I just read about how CIOs are taking it slow regarding generative AI in the enterprise. “The hard thing about this is you don’t have data on people’s level of productivity,” said Sharon Mandell, CIO at Juniper Networks, while  Kathy Kay, executive VP and CIO at Principal Financial Group, created a study group of “a combination of engineers and business people and are curating probably 25 use cases.” The article quotes a Morgan Stanley study finding that only 4% of large company CIOs launched significant generative AI projects.

CIOs should be cautious about where they are making big investment bets, especially when it concerns the enterprise’s intellectual property or sensitive data sets. In my recent post on how CIOs deliver short-term wins and plan visionary impacts with gen AI, I advised finding short-term wins in IT, increasing the frequency of blue-sky planning, and investing opportunities using large language models (LLMs) on proprietary data.

CIOs must consider strategic opportunities and risks – but Digital Trailblazers – the lieutenants leading digital transformation initiatives are the leaders who should identify value opportunities, experiment, learn, and project where generative AI can provide quick and longer-term wins. For example, in my post on 10+ awesome LLM and gen AI capabilities for DevOps and IT Ops, reducing incident resolution times, generating IT documentation, and increasing dev velocity with copilots are all short-term productivity opportunities.

The Digital Trailblazer’s role in AI: Find value and experiment

Where else can Digital Trailblazers look for innovation opportunities with generative AI?

One source of inspiration should come from SaaS and other technology company CTOs, who are more likely to experiment with emerging tech. Many SaaS CTOs embed gen AI capabilities in their products, such as Quickbase AI (Smart Builder, Data Analyzer, and Data Scanner), Coveo’s Relative Generative Answering, or  LaunchDarkly’s Product Experimentation. CTOs can provide insights into what’s working, what’s hype, what tools to try, and how to avoid the gotchas.

I’ve shared some of what I learned from SaaS CTOs in older posts, including ten things enterprise CIOs can learn from startups about innovation and disruption, stemming from a panel I moderated at the former CIO Global Forum, now the Spark Executive Forum, that included Dries Buytaert (Acquia / Drupal) and Jeremy Howard (Kaggle, now fast.ai and Answer.AI). More recently, I wrote about what CIOs can Learn from low-code platform CTOs.

So, let’s look at some examples of SaaS CTOs and engineering leaders leveraging generative AI capabilities.

Reducing bias when recruiting for diversity

Last month, I moderated a panel at SINC USA’s IT and Security Leaders Forum on diving into talent acquisition, diversity, and retention. Digital Trailblazers reported some improvements and their struggles in getting a diverse candidate pool when hiring. For example, leaders reported success in hiring women application developers, but the candidate pool for infrastructure and security jobs was far less diverse. One leader received 260 applications for a security role – all male. Other reported issues in recruiting African Americans, Hispanics, Latinos, and people from diverse backgrounds into IT and security roles.

Bias in job descriptions is often cited as a recruiting issue. Vlad Iacob, VP of engineering at hackajob, says, “While AI in hiring is often criticized for relying on biased data and skewing results to favor certain candidates, the use of gen AI reverses this by analyzing existing data sets to remove bias from the start of the hiring process, enabling hiring managers to remove bias from job descriptions, ensuring that all candidates are equally accessing the information.”

We all have unconscious biases that can impact how we write job descriptions and review job applications. Generative AI technologies are options when used to augment the recruiting and hiring process.

Vlad suggests leaders “should not let gen AI have full autonomy over any aspect of hiring. Instead, it should complement the hiring lifecycle and be used to improve what is already in place.”

Find and reduce toil in dev, ops, and data science

Developers like to solve problems, IT operations want to address performance issues proactively, and data scientists want to find insights in new data sets. Engineers take on many technical tasks to achieve these goals, some value-adding, others necessary toil to get the job done.

Copilots and other generative AI IT capabilities aim to improve productivity, simplify skill sets, and expose existing solutions for reuse.

“I’m excited by the new generation of AI learning algorithms because they allow us to train models faster, optimize code, and generate boilerplate code automatically,” says Darko Fabijan, co-founder of Semaphore CI/CD. “This frees up my team to focus on creative problem-solving and delivering value for customers.”

Digital Trailblazers should ask questions – where is gen AI capability providing tangible value, and what’s the impact? For example, when gen AI productivity improvements help teams address customer needs faster and better, then that’s a positive impact.

Step out of your comfort zone and ask the CTOs

Here are some other examples of innovative use cases:

• “New services offer extremely valuable benefits for rapid prototyping, testing, and market validation in a much shorter timeframe,” says Erik Reeves, CTO of Anaqua, in what CTOs are learning from generative AI.
• “The nature of how we build software itself is being completely transformed dramatically by gen AI. It involves how gen AI can augment what the developer is doing by helping with code snippets, or writing test cases, or even creating automatic pull requests,” says Pandurang Kamat, CTO of Persistent Systems, in his interview on generative AI in the enterprise.
• “We chose use cases because we felt at the time it was right in the wheelhouse of what generative AI is known to cover pretty well. When it comes to text summarization and having the human still involved to verify, we felt really good about that,” says Tim Armandpour, CTO of PagerDuty, in an interview on minimizing collateral damage.

In my book, Digital Trailblazer, I advise readers to “step out of your comfort zone and broaden your perspective by seeking outside‐in learning opportunities.” You can learn a lot from research and want to experiment with gen AI tools. In between these steps are opportunities to reach out to experts, such as SaaS and tech company CTOs, and learn from their experiences.