Chat buddy app: Tarpit idea?

Thoughts on an chat app being a tarpit idea? Basically my stance is GPT API wrappers are all tarpit ideas.

Definition of tarpit idea:

A tarpit business is a term used to describe a business model or idea that appears attractive or promising at first glance but ultimately proves to be difficult, unprofitable, or even detrimental to the company in the long run. The concept is derived from the idea of a tar pit, where animals get stuck and struggle to escape.

Characteristics of a tarpit business idea include:

  1. High initial investment: The business requires significant upfront capital, making it difficult for the company to break even or turn a profit.

  2. Low barriers to entry: The market is easily accessible to competitors, leading to increased competition and reduced market share.

  3. Lack of differentiation: The product or service offered is not unique or lacks a clear competitive advantage, making it challenging to stand out in the market.

  4. Difficulty in scaling: The business model is not easily scalable, limiting growth potential and the ability to expand into new markets.

  5. Regulatory challenges: The industry is heavily regulated, creating compliance issues and potential legal obstacles.

  6. Dependence on external factors: The success of the business relies heavily on factors outside the company’s control, such as economic conditions, technological advancements, or shifting consumer preferences.

  7. Short-term focus: The business prioritizes short-term gains over long-term sustainability, leading to decisions that may harm the company’s future prospects.

What are your guys stances? Is anyone really making money off of an API wrapper? What business ideas that revolve around the gpt api are really taking off?

It’s a huge risk to release an LLM wrapper while:


  • continues to build their walled garden and desperately seek a moat
  • is okay with burning A LOT of money to give access to everyone
  • does NOT release available ChatGPT features for the API in a timely manner

In all manners any developer is at a huge disadvantage while GPT is the leading LLM. You won’t be able to match their pricing, you won’t be able to match their features, and you will be playing constant catch-up as they do not release any technical information of upcoming updates.

Anything that doesn’t use GPT as a driver, but as an assistant to augment current services.

This doesn’t mean “Chatbot to do dumb things like write SQL queries or improve my resume”. But something that is added to the pipeline to help transform content. ESPECIALLY when it comes to analyzing/working with unstructured semantics.

I think from our time being here (nice seeing you again) that we see OpenAI constantly bulldozing upcoming business ideas and applications using their technology.

An API “wrapper” is only useful for those clients that OpenAI doesn’t want to do business with, or who find the fraud resistance of OpenAI too high.

An API product also leaves you vulnerable to those who compete directly against you with free service enticements and for what you must pay for. See Forefront chat - used to have monthly subscriptions and AI personalities: shut down.

Microsoft putting GPT-4 in every OS, as a reward for turning your OS into a payment system vending machine.