I love LLMs, I hate hype (geohot.github.io)

In a reflective blog post, an AI enthusiast expresses enthusiasm for advancements in AI technologies like large language models (LLMs), self-driving cars, and coding agents, emphasizing the significant strides made in AI since the 2007-2014 period. They detail a personal experience setting up a Linux environment with the GLM-5.2 model, highlighting its convenience, but counter the prevalent negativity surrounding AI's trajectory. The author dismisses claims of a technological “window closing” or an impending AI dominance event, arguing that such narratives are not only false but also drive unnecessary fear and anxiety within the community. Significantly, the post critiques the commercial AI landscape, asserting that the progress in AI should be credited to ongoing advancements in computing, such as Moore’s Law, rather than the efforts of elite labs. The writer points out that while AI is undeniably valuable, the control of that value by a few organizations may be overstated. They acknowledge the evolving nature of programming, noting AI can enhance productivity, though they caution against the cognitive overload it may introduce. Ultimately, the blog champions AI as a vital continuation of the computer revolution, advocating for open-source solutions and a more positive perspective on the future of technology.