• Can chatbots craft correct code? • I recently attended the AI Engineer Code Summit in New York, an invite-only gathering of AI leaders and engineers. • One theme emerged repeatedly in conversations with attendees building with AI: the belief that we’re approaching a future where developers will never need to look at code again. • When I pressed these proponents, several made a similar argument: Forty years ago, when high-level programming languages like C became increasingly popular, some of the old guard resisted because C gave you less control than assembly. • The same thing is happening now with LLMs. • On its face, this analogy seems reasonable.
Article Summaries:
- At the AI Engineer Code Summit in New York, some attendees argued that large‑language models (LLMs) will soon eliminate the need for developers to read code. The author counters this view by highlighting a key difference between traditional compilers and LLMs: determinism. Compilers translate code while preserving the programmer’s intended semantics, ensuring predictable behavior-an essential property for security and correctness. In contrast, LLMs are nondeterministic by design, lacking guarantees that generated code will behave as intended. The piece notes recent compiler extensions (e.g., Trail of Bits’ LLVM work) that can enforce properties like constant‑time execution, underscoring the importance of deterministic tools in software development.
Sources: