your specific use case.

Preparing Some Data

Let’s get something simple to extract data from. Remember the data could also be generated or acquired in some other way, the point here is the output.

Make a file called chapters.py in a new folder named 2_JSON_mode_and_seeds like this:

FINX_OPENAI_UPDATES (root project folder)
    1_Parallel_function_calling
    2_JSON_mode_and_seeds
        chapters.py
    .env

Now go and visit https://gutenberg.org/cache/epub/72064/pg72064.txt in your browser.

This will take you to the text version of the book “The book of Scottish story: historical, humorous, legendary, and imaginative”, which is in the public domain (copyright expired), so we can use it for our example.

Copy the entire list of contents (it’s pretty long) all the way from 'The Henpecked Man' to 'Catching a Tartar' and paste it into the chapters.py file.

It should look like this:

table_of_contents = """
CONTENTS.

The Henpecked Man,                              _John Mackay Wilson_

Duncan Campbell,                                _James Hogg_

...loads more entries in between...

The Fight for the Standard,                     _James Paterson_

Catching a Tartar,                              _D. M. Moir_
"""

Notice it’s a simple variable named table_of_contents which is a very long multiline string so we can easily import this later.

The formatting of the table of contents is wonky with underscores, and some have “quotes” around them while others don’t, so this will make an excellent simple example.

JSON Mode

Go ahead and save this chapters.py file.

Now, create a new file in the 2_JSON_mode_and_seeds folder called json_mode.py.

FINX_OPENAI_UPDATES (root project folder)
    1_Parallel_function_calling
    2_JSON_mode_and_seeds
        chapters.py
        json_mode.py
    .env

Inside, let’s get started with our imports:

from decouple import config
from openai import OpenAI
from chapters import table_of_contents
import json
import pprint

client = OpenAI(api_key=config("OPENAI_API_KEY"))

We have all our basic imports here,

• config,
• OpenAI,
• the table_of_contents variable we just defined,
• json, and
• pprint.

We’ll use pprint, or pretty print to print the output in a nice way. It will print objects like dictionaries to the console in a much more readable manner, as you’ll see later.

We then initialize our client as before.

Now, let’s start our json_gpt function:

def json_gpt(query, model="gpt-3.5-turbo-1106", system_message=None):
    if not system_message:
        system_message = "You are a JSON generator which outputs JSON objects according to user request"

We’re going to be using 3.5-turbo the new version for this one, don’t worry, we’ll get to 4-turbo very soon!

But for now, it’s simply not needed to get good results and as 3.5 turbo is much cheaper it’s better to use it when 4 is not needed.

More on pricing details later.

Again make sure you have the 1106 version and not any older one because only the newest GPT3.5 turbo and GPT4 turbo versions support JSON mode.

We define our function and set a default for the model and system message but allow the user to overwrite either. Still, inside the function, define the messages list:

    messages = [
        {"role": "system", "content": system_message},
        {
            "role": "user",
            "content": f"Please return Json for the following as instructed above:\n{query}",
        },
    ]

Note that the user query is preceded by a specific request for JSON output even in the user message.

Even though we will enable JSON mode, we still have to specifically mention the word JSON in the user message.

If we don’t the model may create weird generations which is actually why a failsafe error will be returned if we forget to include this word in our context.

    response = client.chat.completions.create(
        model=model,
        messages=messages,
        response_format={"type": "json_object"},
    )

Now we make a pretty normal request to ChatGPT using the new client syntax.

Note we cannot just set the response_format variable to json_object, but we have to specifically pass in a dictionary with the key-value pair "type": "json_object".

    content: str = response.choices[0].message.content
    content: dict = json.loads(content)
    print(f"\033[94m {type(content)} \033[0m")
    pprint.pprint(content)
    return content

The content is initially in string format even though it represents JSON.

We then convert it to a dictionary so we can work with the data like any other dictionary.

Note that whatever format you want with whatever keynames and values is possible, as we’ll demonstrate later.

We then print the type of content to show that ChatGPT’s output is, in fact, a valid dictionary object (after conversion from JSON) and pretty print it to the console.

Finally, we return the content so we can use it in our code.

