Accelerating the search for antimicrobial peptides with Google Antigravity
deepmind.google·2h ago
AI Engineering
Hacker News·2d ago·5 min read
AI Is a Tool. Using It Well Takes Skill.
TL;DR
Rich Hickey argues that large language models (LLMs) are automation rather than tools, but effective use of LLMs requires skill and intention, similar to traditional tools.
✦ Why It Matters
Develop a structured approach to interact with LLMs to maximize their effectiveness in your projects.
Key Takeaways
- Recognize LLMs as tools that require skillful use.
- Develop a feedback loop when working with LLMs.
- Evaluate LLM outputs critically to ensure accuracy.
- Understand that automation can enhance tool functionality.
Full Summary
Rich Hickey's argument that large language models (LLMs) are merely automation and not tools overlooks the skill required to use them effectively. While he suggests that LLMs lack the qualities of traditional tools, such as predictability and reliability, this does not disqualify them from being considered tools.
Just as a saw or chisel requires skill to produce desired results, LLMs also demand operator expertise to navigate their unpredictability. The feedback loop between a programmer and an LLM—where the programmer refines requests and evaluates responses—demonstrates intentional use.
Furthermore, automation can enhance tools, as seen in compilers and spreadsheets, which perform tasks without constant human input while still being directed by user specifications. Thus, LLMs should be recognized as tools that can amplify both good and poor judgment, depending on the user's skill level.