Combining Low-Code and GenAI for App Delivery
Artificial intelligence (AI) is setting new standards for software development, but despite the opportunities to drive business value, barriers to more widespread adoption remain. In this blog, we look at what is driving AI forward in organizations and what’s holding them back.
The AI revolution is changing business, and there’s no turning back
The business case for embedding AI in applications is becoming increasingly clear. When organizations integrate generative AI throughout their applications, a recent McKinsey report found, they see a 16% increase in revenue, 15% cost savings, and 23% higher productivity on average. These gains come from AI's ability to handle routine tasks and enhance user experiences across different domains.
The applications of AI in software span several key areas. In customer service operations, AI-enhanced applications are helping organizations provide consistent support across channels while reducing the workload on service teams. For research and development, AI tools are assisting with everything from literature review to data analysis, freeing up researchers to focus on insights rather than data collection.
Marketing teams are seeing particular benefits from AI integration, with over three-quarters of marketers now using AI for content creation. AI-enhanced tools help identify customer behavior patterns and inform decisions about targeting and segmentation. As a result, AI and generative AI are here to stay in business, but there’s more to it than that. There’s no avoiding it in IT, either.
How AI is transforming software development
In software development specifically, AI tools are reducing development time by more than half in some cases. This acceleration allows development teams to focus on higher-value work like system architecture and solving complex business problems. In fact, according to a recent survey by OutSystems and CIO Dive, 93% of software executives report that their organizations plan to increase their investments in AI for software development. Among the early adopters were the IT and consulting services sector, where AI is already well-established in use cases testing, quality assurance, and security vulnerability detection.
“What surprises me is that there isn’t yet more investment in AI for other parts of the software development lifecycle (SDLC),” says Rodrigo Coutinho, our AI product manager, and one of our co-founders. “There are many other areas that can benefit from these technologies, like coding assistance, analysis, design, deployment, and maintenance. The survey shows that IT leaders consider these areas important, so I would expect more experimentation and advances there.”
Why barriers to adoption of generative AI remain
Fears of job losses to AI abound, with 87% of executives reporting that at least some roles will be eliminated. However, what’s more likely in the longer term is that a new type of developer will emerge—one equipped with AI-specific skill sets—especially as executives focus less on saving time and reducing costs and more on risk management and business outcomes.
Many software executives use or plan to use generative AI in areas like DevOps optimization, code generation, document generation, and user interface design. Nonetheless, while they’re broadly confident in the quality of AI-generated code, concerns around data privacy, security, and AI explainability persist.
“Not all code is born equal, and there are different levels of confidentiality depending on what the code addresses,” says Coutinho. “I recommend starting by using AI to help with non-mission-critical projects. Then, you can analyze the gains in productivity to better understand how these models work. It’s about earning trust and, in doing so, iteratively opening up the usage of AI for other projects.”
A future filled with the potential of AI
According to the report, over a third of software executives anticipate that integrating AI in low-code workflows will enhance efficiency and scalability, reduce costs, and lead to faster development cycles. Low-code and AI are a natural fit for one another, driving democratization in software development.
Even more importantly, visual modeling and low-code solutions can help verify AI-generated outputs, enhance explainability, and mitigate the risk of hallucinations and AI-generated bugs.
“What excites me the most is the ability of these technologies to transform the relationship between developers and users,” says Coutinho. “By using AI to compress the time it takes to build an app prototype, the feedback process can start much sooner. This means developers will have a deeper understanding of what stakeholders really want, while stakeholders will have a clearer vision of where the project is heading.”
Ultimately, we’re looking at a future where AI will become ever more deeply ingrained in every stage of the SDLC, but one where AI and human decision-making work in tandem to fully realize the AI opportunity.
Get the AI report for complete access to the 2024 OutSystems and CIO Dive survey
About the survey
CIO Dive surveyed 555 software executives across several regions and industries to learn about their current adoption of AI, how they plan to use it, and the challenges they face.
