CIO

The blistering pace of advancement in generative AI leaves companies struggling to effectively implement and measure the technology, while guarding against bias and risk.

Generative AI has created an unprecedented pace of technological change, but complications of choosing AI vendors have grown in equal measure. So tech leaders need to be equipped with the right questions — and be prepared for answers — to help enterprises prosper.

As companies race to add generative models into products and workflows, massive investment to upskill workforces on prompt engineering and ethical AI emerges as a top priority.

Building a new large language model (LLM) from scratch can cost a company millions — or even hundreds of millions. But there are several ways to deploy customized LLMs that are faster, easier, and, most importantly, cheaper.

Data issues are still among the chief reasons why AI projects fall short of expectations, but the advent of generative AI has added a few new twists.

In 2022, the big AI story was the technology emerging from research labs and proofs-of-concept, to it being deployed throughout enterprises to get business value. This year started out about the same, with slightly better ML algorithms and improved data pipelines. But that seems so long ago now. And in the few months left of 2023, change could further offset what we’ve come to expect.

Enterprises—and their technology vendors—continue to fall over themselves to add gen AI capabilities to tools and platforms. But moving too fast can expose companies to present and future compliance risks, as well as potential legal liabilities.

Beyond the household names for generative AI models, a burst of other options are emerging for the enterprise. Here’s a look at some that might be the right fit where others fall short.

Getting business value out of data requires data that’s accurate, comprehensive, and unbiased. Having a team composed of people with varying demographic, educational, and professional backgrounds can help companies reach that goal.

When asked, IT leaders, and the companies they represent, must be able to immediately harness the power of generative AI, and understand, as well as manage, potential vulnerabilities, biases, and other risks to protect sensitive data and maintain customer trust.

Companies that embrace data-driven approaches stand to perform much better than those that don’t, yet they’re still in the minority. What’s standing in the way?

To become data-driven, companies need a democratization strategy that’s equal parts disciplined and diverse. Data collection, selecting a platform and employee training are just the beginning.

Companies embarking on their automation journeys need to figure out the right process to automate, deal with organizational change, and watch out for security and compliance issues.

Taking stock of AI ROI is challenging but essential. IT leaders and industry observers lend insights on how to get a clear idea of whether your AI efforts are paying off.

The chances are that most of the data you collect — from human communications to machine logs — is piling up with little plan for actualizing its potential. Good governance and AI can help.