Table of Links
2. Prior conceptualisations of intelligent assistance for programmers
3. A brief overview of large language models for code generation
4. Commercial programming tools that use large language models
5. Reliability, safety, and security implications of code-generating AI models
6. Usability and design studies of AI-assisted programming
7. Experience reports and 7.1. Writing effective prompts is hard
7.2. The activity of programming shifts towards checking and unfamiliar debugging
7.3. These tools are useful for boilerplate and code reuse
8. The inadequacy of existing metaphors for AI-assisted programming
8.2. AI assistance as compilation
8.3. AI assistance as pair programming
8.4. A distinct way of programming
9. Issues with application to end-user programming
9.1. Issue 1: Intent specification, problem decomposition and computational thinking
9.2. Issue 2: Code correctness, quality and (over)confidence
9.3. Issue 3: Code comprehension and maintenance
9.4. Issue 4: Consequences of automation in end-user programming
9.5. Issue 5: No code, and the dilemma of the direct answer
10. Conclusion
Large language models have initiated a significant change in the scope and quality of program code that can be automatically generated, compared to previous approaches. Experience with commercially available tools built on these models suggests that a they represent a new way of programming. LLM assistance transforms almost every aspect of the experience of programming, including planning, authoring, reuse, modification, comprehension, and debugging.
In some aspects, LLM assistance resembles a highly intelligent and flexible compiler, or a partner in pair programming, or a seamless search-and-reuse feature. Yet in other aspects, LLM-assisted programming has a flavour all of its own, which presents new challenges and opportunities for human-centric programming research. Moreover, there are even greater challenges in helping non-expert end users benefit from such tools.
Authors:
(1) Advait Sarkar, Microsoft Research, University of Cambridge ([email protected]);
(2) Andrew D. Gordon, Microsoft Research, University of Edinburgh ([email protected]);
(3) Carina Negreanu, Microsoft Research ([email protected]);
(4) Christian Poelitz, Microsoft Research ([email protected]);
(5) Sruti Srinivasa Ragavan, Microsoft Research ([email protected]);
(6) Ben Zorn, Microsoft Research ([email protected]).
This paper is