AI Coding Tools Stifle Tech Adoption

Declan Chidlow expresses concern that AI is stifling tech adoption.

He writes:

I think it is evident that AI models are influencing technology, and that the technologies currently in use – especially those that reached popularity before November 2022, when ChatGPT was released, or that are otherwise in current data sets – will be around for a long time to come, and that AI models’ preferential treatment of them will expand their adoption and lifespan.

Why is this the case?

To begin, large language models do not have access to training data on technology developed after a given point.

By the time everything has been scraped and a dataset has been built, the set is on some level already obsolete. Then, before a model can reach the hands of consumers, time must be taken to train and evaluate it, and then even more to finally deploy it. Once it has finally released, it usually remains stagnant in terms of having its knowledge updated. This creates an AI knowledge gap.

This concept has already been well-established, where models perform poorly in solving problems involving newer frameworks and versions.

Perhaps knowing this, companies providing LLM-powered coding tools write system prompts to influence different LLMs to prefer certain tech stacks.

Thus, the use of LLM-assisted code tools however creates a negative feedback cycle for innovation:

if people are reluctant to adopt a new technology because of a lack of AI support, there will be fewer people likely to produce material regarding said technology, which leads to an overall inverse feedback effect. Lack of AI support prevents a technology from gaining the required critical adoption mass, which in turn prevents a technology from entering use and having material made for it, which in turn starves the model of training data, which in turn disincentivises selecting that technology, and so on and so forth.

— via Baldar Bjarnason