How generative AI will change low-code development

It’s been over two decades since I developed my first low-code application. Since then, I’ve seen platform capabilities evolve to make it easier for both software developers and citizen developers to build and enhance applications. Low-code and no-code can help developers build apps faster, enable business users to convert spreadsheets to workflows, and help IT departments accelerate application modernization. Beyond applications, these platforms can accelerate the development of integrations, dashboards, IoT data streams, and other capabilities.

Evolutions in technology often drive changes in application development and modernization. For example, the release of smartphones and app stores required pivoting to mobile-first development strategies, while cloud infrastructure enabled many non-technology businesses to improve applications and develop analytics capabilities. Now, we are in the early stages of seeing the same pattern with generative AI. The question is, how will genAI impact the adoption and use of low-code platforms?

How genAI impacts software development

I recently wrote about 10 ways generative AI will transform software development. One of my points was that today’s code generators may evolve the software development lifecycle (SDLC) into a manufacturing process where developers prompt for application components and assemble them into applications and services. That may sound futuristic, but code generators are already making significant impact. GitHub found that 88% of developers reported improved productivity, 74% focused on more satisfying work, and over 87%  said they completed tasks faster using GitHub Copilot.

Currently, low-code and no-code platforms are used to simplify development, expand the number of people who can develop applications, and evolve the skills required to customize user experiences. So, how will genAI impact these platforms?

“In the future, everyone will be generating software, but they just won’t realize that’s what they’re doing,” says Jon Kennedy, senior VP of engineering at Quickbase. “For example, if you know how to ask the right questions of a copilot, you can have it quickly build an app or deploy a solution.”

While natural language querying and prompting enables software developers to generate code and improve productivity, low-code and no-code platforms are adding their own copilot development capabilities.

“Coding will become almost entirely automated, and UX designers will become the de facto front-end developer,” says David Brooks, senior VP and lead evangelist at Copado. “Instead of graphics tools like Figma to mock up UI, they will work with genAI tools to generate working UI prototypes in the company’s framework of choice.”

Will code generators replace low-code platforms?

GitHub’s research shows that users accept 30% of the code its Copilot suggests and that less experienced developers have a greater advantage with AI. This leads some to believe that genAI may spell the end for low-code platforms.

“Low code is dying in the enterprise, and AI will kill it,” says Anand Kulkarni, CEO and founder of Crowdbotics. “The big question is, why would you want to use low-code when you can use AI to create full code with the same effort?”

Michael Beckley, co-founder and CTO of Appian, sees things differently. “No, code generators are part of the problem. AI copilots make it easy to create lots of apps which only increases the need for a low-code platform to connect and govern them all to ensure you aren’t creating data silos and security issues.”

Beckley takes a wider view of how genAI will expand the need for low-code and its use cases. “Low-code makes it easy to deploy AI assistants, but AI is only as good as its data. Low-code platforms are evolving to include data fabrics to create private AIs that can access all your data and keep your secrets.”

Another response comes from Manish Rai, VP of product marketing at SnapLogic. “AI and machine learning have paved the way for new, innovative ways to make business process automation and data and application integration easier to implement, more accessible to non-technical users, and more efficient.”

Ultimately, organizations need greater AI innovations, more personalized experiences, shorter development cycles, and greater business value delivered from software investments. Increased expectations and scope will likely drive technology leaders to build software capabilities with both code and low-code options. 

Sid Misra, SAP vice president of product marketing, emphasizes the potential of combining low/no-code development with AI and mobile technology for groundbreaking applications. “Low/no-code development, when integrated with AI, enables rapid prototyping and sophisticated solution development, transcending traditional limitations. In healthcare, for instance, developers leverage these tools to quickly build apps that significantly enhance Parkinson’s disease diagnosis, utilizing AI to detect patterns for more accurate, swift diagnoses.”

How will genAI drive developer skillsets?

GenAI can generate code, test cases, documentation, and other artifacts needed to develop software. How will that impact the skills to build software capabilities with low-code and no-code platforms?

Dinesh Varadharajan, chief product officer of Kissflow, says, “Coding will shift from traditional syntax to contextual awareness and intelligent constructs, empowering business users to create applications with little programming skills.”

If developers are coding less, what other skills become more important?

“Skill sets will evolve to encompass a blend of traditional coding expertise, along with proficiency in utilizing low/no-code platforms, understanding how to integrate AI technologies, and effectively collaborating in teams using these tools,” says Ed Macosky, chief product and technology officer at Boomi. “The combination of low code alongside copilots will allow developers to enhance their skills and focus on supporting business outcomes, rather than spending the bulk of their time learning different coding languages.”

Armon Petrossian, CEO and co-founder of Coalesce, adds, “There will be a greater emphasis on analytical thinking, problem-solving, and design thinking with less of a burden on the technical barrier of solving these types of issues.”

Today, code generators can produce code suggestions, single lines of code, and small modules. Developers must still evaluate the code generated to adjust interfaces, understand boundary conditions, and evaluate security risks. But what might software development look like as prompting, code generation, and AI assistants in low-code improve?

“As programming interfaces become conversational, there’s a convergence between low-code platforms and copilot-type tools,” says Srikumar Ramanathan, chief solutions officer at Mphasis. “The evolving skill set sees developers embracing AI principles while citizen developers focus on business logic, aiming to enhance quality through collaborative AI-driven efficiency and customized solutions.”

Will software quality improve or get worse?

As more people with different skill sets leverage AI assistants to build and enhance software, should we expect software quality and end-user experiences to improve or worsen? A related question is whether we will see defects released to production, mounting technical debt, and greater security vulnerabilities as AI enables more people to release more code.

“We’re already seeing lots of apps built by non-developers proliferate throughout organizations, so we know it’s a simple process,” says Kennedy of Quickbase. “This is exciting but comes with some caution—as these apps and copilots become common, organizations must ensure that the ease of building ‘an app for that’ doesn’t lead to the sprawl that can undermine productivity or introduce security risks.”

One answer may come from low-code platforms that extend testing, governance, and other guardrails to their AI assistance capabilities. 

“Developers are using generative AI alongside tools such as low-code to create applications at unprecedented speeds and do more with the same resources,” says Sílvia Rocha, VP of engineering at OutSystems. “These technologies’ built-in guardrails foster experimentation while eliminating the privacy and security risks associated with public AI models.”

AI assistants will likely help development teams shift left by bridging the gaps between writing requirements and generating development artifacts. “GenAI also has the opportunity to perform most of the tasks directly from a well-written user story. Instead of using the custom object/field tools, a co-pilot can create the metadata required and insert it directly into the platform,” says Brooks of Copado.

But back to today’s reality, where AI-generated code doesn’t mean defectless, security-clear, cost-free, or humanless code. “There is a strong need for a qualified human to verify the output of genAI, whether that’s writing lines of code or generating no-code workflows,” says Ben Dechrai, developer advocate at Sonar.

Will organizations build more applications with genAI?

As manufacturing assembly lines, electronic device design, and construction projects became streamlined, opportunities for growth and expansion in these industries opened up. The same is likely true for software development, and genAI is the next evolution.

“In recent years, we have seen how the traditional SDLC is being outshined by the low-code application platform,” says Varun Goswami, VP of product management at Newgen Software. “This shift has significantly streamlined lifecycles, enabling enterprises to expedite their go-to-market strategies. Today, with the advent of generative AI in application development, the lifecycle has not just evolved; it has taken flight.”

Many businesses will benefit if this prediction proves true, though I believe low-code and no-code platforms will be of greater value and importance in building, testing, and extending software developed with AI assistants.