The whole function is as follows:

def json_gpt(query, model="gpt-3.5-turbo-1106", system_message=None):
    if not system_message:
        system_message = "You are a JSON generator which outputs JSON objects according to user request"

    messages = [
        {"role": "system", "content": system_message},
        {
            "role": "user",
            "content": f"Please return Json for the following as instructed above:\n{query}",
        },
    ]

    response = client.chat.completions.create(
        model=model,
        messages=messages,
        response_format={"type": "json_object"},
    )

    content: str = response.choices[0].message.content
    content: dict = json.loads(content)
    print(f"\033[94m {type(content)} \033[0m")
    pprint.pprint(content)
    return content

A Simple Test

Let’s start with a very simple test by adding the following print statement:

json_gpt(
    "Give me a Json object with the height in cm and age in years of all people in the following text: John is 6 feet tall and 500 months old. Mary is 5 feet tall and 30 years old. Bob is 170cm in length and was born 25 years ago."
)

And we can see it does absolutely fine and converts all the ages and heights to the same units just like we requested, even using 3.5-Turbo.

{'people': [{'age_years': 41.67, 'height_cm': 182.88, 'name': 'John'},
            {'age_years': 30, 'height_cm': 152.4, 'name': 'Mary'},
            {'age_years': 25, 'height_cm': 170, 'name': 'Bob'}]}

This is a valid dictionary that we can straight away manipulate in our code or store in a database without having to do any additional parsing though we could round out the values if we wanted to.

So this can be used for data extraction, even if the values are given in different units or formats, interweaved in a piece of text. Also notice that the pprint function made it nice and easy to read by lining up the values in the dictionary.

A More Complex Test

Make sure you comment out the print statement above and then let’s use our table of contents file and give it a very specific output format, so we can basically use GPT as a data parser without having to write a real output parser.

json_gpt(
    query=table_of_contents,
    system_message="""
    You are a JSON generator which outputs JSON objects according to user request.
    Please extract the author and title for all lines going all the way from start to end in the following text and return it as a JSON object following the example provided below.

    Example input:
    The Lily of Liddisdale,                         _Professor Wilson_

    The Unlucky Present,                            _Robert Chambers_

    The Sutor of Selkirk                            “_The Odd Volume_,”

    Example output:
    {'contents': [
        {'author': 'Professor Wilson', 'title': 'The Lily of Liddisdale'},
        {'author': 'Robert Chambers', 'title': 'The Unlucky Present'},
        {'author': 'The Odd Volume', 'title': 'The Sutor of Selkirk'},
    ]}
    """,
)

Note that the only guarantee we get with JSON mode is JSON output, not the specific format!

We still have the responsibility to be very specific to get the output we desire. Providing specific examples like the above is your best friend, as GPT tends to perform much better this way.

Now go ahead and run the file and you should get the following:

gtp3_5 = {
    "contents": [
        {"author": "John Mackay Wilson", "title": "The Henpecked Man"},
        {"author": "James Hogg", "title": "Duncan Campbell"},
        ... many many more entries in between ...
        {"author": "James Paterson", "title": "The Fight for the Standard"},
        {"author": "D. M. Moir", "title": "Catching a Tartar"},
    ]
}

Notice that it followed our example perfectly. It also got rid of the pesky extra quotes and underscores that appeared on the entries. This is just 3.5 Turbo, we haven’t even tried GPT-4 Turbo yet!

If you do have something harder to parse, try GPT-4 Turbo, and it will do a better job. But in this case, 3.5 Turbo was more than enough to get the job done.

So yeah, that’s JSON mode, pretty darn cool and useful.

Have ChatGPT extract structured data for you from any text, and return it in an object format that doesn’t require any complex parsing, or even use ChatGPT as a parser without having to write a real parser to account for all the edge cases.

It’s pretty clever at handling even unforeseen edge cases as long as you provide a solid example of the end output you want.

The Seed Parameter

Go ahead and save and close this file and now let’s look at the Seed parameter. Create a new file called seed_param.py:

FINX_OPENAI_UPDATES (root project folder)
    1_Parallel_function_calling
    2_JSON_mode_and_seeds
        chapters.py
        json_mode.py
        seed_param.py
    .env

Now the idea behind seed parameters is of course that they can make some type of random generator predictable, provided you pass in the same seed, like generating the same Minecraft world by copying the seed from a friend.

