If you believe the hype, generative AI has the potential to transform how we work and play with digital technologies.
Today’s eye-popping text-and-image generating classes of AI capture most of the limelight, but this newfangled automation is also coming to software development.
It is too soon to say what impact this emerging class of code-generating AI will have on the digital world. Descriptors ranging from “significant” to “profound” are regularly tossed around.
What we do know: IT must take a more proactive role in supporting software developers as they begin to experiment with these emergent technologies.
Generative AI Could Change the Game
Many generative AI coding tools have come to the fore, but perhaps none possesses more pedigree than Copilot, developed by Microsoft’s GitHub coding project management hub.
A type of virtual assistant, Copilot uses machine learning to recommend the next line of code a programmer might write. Just as OpenAI’s ChatGPT gathers reams of text from large corpuses of Web content, Copilot takes its bits and bytes insights from the large body of software projects stored on GitHub.
Although it’s early days for such tools, developers are excited about Copilot’s potential for enhanced workflows, productivity gained and time saved. Empirical and anecdotal evidence suggests it can shave anywhere between 10% and 55% of time coding, depending on who you listen to.
Today Copilot is targeted at professional programmers who have mastered GitHub and committed countless hours to creating and poring over code. Yet it’s quite possible that Copilot and other tools like it will follow the money and migrate downstream to accommodate so-called citizen developers.
DIY AI, for Non-Coders
Typically sitting in a business function such as sales or marketing, citizen-developers (cit-devs) are non-professional programmers who use low-code or no-code software to build field service, market and analytics apps through drag-and-drop interfaces rather than via the rigors of traditional hand-coding.
If the low-code/no-code evolution has come to your company, you may have marveled at how this capability freed your staff to focus on other tasks, even as you helped these erstwhile developers color within the governance lines.
Considering their all-around efficacy, self-service, do-it-yourself tools are in-demand: The market for low-code and no-code platforms is poised to top $27 billion market in 2023, according to Gartner.
Now imagine what organizations will pony up for similar tools that harness AI to strap rocket boosters onto software development for non-techie coders. In the interest of catering to these staff, GitHub, OpenAI and others will likely create versions of their coding assistants that streamline development for cit-devs. GitHub, for example, is adding voice and chat interfaces to simplify its UX even more.
It’s not hard to imagine where it goes from there. Just as the API economy fostered new ecosystems of software interoperability, generative AI plugins will facilitate more intelligent information services for big brands. Already OpenAI plugins are connecting ChatGPT to third-party applications, enabling the conversational AI to interact with APIs defined by developers.
One imagines this AI-styled plug-and-play will broaden the potential for developers, both of the casual and professional persuasion. Workers will copilot coding tasks alongside generative AI, ideally enhancing their workflows. This emerging class of content creation tools will foster exciting use cases and innovation while affording your developers teams with more options for how they execute their work. This will also mean development will continue to become more decentralized outside the realm of IT.
Keep an Open Mind for the Future
The coming convergence of generative AI and software development will have broad implications and pose new challenges for your IT organization.
As an IT leader, you will have to strike the balance between your human coders—be they professionals or cit-devs—and their digital coworkers to ensure optimal productivity. You must provide your staff guidance and guardrails that are typical of organizations adopting new and experimental AI.
Use good judgment. Don’t enter proprietary or otherwise corporate information and assets into these tools.
Make sure the output aligns with the input, which will require understanding of what you hope to achieve. This step, aimed at pro programmers with knowledge of garbage in/garbage out practices, will help catch some of the pitfalls associated with new technologies.
When in doubt give IT a shout.
Or however you choose to lay down the law on responsible AI use. Regardless of your stance, the rise of generative AI underscores how software is poised for its biggest evolution since the digital Wild West known as Web 2.0.
No one knows what the generative AI landscape will look like a few months from now let alone how it will impact businesses worldwide.
Is your IT house in order? Are you prepared to shepherd your organization through this exciting but uncertain future?
Learn more about our Dell Technologies APEX portfolio of cloud experiences, which affords developers more options for how and where to run workloads while meeting corporate safeguards: Dell Technologies APEX.
