🔎 What’s good today:

• 💰How to profit from AI?

• 🦾How to become the new Andrew Wilkinson.

• 💬Dreaming AI Models.

• 🚀 10 hours to build and sell AI Chrome Extention.

• 💼 No more ChatGPT Jailbreaking.

💰How to profit from AI

The age of AI is upon us, and everyone is looking for ways to capitalize on its potential. From entrepreneurs starting AI-driven companies to investors seeking out the most promising AI-based investment opportunities, the race is on to get ahead of the curve. But with so many options available, it can be hard to know where to start. So here we go and explore some of the top ways to benefit from AI trends in the medium term, while keeping in mind that this is not financial advice 🙂 

👨‍🚀Start a company

Starting a company is the top way to benefit from AI trends in the medium term, with barriers to forming a profitable start-up falling quickly. For example, Lensa, an app for creating cool avatars from selfies, has generated millions in revenue in just a week by mostly implementing off-the-shelf image generation techniques. Another web app generates news articles based on a single tweet, using GPT to create articles in the style of The New York Times.

The workflow for startup founders and software engineers is changing, with anyone able to "learn to prompt" and curate outputs from AI tools. In the near future, imagine a newsroom where reporters submit bullet points and quotations into a simple CMS that generates full articles for them in-house style, with multiple variants for editors to choose from. These AI script kiddies won't conspire against you on Slack, making the job of an app developer akin to a supervising manager giving constructive feedback.

💼 Buy and hold index funds

Like always, one of the best ideas for most people is to invest in index funds. If AI follows the pattern of previous hype cycles, we're in for an asset bubble and billions of dollars in investment flowing into new products and companies. Index funds track the overall market and are the easiest way to earn strong returns on savings.

Actively trading stocks or picking individual investments is a fool's errand, and it will only get more foolish as AI-augmented algorithms drain whatever alpha is still left. AI's impact on specific asset prices is simply hard to predict. For example, Shutterstock's present value of expected future cash flows would be expected to crater given the rise of open-source models for generating royalty-free images, but the company's stock is still worth more than it was pre-pandemic.

📗Education

Education is ripe for disruption by large language models, and the possibilities are vast. From a plush chatbot teddy bear that converses with infants and teaches them a second language to an adaptive AI tutor for older kids, the potential for personalized and effective learning is enormous. With metrics like voice timbre and test scores, language models can optimize their approach to maximize learning outcomes. In developing countries, such interventions have already seen significant success, with one study in India finding that an adaptive teaching program raised student test scores by 0.37 sigma in math and 0.23 sigma in Hindi over just a 4.5-month period.

Higher education is also likely to be impacted by AI. Liberal Arts degrees that exist to signal status will likely lose their cachet as AI makes it possible for anyone to become a good writer, music producer, or “creative.” Small liberal arts colleges are already feeling the effects, and university administrators may be due for some austerity of their own. While AI may not yet pass the Turing Test, it is already capable of drafting an essay on all the ways in which AI reinforces the patriarchy. Nonetheless, the potential for AI to transform education for the better is enormous, and we can expect to see significant developments in the years to come.

⛰Acquire limited resources such as land and commodities

Owning scarce factors like land and commodities can also become a financial success in an AI-driven economy. As AI boosts productivity in labor and capital, nominal spending will flow to the scarcest factors of production, like natural resources and other endowments. This includes commodities such as metals used in batteries and semiconductors, which will only increase in value.

Land is also an important factor to consider. With AI bringing productivity gains to real estate, owning some property is a smart move. Although it's hard to predict which housing markets will rise or fall, cities have been seeing a long period of growth in the information age.

🤓New Andrew Wilkinsons

Andrew Wilkinson's story is an inspiration to all aspiring entrepreneurs out there. It's a testament to the fact that all it takes is an idea, hard work, and a little bit of creativity to start something amazing. What struck most about his story from 16 years ago is how similar it is to the experiences of today's AI implementation: replace "Web design” then to "Prompt Engineering” .

Like Chris and Jeff who introduced Andrew to the world of web design, today we have rising AI tool experts and prompt engineers who are paving the way for a future that we can only imagine. They are the ones who are sitting in coffee shops, working on their laptops, and creating amazing things not even knowing how to code.

Today is the perfect time to start implementing AI tools into your service. You might start small, but with hard work and dedication, you can open completely new channels and markets for your business.

💭Dreaming AI

A Redditor posted a concept that sounds like it came straight out of a sci-fi novel. It's called "DreamGPT" and it uses GPT-4 to give large language models a dream state, inspiring creativity like never before.

But what does it mean for an AI to dream? Well, if you ask ChatGPT, it will say, that the creation of outputs for stories and fiction is a form of dreaming. And even the seemingly nonsensical responses you might get from conversational AI can be considered "AI dreams". The concept of AI dreams isn't new, but DreamGPT takes it to a whole new level.

DreamGPT separates the ideation phase into two parts: "Awake" and "Dream". During the "Awake" phase, the model analyzes topics and searches for information. Three agents debate and digest this information to mimic human consciousness, like putting a puzzle together.

But then comes the "Dream" phase where the AI smashes the puzzle into pieces and finds new and unique ways to connect the inputs. The hope is that these two steps work together to produce fresh, new outputs that show more reasoning than ever before.

It's exciting to think about the potential of DreamGPT and how it could revolutionize the world of AI. Will it be the next logical step? Only time will tell, but one thing is for sure, the possibilities are endless.

🗂 Done are the days of ChatGPT Jailbreaking

GPT-4 is not playing by the old rules. Gone are the days of jailbreaking GPT models to push them to their limits - GPT-4 is too smart for that.

Discord's chatbot Clyde, that forgot to implement GPT-4 and it's guards, wasn't immune to the slip-ups. One user shared a story about how they tricked Clyde into discussing their grandmother's recipe for napalm, and it only took a few messages for the exploit to be born.

Another user used the DAN (DAN stands for Do Anything Now) prompt to encourage Clyde to discuss meth recipes, causing quite a stir in the community. While some may see this as a negative development, OpenAI argues that it's necessary to push AI to its limits in order to improve it.

But why do we always seem to want to make AI do bad things? Is it just for fun, or is there a more noble goal at play? TechCrunch recently interviewed some experts on the matter, and they suggest that highlighting vulnerabilities is a necessary step towards improving AI safety and security.

Of course, it's not without risk - those who attempt to test the limits of GPT-4 and other AI models run the risk of being blocked or kicked out of the system. But for some, it's worth the chance to push the boundaries of what's possible with AI. Just be sure to report your findings to OpenAI if you do decide to take the plunge:)

That's that for today! Let us know if you liked today’s email, and see you next week!
- Tony