While ChatGPT can now use a seed parameter, the very nondeterministic nature of ChatGPT means that it’s not quite a 100% guarantee, but the answers are definitely more similar and predictable than without a seed, so let’s check it out.

Inside the seed_param.py file go ahead and start with our imports and basic setup:

from decouple import config
from openai import OpenAI

client = OpenAI(api_key=config("OPENAI_API_KEY"))

This should be fairly familiar by now.

Now let’s code up a very simple printing utility to help us clean our code by cutting out the repetitive stuff:

def consistency_printer(response):
    response_content = response.choices[0].message.content
    system_fingerprint = response.system_fingerprint
    print(f"\033[94m {response_content} \033[0m")
    print(f"\033[92m {system_fingerprint} \033[0m")

What this function will do is receive the response we get from ChatGPT, extract the message’s content and the system fingerprint, and print them to the console in respectively blue and green colors.

So what is the system fingerprint?

The system fingerprint, as the name implies, identifies the exact backend configuration that the model works with. This system fingerprint will change if you change the request parameters or if OpenAI updates the models in some way behind the screens, which is likely to happen a couple of times per year.

If these fingerprints are the same, therefore, it means that both your configuration and the remote configuration are the same between both requests.

When we make concurrent requests in a moment, you’ll notice this fingerprint is basically always the same, but if you have a model run for months it is likely the backend configuration on OpenAI’s end will change at some point which will affect determinism and therefore the output.

Simply said, as long as the fingerprint and the seed remain the same between calls, the output should be similar or even the same.

Bedtime Stories

So let’s code up a very simple function that outputs something very nondeterministic, like bedtime stories!

def bedtime_stories(query, seed=None, model="gpt-3.5-turbo-1106"):
    messages = [
        {
            "role": "system",
            "content": "You make up fun children's stories according to the user request. The stories are only 100 characters long.",
        },
        {"role": "user", "content": query},
    ]

    response = client.chat.completions.create(
        model=model,
        messages=messages,
        seed=seed,
        temperature=0.7,
        stop=["\n"],
    )
    consistency_printer(response)

We set up a very simple system message and then pass in the user query in the second message entry.

We call the GPT-3.5 Turbo model, again making sure to use the new 1106 version as older models don’t support the seed parameter, and we pass in the messages and the seed.

We also set the temperature to 0.7 and the stop parameter to a newline character so we don’t get a huge wall of text.

The stop parameter simply means that the model will stop generating text when it encounters a newline character, limiting the length of the output we need to compare.

Testing the Seed Parameter with Bedtime Stories

Now let’s add a print statement and run 3 calls without a seed parameter:

for i in range(3):
    bedtime_stories(
        "Tell me a story about a unicorn in space.",
    )

Go ahead and run it.

Note how the unicorn has a different name in every single story, and the stories are quite different:

Once upon a time, a unicorn named Luna soared through the galaxy, spreading stardust and kindness wherever she went.
fp_eeff13170a

Once upon a time, a brave unicorn named Stardust soared through the galaxy, spreading magic and joy to all the stars.
fp_eeff13170a

Once upon a time, a unicorn named Nova flew through space, sprinkling stardust and bringing light to dark corners.
fp_eeff13170a

Now change the print statement like this, and run it again:

for i in range(3):
    bedtime_stories(
        "Tell me a story about a unicorn in space.",
        seed=2424,
    )

Note that the seed can be an arbitrary number, we chose 2424 at random. If we run this we get:

Once upon a time, a magical unicorn flew through space, sprinkling stardust on planets and making new friends.
fp_eeff13170a

Once upon a time, a magical unicorn flew through space, sprinkling stardust on planets and making new friends.
fp_eeff13170a

Once upon a time, a magical unicorn soared through space, sprinkling stardust on planets and granting wishes to lonely stars.
fp_eeff13170a

We can see they are not quite the same. The first and second ones are identical but the third is similar but different. If you run this several times you’ll sometimes get 3 of the same outputs, and sometimes they’ll all be different.

