Why Go Developers Are Turning to AI Despite Python Dominance

A recent official survey of Go developers shows growing interest in building AI applications, yet many feel frustrated because Python dominates the field with a richer set of libraries.

About a quarter of respondents work in organizations exploring or building AI, and more than half cite the "Python‑centric ecosystem" as their biggest challenge.

Developers want to use Go for AI; one participant noted that Go's robustness, simplicity, performance, and native binaries make it a compelling choice for AI workloads.

The survey data were collected from feedback submitted in January and February 2024.

The most common AI services built are text summarization (56%), text generation (55%) and chatbots (46%), typically for internal enterprise use.

Python's dominance is seen as an obstacle: "Go lacks many AI libraries. If I have an LLM PyTorch model, I can't provide it—or I don't know how. In Python it’s just a few lines of code," said a Go developer.

Developers' top request is for Go to have libraries comparable to Python's, followed by tools to migrate Python code to Go and better interoperability with Python.

The Go team received 6,224 responses through the Go blog, the VS Code Go extension, and prompts in the JetBrains GoLand IDE.

Go is a compiled language with static typing, garbage collection, and memory safety, created by Google. It excels at concurrent programming and offers good performance, though it is generally slower than low‑level languages like C, C++ or Rust.

Performance matters: over 70% consider it at least moderately important, and nearly 40% view it as very or extremely important. However, diagnosing performance issues is challenging—30% find it difficult, while only 24% find it easy.

Fifteen percent of respondents say the biggest challenge is learning to write effective Go code, and 13% point to verbose error handling.

VS Code is the most popular IDE among Go developers (43%), followed by GoLand (33%) and Vim/Neovim (17%). Some developers are migrating to GoLand for its richer experience.

For cloud deployment, AWS is the preferred platform (52%), with GCP second (27%) and Azure third (13%). Satisfaction rates are 77% for AWS and GCP, but only 57% for Azure.