Tag
This article discusses the current limitations of AI in research-level work, arguing that while AI excels at using existing packages and engineering solutions, it still struggles with the deep hypothesis-driven iteration required for genuine research. The author also warns against extreme views on AI's capabilities and uses AlphaFold as an example to illustrate that structuring the problem is the hardest part, not the optimization.
datasette-llm-limits 0.1a0 is a new plugin for Datasette that allows configuring per-user or global spending limits for LLM usage, working alongside datasette-llm and datasette-llm-accountant.