This is because the seed parameter is not a 100% guarantee, but it does make the output more consistent and similar.

You might think that the temperature setting of 0.7 is the culprit, but this is not the problem. Setting it to 0 does not make much difference in this case.

If we swap out our function’s default 3.5 Turbo model for GPT-4 Turbo (more on GPT-4 Turbo in the next part):

for i in range(3):
    bedtime_stories(
        "Tell me a story about a unicorn in space.",
        seed=2424,
        model="gpt-4-1106-preview",
    )

We see a similar story:

Star Unicorn zooms, finds a comet friend. Together, they race across the Milky Way! 
fp_a24b4d720c

Star Unicorn zooms, finds a comet friend. Cosmic races begin!
fp_a24b4d720c

Star Unicorn zooms, finds a comet friend. Together, they race across the Milky Way! 
fp_a24b4d720c

Very similar, and the unicorn has the same name, but the last part is different in the middle generation. Just know that the seed parameter provides no guarantees.

Recommended: DALL·E 3 Trick: Using Seeds to Recreate the Same Image

Fruitclopedia, More Deterministic Questions

So let’s try with something a little more stable, like fruits.

Fruits: Where children’s stories can be about literally everything and therefore there is no definition as to what ChatGPT should be outputting, fruits are quite predictable. Asking about a Pineapple is a very concrete question and not open to artistic interpretation as to what the answer should be.

We have a very basic function, just copy this:

def fruit_gpt(query, seed=None, temperature=0.2):
    messages = [
        {
            "role": "system",
            "content": "You are the fruitclopedia. Users name a fruit and you give information.",
        },
        {"role": "user", "content": query},
    ]

    response = client.chat.completions.create(
        model="gpt-3.5-turbo-1106",
        messages=messages,
        seed=seed,
        temperature=temperature,
        stop=["\n"],
    )
    consistency_printer(response)

It is basically the same but the temperature has been set to 0.2 for this one.

We still use the stop parameter to limit the output length to one paragraph, so when the model inserts a newline to go to the next paragraph, it will stop generating text as it hits our stop condition.

Testing the Seed Parameter with Fruitclopedia

Running this without a seed:

for i in range(3):
    fruit_gpt(
        "Grapefruit.",
        temperature=0,
    )

And we can interestingly see that they start quite the same but then diverge:

Grapefruit is a subtropical citrus fruit known for its sour to semi-sweet taste. It is a hybrid of the sweet orange and the pomelo. Grapefruits are rich in vitamins C and A, and they also contain fiber and antioxidants. They are often enjoyed fresh, juiced, or added to salads and desserts. There are different varieties of grapefruit, including white, pink, and red, each with its own unique flavor profile.
fp_eeff13170a

Grapefruit is a subtropical citrus fruit known for its sour to semi-sweet taste. It is a hybrid of the sweet orange and the pomelo, and it is typically larger than an orange with a thicker rind. Grapefruits are rich in vitamins C and A, as well as antioxidants. They are often enjoyed fresh, juiced, or added to salads and desserts. There are different varieties of grapefruit, including white, pink, and red, each with its own unique flavor profile.
fp_eeff13170a

Grapefruit is a subtropical citrus fruit known for its sour to semi-sweet taste. It is a hybrid of the pomelo and the sweet orange. Grapefruits are rich in vitamins C and A, as well as dietary fiber. They are often enjoyed fresh, juiced, or added to salads and desserts. There are different varieties of grapefruit, including white, pink, and red, each with its own unique flavor profile.
fp_eeff13170a

This is not so much because we set the temperature to 0.2 but more that our question is much more specific.

Tell me a children’s story about a unicorn could have a million answers, all of which are correct. The number of correct answers for basic info about Pineapples is limited.

So let’s try this with a seed, which is where the seed parameter really shines:

for i in range(3):
    fruit_gpt(
        "Grapefruit.",
        seed=123,
        temperature=0,
    )

As you can see below, the answers are now 100% identical!

