What Successful Early Adopters Of Generative AI Have Learned

Innovation Enterprise Tech What Successful Early Adopters Of Generative AI Have Learned Joe McKendrick Senior Contributor Opinions expressed by Forbes Contributors are their own. I track how technology innovations move markets and careers Following Click to save this article. You’ll be asked to sign into your Forbes account.

Nov 10, 2023, 01:18pm EST Share to Facebook Share to Twitter Share to Linkedin Generative AI starts to deliver getty There has been no shortage of gushing over the past year about generative AI, but is it delivering business results? Maybe it’s time to set our perspectives straight. As with any promising technology, it can deliver amazing results, or turn out to be a colossal and expensive disappointment. We need to better understand what sets forward-thinkers in generative AI apart, and how others can emulate their success.

The folks at LTIMindtree did just that, looking at what has been working so far with generative AI. Their survey , conducted among 450 global executives in August 2023, finds widespread experimentation with generative AI tools. Early adopters seek to elevate customer experience, drive product design, and unlock data analytics.

It has led to increased revenue (80% of leaders report at least 5% growth) and cost savings (up to 40%), the survey shows. Still, generative AI is still in the hype phase, and “the real challenge lies in how businesses can adapt, scale, and ethically deploy these tools to realize tangible outcomes,” the survey report’s author, Pascal Bornet , writes. Early adopters are aware that AI expertise is still few and far between, requiring outside help.

At least 78% say they lack suitable skills, expertise, or knowledge to move forward with AI. A majority, 52%, sought outside expertise. MORE FOR YOU Are We Ready To Face Down The Risk Of AI Singularity? How Small Business Entrepreneurs Can Level Up With AI This Week In XR: Open AI Dev Day Bombshells, Aleph Alpha AI Raises $500 Million, xAI’s Grok It’s skills that deliver the most success with AI efforts.

Seven in ten executives say skills availability is the leading factor in delivering results from AI. Leadership and support is the second-ranked success factor, cited by six in ten. What Delivers Success in Generative AI? Access to skilled personnel 69% Strong leadership & management support 60% Effective training & education 59% User-friendly AI technology 53% Effective communication & change management 51% Companies leading the way with AI have a strong commitment to AI training, with 71% indicating they are taking measures to upskill their workforces.

Barriers to moving forward with AI include operational costs, challenges related to data quality and availability, and the need for precise identification of suitable use cases. A majority of AI leaders, 79%, cite ethical, security, or regulatory issues as barriers to the successful adoption or scaling of generative AI. Accordingly, they make responsible AI a top priority — 79% regularly audit their generative AI usage for ethical compliance and potential bias, focusing on reliability, safety, and ethical use of the technology.

Barriers Encountered High operational costs 85% Data quality or availability issues 75% AI strategy gap 72% Difficulty identifying suitable use cases 72% Technical infrastructure challenges 72% “The accuracy and effectiveness of generative AI solutions hinge on the quality, diversity, and relevance of the underlying data used to train AI models,” Bornet states. “Developing and fine-tuning narrow, industry-specific language models are crucial. These models, enriched with industry-specific knowledge, enhance content generation and task performance.

” As an example, a law domain-specific model needs “specialized training encompassing terminology, legal texts, cases, and specific terminology,” he illustrates. “Once fine-tuned, these industry-specific models excel in generating content, answering questions, or performing tasks tailored to their respective industries. ” Follow me on Twitter .

Joe McKendrick Editorial Standards Print Reprints & Permissions.