GPT-4 apps BabyAGI and AutoGPT may have disruptive implications for crypto

A recent wave of applications built on OpenAI’s GPT-4 API has the crypto community buzzing with designs toward the development of a fully autonomous, self-correcting cryptocurrency trading robot.

Two such apps, called ‘BabyAGI’ and ‘AutoGPT’, have gained particular attention with many users attempting to build crypto trading applications on top of them.

The big idea behind both apps involves task management for GPT-4. Currently, GPT-4 excels at natural language processing, as evidenced by the demonstrable utility of the ChatGPT interface, but it has no capacity for memorization.

Applications built on the GPT API are initially limited to use in one session, which means that the model cannot remember information from previous interactions. This has to do with the amount of data (referred to as the number of ‘tokens’) individual queries require, and GPT’s tendency to hallucinate – a problem that becomes increasingly noticeable as the number of tokens increases.

Users essentially start with a clean slate every time they ask for the machine. When it comes to building a crypto trading application capable of self-correction and historiographic analysis – adapting to market conditions in real-time while keeping short and long-term trends in focus – this means that even the most robust bot built on the GPT API will typically require heavy human supervision.

Related: Tech giant Alibaba to launch ChatGPT competitor AI

Some clever developers may have discovered a potential method to circumvent these limitations by building applications that take advantage of GPT’s ability to generate code and connect to external sources.

We’ve seen our fair share of trading bots, but the goal of these particular apps goes beyond just automating crypto news aggregation or teaching a machine learning agent how to recognize the downturn.

AutoGPT, for example, uses GPT-4 to generate code and then leverages GPT-3.5 as what appears to be a virtual artificial memory space where information is combined and passed between the two.

Another effort, BabyAGI, combines GPT-4 with LangChain, a coding framework, and Pinecone, a vector database, to create new agents to complete complex tasks without losing focus on the original goal.

Both apps could have the potential to act as the backbone of a multi-agent, fire-and-forget AI application capable of managing a top-to-bottom crypto portfolio based on little more than plain language.

While neither app appears to be specifically designed with the cryptocurrency market in mind, we’ve seen several attempts across social media and on GitHub to adapt one or both for autonomous trading.