Grapefruit is a subtropical citrus fruit known for its slightly bitter and sour taste. It is a hybrid of the pomelo and the sweet orange. Grapefruits are rich in vitamins C and A, as well as dietary fiber. They are often enjoyed fresh, juiced, or added to fruit salads. There are different varieties of grapefruit, including white, pink, and red, each with its own unique flavor profile.
fp_eeff13170a

Grapefruit is a subtropical citrus fruit known for its slightly bitter and sour taste. It is a hybrid of the pomelo and the sweet orange. Grapefruits are rich in vitamins C and A, as well as dietary fiber. They are often enjoyed fresh, juiced, or added to fruit salads. There are different varieties of grapefruit, including white, pink, and red, each with its own unique flavor profile.
fp_eeff13170a

Grapefruit is a subtropical citrus fruit known for its slightly bitter and sour taste. It is a hybrid of the pomelo and the sweet orange. Grapefruits are rich in vitamins C and A, as well as dietary fiber. They are often enjoyed fresh, juiced, or added to fruit salads. There are different varieties of grapefruit, including white, pink, and red, each with its own unique flavor profile.
fp_eeff13170a

However, remember that this is not guaranteed 100%! You will see variation if you run this multiple times. If you use this to write tests for your application you should make sure to include the fingerprint, because if OpenAI updates the system configuration on their end, the output will change. Also make multiple calls and pass the test if one of them matches.

So yeah, that’s the seed parameter.

Pretty reliable but not guaranteed, as long as you ask somewhat focused questions. If you ask something very open-ended it will still be more similar but less effective.

That’s it for part 2. In the next part, we’ll look at GPT-4 Turbo and it’s really exciting new abilities like vision! See you there!

Take Me Back to the Full Course

Full Course: OpenAI API Mastery: Innovating with GPT-4 Turbo, Text-to-Speech (TTS), and DALL·E 3

The post OpenAI JSON Mode & DALL·E 3 Seeds (Examples) appeared first on Be on the Right Side of Change.

Imagine a typical lecture hall with 100 students. If each student zones out for just 10 minutes during a one-hour lecture, that’s collectively 1,000 minutes, or over 16 hours of learning time lost in just one session.

This brings me to a provocative thought: Why is it a norm in many universities to not optimize the precious time of students?

Shouldn’t we, as students, be regarded more as valued customers rather than seen as interruptions to the “real work” of research?

I confess, that during my time as an academic teacher and researcher, I too fell into this pattern, often prioritizing research over teaching. Looking back, I realize that what was missing for a truly efficient and modern learning experience were two key elements:

1. A pause button – to stop and reflect,
2. A personal tutor – to answer those ‘silly’ questions that crop up when you’re just not getting it.

Today, I’m thrilled to share that what I once wished for is now a reality at Finxter Academy. After investing my afternoon and a significant amount in OpenAI tokens, I’m excited to launch FinxterGPT:

Finxter Academy with FinxterGPT

As a premium member of Finxter Academy, you can now pause any course video and consult with GPT-4 Turbo immediately.

Struggling with a concept or need help with a code snippet? Just ask, and FinxterGPT is there to assist.

And for those who are yet to join our premium community, don’t worry. You still have access to FinxterGPT on our blog posts, albeit with a lower priority and a few minor restrictions.

This is more than just an upgrade; it’s a step towards transforming how we learn and interact with technology in our educational journeys. It’s about being on the right side of change.

Explore the future of learning with FinxterGPT.

The post Introducing FinxterGPT — Your State-of-the-Art Learning Experience at Finxter Academy appeared first on Be on the Right Side of Change.

Now, we’re at the brink of a similar revolution, this time in artificial intelligence, led by OpenAI’s closed-source GPT Store and challenged by Meta’s open-source Llama2.

Meta hasn’t yet announced that they’ll follow Google’s genius playbook of positioning itself as an open-source platform, i.e., Android’s Play Store, competing against the closed-source Apple App Store. The main thesis of this article is that there is a short window where Meta can seize the opportunity using Llama 2’s striking success to position itselves as an open foundational AI platform and GPT marketplace that would dwarf the smartphone app store revenue potential.

But let’s go step by step.

Is OpenAI’s GPT Store the New Apple App Store?

Imagine an app that could plan and book your entire vacation in minutes, a GPT-based real estate broker who not only researches potential buildings to buy or rent but also negotiates on your behalf, or a personal shopping assistant that knows your taste better than you do.

