Low/ No Code Development and Generative AI
Authors
Sido, Nourjan ; Emon, Eksan Ahmed
Term
4. term
Publication year
2024
Submitted on
2024-05-31
Abstract
This thesis examines how low/no-code development and generative AI together may shape the future of software development. Guided by the research question “How will the offering of low/no-code solutions, enhanced by generative AI, affect the future of software development?”, it conducts a structured literature and conceptual analysis. The work reviews the background of traditional and modern development methodologies, the state of the art in low/no-code and generative AI, and the distinctions between low and no code. It includes a survey of existing platforms (e.g., OutSystems, Joget, and Mendix), a technical analysis of platform architecture, the low-code development lifecycle, and stages where generative AI can contribute (such as code generation, testing, and documentation), as well as a business analysis covering stakeholders, market trends, SWOT, Porter’s Five Forces, and Value Proposition/Business Model Canvases. The study indicates substantial potential to democratize development, empower citizen developers, speed prototyping, and reduce costs via automation and faster time-to-market. It also identifies key challenges, including the quality and security of AI-generated code, ethical concerns, disruptions to established processes, and current limitations of low/no-code platforms. The research is primarily theoretical without implementation and has limited primary data due to low regional adoption and declined industry collaboration. Overall, the thesis outlines opportunities and risks in integrating generative AI with low/no-code and highlights the need for further empirical work and practical validation.
Dette speciale undersøger, hvordan low/no-code-udvikling og generativ AI tilsammen kan påvirke fremtidens softwareudvikling. Med udgangspunkt i problemformuleringen “Hvordan vil udbuddet af low/no-code-løsninger, forbedret med generativ AI, påvirke fremtidens softwareudvikling?” gennemføres en struktureret litteratur- og begrebsanalyse. Specialet dækker baggrunden for traditionelle og moderne udviklingsmetoder, state-of-the-art for low/no-code og generativ AI, og skelnen mellem low og no code. Det omfatter en platformsgennemgang (bl.a. OutSystems, Joget og Mendix), en teknisk analyse af platformarkitektur, udviklingslivscyklus og de faser, hvor generativ AI kan bidrage (fx kodegenerering, test og dokumentation), samt en forretningsanalyse med interessenter, markedsudvikling, SWOT, Porters fem kræfter og Value Proposition/Business Model Canvas. Undersøgelsen peger på et betydeligt potentiale for at demokratisere udvikling, styrke “citizen developers”, accelerere prototyper og reducere omkostninger gennem automatisering og hurtigere time-to-market. Samtidig identificeres centrale udfordringer, herunder kvalitet og sikkerhed af AI-genereret kode, etiske problemstillinger og mulige forstyrrelser af etablerede processer samt nuværende begrænsninger i low/no-code-platforme. Arbejdet er primært teoretisk uden implementering og med begrænset adgang til primærdata, bl.a. på grund af lav regional udbredelse og afslag på virksomheds-samarbejde. Specialet afgrænser muligheder og risici ved integrationen af generativ AI i low/no-code og peger på behov for yderligere empiriske studier og praktiske afprøvninger.
[This apstract has been generated with the help of AI directly from the project full text]