Could Recent Leap in AI Lower Barrier to Start Up New Businesses?
TL;DR Artificial Intelligence will be game changer in leveling up next-gen entrepreneurship
What Artificial Intelligence actually is?
Despite all the recent hype, Artificial Intelligence has been de-facto assisting and co-piloting our work & everyday life for number of past decades. When talking about AI, people tend to relate it of something complicated. However, many don’t know that in it simplest form, AI is just a set of programmatic rules, logic and algorithms which are based on mathematics and statistics we’ve all been taught in school.
But as we know - research never sleeps. Constant improvements in material technology lead to improvements in transistors which leads to faster processors (check out Moore’s Law on this / PS. sad hearing that Moore just passed away 😔) which, in turn, help innovating on faster and more efficient algorithms to be utilized. Basically this is short version of how we got in todays’ reality with GPT-kind technology. If you think development it’s over - think twice. While GPT cannot check all the marks for real AGI1 - reality is that full-blown AGI-implementations are in the works as of writing this post and are waiting perhaps already around the corner.
The change is real
Early Web3 adopter, investor and founder of Outlier Ventures Jamie Burke discussed this topic of next “unicorn-companies” being kickstarted with simply power of AI. He described so-called “solo founder startup with no employees” will leverage GPT & AGI like technologies to mass-reproduce all the technological stack, marketing and sales processes up to the point “where venture capital won’t be needed”.
Building a business with the help of OpenAI’s ChatGPT
To understand how realistic this would be, I decided to give a try. It was decision for me to narrow down by scoping into Web3-context.
The business idea
To start with, obviously I need a business idea. I got number of alternatives: NFT Marketplace, Decentralized Social Network, Web3-related Education-business, Web3-enabled e-commerce and DAO consultancy.
Most appealing to me was the idea of Web3-enabled e-commerce business because it’s still true that platform such as Amazon or Shopify are dominating in this space. I could create a platform that allows for peer-to-peer transactions, thus eliminating middlemen, and providing a more transparent and secure way for buyers and sellers to exchange goods and services.
Technology stack
Answer to this one was quite broad and I had to make some additional questions to really deep dive specific answers.
Basis:
Ethereum-based blockchain stack, preferably utilizing Polygon L2-technology
Smart contracts:
Solidity programming language + Web3.js library to interact with Ethereum node
Data storage layer:
IPFS protocol, more specifically Pinata
DeFi protocols:
Uniswap protocol to access to facilitate decentralized exchange of tokens + Aave to provide decentralized lending and borrowing services.
User authentication & blockchain data stream via API:
Moralis.io
Overall solid suggestions from what I’m familiar with and been experimenting with them.
Marketing strategy
There are tons of different approaches when it comes to creating marketing strategy for the product. Many experienced marketers have stated that “there are as many marketing strategies as marketers themselves”. This means that there’s “no one-size fits all”-approach.
Speaking of which, GPT suggested me very generic set of tools such as
scoping the target audience
developing brand identity
utilising all possible social channels (Discord, Twitter, Lens, Meta)
leveraging influencer marketing per need
Marketing material like text, landing page, forms, email campaigns and whole brand identity will be created by tools such as DALL-E, ChatGPT and Microsoft 365 copilot.
On general, marketing-topic seemed to be difficult topic for GPT since this is the most creative part which needs extra more drill downs to reap the benefits. However, this is completely doable.
Working foundation for e-commerce store landing page was done under 30 seconds - that’s impressive! The rest can be done by hand but that’s besides the point. The fact that these tools can really save us valueable time to deliver faster - is real.
Analytics
To understand the flows and potential customer behavior, I need tools and strategy. I got advices to use tools such as SEO (=Google), survey tools (=Typeform), predictive analytics platforms (=Google), CRM (=HubSpot).
When it comes to analytics & customer strategy, GPT provided me the with very general approaches ones courtesy of Reddit, Quora and Stackoverflow 😃
Paradigm shift - from consumption to creation mode?
What I’ve learned from this experiment? A ton. If you are seasoned professional with background working natively with IT-tech, these tools works like cheat codes for Gran Theft Auto😎 On the other hand, novelty of creation has, to a degree, disappeared for me. And this is more worrying.
There are officially seven wonders in the World 🗺️
There are officially seven wonders in the World 🗺️
8th unofficial wonder is compounding effect 📈
9th unofficial is AI - those who utilize it will stay ahead of the game 🌅
Afore mentioned example experiment is simple but showcases how AI revolutionized in the matters of months the way how virtual online service/business could be built. This means that today, we are able to create ton of simple little “helper” servicing apps for needs: learning, productivity, outsourcing recurring tasks. Gone are days of long, costly application/service development. I’m able to have a conversational partner who’s always ready to advise me on whatever I need. It helps me to create anything I have in mind. Therefore I think GPT-, generative and conversational AI is a revolutionising technologies: instead of passive consumption like social media, these tools provide the ability to create up to meaningful things that will help us to focus on other more intelligent matters in our lives (where AI is still not smart enough) 😅
Wrap up and what I learned?
In a sense, GPT AI-technology is not so special Large Language Model-based model2. Its power is based on huge datasets collected over years and in fact - wasn't even planned to be realized in public. In the recent news we have heard how GPT-4 technology is able to successfully pass law & medical exams scoring up to in 10 highest percentile. This is non disputable achievement. Doesn’t this just basically mean that these exams are too “dumb” for being measured as “gold-standard” tests? It feels like they lack pinch of creativity where human being is still able to dominate over AI?
Other, a bit controversial fact was GPT’s sudden emerge into markets with huge backing of Microsoft and number of companies like Duolingo utilizing the model for language education right from day one of announcement. It just feels like all of this was well planned co-piloted global marketing effort. And social media (=people/users) contributed to this without needs for big marketing budgets 😃
One takeaway for me from using ChatGPT is the fact that we are still so in web2-mode. Majority of the technical solutions to build business processes are still very much utilising old-traditional methods of centralization, thinking and monetization principles. Sure, perhaps this is how people nowadays are taught to think. But made me actually think and realize that this “ideal” web3-world with all the concepts of decentralization and personal ownership are still too early.
As of writing of this, storm has been raising over this week countries wanting to ban OpenAI’s GPT because of legislative data-violations. Additionally, there has been unexpected open petition from big names such as Elon Musk of Tesla and co-founder of Apple Steve Wozniak accompanied with scientist elite to halt development of AI-algorithms worldwide. Ironically, the people who first-hand started all this run - now would like to halt something that’s out of control and unstoppable? Maybe they just realized that there’s real competitive threat to their companies that has to be stopped or at least slowed down?
Thanks for reading this post. Consider sharing, commenting and subscribing to learn about wonders of future and support vonLifted!
What are you thoughts on recent leaps from Artificial Intelligence field? Are they providing benefits into our lives or rather acting as threats? Let’s discuss the topic!🙌
Sparks of Artificial General Intelligence: Early experiments with GPT-4. 2023. https://arxiv.org/abs/2303.12712
OpenAI. Training language models to follow instructions with human feedback. 2022. https://arxiv.org/pdf/2203.02155.pdf