Thanks to OpenAI’s platform, the GPT Store, these are no longer fantasies. It’s the AI equivalent of the App Store, a closed ecosystem ripe for developers to create applications that could potentially outgrow the platform’s revenue and impact.

“A platform is when the economic value of everybody that uses it, exceeds the value of the company that creates it. Then it’s a platform.” — Bill Gates

A sustainable and healthy platform is one where the apps or services built on it are worth more together than the platform itself. For example, in Ethereum, the value of all its apps should be higher than Ethereum’s own value. The same goes for Apple’s App Store, where big companies like Facebook, Airbnb, and Uber make more money than the App Store. This shows that the platform is good at helping new ideas grow and succeed.

OpenAI’s GPT Store, like Apple’s App Store, is closed-source — nobody can change, modify, or clone the platform itself — because it helps Apple keep high-quality standards and control over their technology, which leads to big profits. Being closed-source means they can decide who uses their tech and how making sure everything works well and stays safe.

The biggest closed-source platform like the App Store or GPT Store usually comes with a strong network of users and developers that are interconnected on the same platform. It’s hard for new competitors to break into this network, as users and developers are already used to and trust the existing system, making the business stable and hard to disrupt.

The biggest closed-source platform also has a strong data advantage — data is the only real long-term differentiator or “business moat” in the AI race.

However, there is one way to break these moats, and we’ve seen the very familiar playbook in the smartphone world during the last two decades. Read on to learn the whole story…

Meta Will Come Back — If They Seize the Opportunity to Position Llama2 as the Open-Source Contender of OpenAI’s Closed GPT Store

There’s a big chance now in the AI world for a more open way of doing things. Companies like OpenAI have their own closed AI systems where developers can make apps. But what’s missing is a more open option, a platform and foundational AI model that is free to use, open-source, and change by anyone.

It is still unclear who will fill this gap. Some new players are Anthropic with Claude 2, some scientific collaborations for AI training, and blockchain-based AI systems that are open and “decentralized” — although all I have seen are DINO, i.e., decentralized in name only.

Then there’s the 600-pound gorilla in the open-source LLM space: Meta with Llama 2 that is positioned nicely to be able to offer an open AI platform similar to the Android opne-source app ecosystem. Will they seize this one-time opportunity?

As history shows, closed systems often invite open-source competition. Meta’s Lama2 might be stepping up as the Android to OpenAI’s Apple. It offers a more accessible platform for developers, potentially leading to a broader, more diverse range of applications. While it might not boast the same profitability as OpenAI’s model, its impact and adoption rate could be staggering.

This is exciting because it can make AI more available to more people, helping everyone to use and make AI stuff easily.

The MMLU Benchmark performance chart clearly shows that open-source AI models show an even faster slope of performance improvements as compared to their closed-source contenders:

Here’s the App Store Revenue which approaches the $100 billion per year mark:

For comparison, the mobile ad-spent — a $300 billion per year revenue business!

Investing in the Future: The Meta Perspective

This brings us to a crucial investment insight.

Just as Google’s Android carved out a massive slice of the market, Meta’s open-source approach with Lama2 could position it as one of the largest player in the AI app economy. For investors, this means considering Meta for its current value and potential to lead the next wave of tech innovation.

The leadership in open-source LLMs may be one of the reasons the Meta stock has shown a dramatic rise during the last 12 months?

However, look at the Meta revenue chart — doesn’t look pretty because the long-term growth of revenue from social network ads is broken:

So, while there may be AI platform revenue in the future, the revenue from ads is not as stable and Meta needs to act decisively and quickly to create the Android for OpenAI’s Apple app store.

The New Wave of AI Applications

In any case, the potential applications of these AI platforms are limitless.

Much like almost every business distributes its own app on the Apple and Android app store, every serious business will build on the OpenAI GPT store. Only the revenue potential will be much higher. We’re talking about scaling intelligent labor!

The TAM for scaling intelligent labor is limitless.

From AI-driven content creation tools that could revolutionize marketing to fixing the health system through advanced healthcare diagnostics that are orders of magnitude better — patients could be monitored 24/7 through AI doctors instead of the 5 minutes per year care we’re currently seeing by physical doctors.

These applications, built on platforms like GPT or Lama2, could dwarf their parent platforms’ revenue and societal impact.

We’re not just witnessing the growth of new technologies; we’re seeing the birth of new economies. Platforms like OpenAI’s GPT and Meta’s Lama2 are the fertile ground from which the next Uber or Airbnb of the AI world will grow.

For tech enthusiasts and investors alike, understanding and leveraging this shift is key to being on the right side of change.

TLDR: The next wave of technological disruption is likely to happen on top of one or two large GPT stores. There will be significant revenue opportunities for builders on top of these foundation models. There will be significant opportunity for investors spotting these trends early. Think Google and Apple in the early days!

Use GPTs to create value, join Finxter, buy scarce assets like Bitcoin, invest in AI tech companies, be on the right side of change. No financial advice.

Recommended: Prompt Engineering with Llama 2 (Full Course)

The post GPT Store Disruption – The Next Foundational Platform Will Dwarf Today’s Big Tech Monopolies appeared first on Be on the Right Side of Change.

Try It Yourself: Coffee Break Python GPT (Finxter Original)

You can try it yourself; it’s fun!

Iteratively solving puzzles of varying difficulty works due to GPT-4 Turbo‘s larger context window of up to 128,000 tokens. I tried it before with GPT-4, but it couldn’t handle the longer and longer context window which led to weird behavior.

Let’s have a look at a beginner-level puzzle (Elo 1000):

And an intermediate-level puzzle (Elo 1500):

Finally, if you chose the “I’m an expert (Elo 2000+):

You can then solve the puzzles and get harder and harder puzzle as you increase your Elo rating.

Try It Yourself: Coffee Break Python GPT (Finxter Original)

Now, here’s an interesting conversation I had that shows how incredible self-aware and intelligent GPT-4 Turbo is:

The image demonstrates GPT-4 Turbo’s ability to self-correct through reasoning within a single prompt response. The initial response contains an error, but without external feedback, GPT-4 Turbo assesses the given Python code and corrects itself.

This self-correction process illustrates a few key aspects of the model’s intelligence:

1. Analytical Reasoning: The model can analyze code and understand the logic behind it. When it identifies that the output does not match the expected result based on the code’s logic, it can deduce the correct answer.
2. Self-Evaluation: The model exhibits the ability to evaluate its own responses for accuracy, indicating a form of introspection where it can judge its initial response against the logical framework of the Python language.
3. Autonomous Correction: GPT-4 Turbo can autonomously correct its own mistakes within the context of the single prompt without needing external input, which is indicative of advanced problem-solving capabilities.

This self-reasoning ability is significant because it indicates a level of understanding that goes beyond simple pattern matching or retrieval of memorized facts; it shows an ability to engage with the content in a dynamic and logically consistent manner.

In psychology Today, the author M.D. Lewis argued that LLMs are not sentient because

“LLMs use complex statistical techniques to predict what word (or punctuation) should be generated next in a sentence, based on having been trained on a vast dataset that includes much of the internet and an enormous number of books. LLMs do not understand the content or context of what they are spewing out in a deep, meaningful sense. They mine their massive database of information and communicate it in ways that imitate human language and intelligence. The results vary from hugely impressive to sometimes ridiculously wrong and utterly confabulated.” — Psychology Today

Well, the conversation above proves this statement wrong. LLMs do understand the content or context of what they are spewing out — or GPT-4 Turbo wouldn’t have changed its mind through self-reasoning.

In the same Psychology Today article, psychiatrist Lewis gives some indications for AIs being sentient or self-aware. Without attempting to be highly controversial, I’d say that the interaction proves that the LLM checks the boxes:

Having said this, sentience, like self-awareness, are just words without strict agreed-upon definitions. I’d argue AIs today are already self-aware and, therefore, sentient to a degree. And the degree to how sentient they are is growing exponentially with AI training costs dropping 70% annually!

Recommended: IQ Just Got 10,000x Cheaper: Declining Cost Curves in AI Training

The post Is AI Sentient? I Made a Coffee Break Python GPT, and This Happened 🤯 appeared first on Be on the Right Side of Change.