sdecoret -

Generative AI in business: Fast uptake, earmarked funding

More than half of IT and business decision-makers said they have generative AI on the near-term adoption track, according to a report from TechTarget's Enterprise Strategy Group.

Generative AI became a household word in late 2022 and has since seen mass adoption. Millions of consumers already use generative AI, and businesses are exploring how they can employ the technology for competitive advantage. This wave of artificial intelligence presents a paradox: On the one hand, it's intuitive enough for people to grasp with little or no training. On the other hand, its potential uses -- and abuses -- are such that organizations face considerable learning curves and deployment challenges.

To better understand the business implications of this high-profile technology, TechTarget’s Enterprise Strategy Group surveyed 670 technology and business decision-makers working with generative AI in their organizations. The global survey canvassed C-level executives, directors, managers and staff, among other professionals.

The resulting report, "Beyond the GenAI Hype: Real-world Investments, Use Cases and Concerns," quantifies generative AI's wide and early acceptance among businesses. In this interview, Mike Leone, principal analyst of analytics and AI at Enterprise Strategy Group, discussed the technology's adoption status, funding commitment and broad stakeholder ecosystem.

What do you see as the key takeaways from the research?

Mike Leone, principal analyst, analytics and AI, Enterprise Strategy GroupMike Leone

Mike Leone: We're seeing broad industry adoption: 54% of organizations will have generative AI adopted in the next 12 months. That might be the fastest enterprise technology adoption rate in my lifetime.

The alarming part about that? Because organizations are putting pressure on different business units to use and leverage this technology, they need to make tradeoffs in how they're approaching it. So, it's actually introducing a fair amount of risk in the organization, whether it's compliance, privacy, governance, security, etc. Organizations are starting to take firmer stances and establish guidance and frameworks when it comes to those items, but because of the pressure to innovate and accelerate GenAI initiatives, we're actually seeing them have to backpedal a little bit.

So, while the rate of enterprise adoption shows this isn't hype, it's still introducing a fair amount of risk.

Chart showing generative AI adoption status
The majority of organizations will have adopted generative AI within 12 months.

'Shadow AI' in a broad stakeholder ecosystem

Another component is really around the stakeholder ecosystem. Traditionally, leveraging AI will come from somebody on the technical side. Maybe it's somebody within the data science team, or maybe even someone in IT that's leading the charge.

What we're seeing now with generative AI is that line-of-business leaders, or just business folks in general, are being empowered to leverage this GenAI technology. Vendors are making it so easy to adopt this technology using pre-trained models with a prompt front-end like ChatGPT, or managed services that enable self-service and take the guesswork out of the underlying infrastructure and/or data management components.

Seventy-eight percent of people like you and I are leveraging generative AI for personal use. That's a massive number. That's a huge risk from a shadow AI standpoint.
Mike LeonePrincipal analyst, analytics and AI, Enterprise Strategy Group

When it comes to decision-making on what to do and when to do it, it's not just data teams, it's not just IT. It's business leaders. It's developers. It's virtually everyone within an organization that has a computer. It's folks in sales and folks in marketing.

That broad stakeholder ecosystem is really interesting, because a lot of those business folks are not involving the folks on the technical sides. Seventy-eight percent of people like you and I are leveraging generative AI for personal use. That's a massive number. That's a huge risk from a shadow AI standpoint. The flip side: Only 36% of organizations have implemented any policies that prevent or limit the use of AI. The scale is tipped in significant favor of personal use and almost forgetting about compliance, privacy and security in some cases. That really needs to balance out.

Graphic depicting high level of generative AI for personal use
Nearly 80% of respondents employ generative AI for personal use.

Leading use cases for generative AI

The other aspect to touch on is the broad set of use cases that spread across all different areas within enterprise technology. I think the use cases that are really emerging today [are] focused on the CX side and transforming the customer experience -- so, virtual assistants, call centers, contact centers.

The other [area] is around just productivity boosting. So that's summarization, content and code generation, and then enterprise search is a really great opportunity, too.

What role might vendors play in generative AI adoption?

Leone: Overcoming the skills gap is going to be very difficult for adopters. And because of that, there is going to be an enormous opportunity in the partner and professional services ecosystem. We're already starting to see that with so many vendors positioning solutions, accompanied by some type of professional services. Vendors and partners are positioning themselves to answer questions like, 'What are your challenges? What are your objectives? What initiatives are underway? What is available in-house?' They'll be able to quickly identify use cases that can impact your business and then provide technical guidance in order to ensure success, whether that's providing access to an open source model or a prebuilt solution.

Vendors are really trying to position the concept of guiding customers from ideation to completion at scale. This really highlights the fact that the partner ecosystem is critical. There are definitely some vendors that are out in front. But you need to partner, and it's likely going to be with more than one vendor, consultancy or third party.

Let's go back to the high percentage of personal use for generative AI. Is there any upside to that? With any emerging technology, user acceptance is usually among the hardest parts.

Leone: One of the biggest challenges when it comes to AI is getting full company buy-in. I think, with the accessibility of GenAI tools, folks are actually willing to experiment; they’re seeing the power of this technology.

The fact that you or I can be empowered to go and mess around with a generative AI tool and really think outside of the box -- that's huge, because in the past, there was so much red tape to be able to do that. Or there were high costs. Or someone else was in control of enabling it, like IT. I think from a consumer standpoint, because it's readily available, that red tape has been taken away to a point where people can see that power and say, 'OK, I've played with this. I know it can add value, not just to me, but to my team or my colleagues. We need to start pursuing this in a more formal way.'

The report lists areas like cybersecurity resiliency, digital transformation, cutting costs, automation and application modernization as strategic corporate initiatives. Is generative AI going to reshape those initiatives? Does digital transformation become AI transformation? Or is it more the case that AI is an enabling technology within digital transformation?

Leone: GenAI will reshape a lot of areas as organizations put in guardrails and people gain more trust in it. I think GenAI capabilities will become table stakes throughout all of those different areas in a way that organizations and people will be able to do much more -- become more productive, become more operationally efficient, accelerate delivery of products, innovate faster, experiment more, etc.

Application modernization is an example, with developers turning to generative AI to help with code generation. It's not taking the job of a developer but enabling a level of productivity boosting for developers to experiment with something now, like providing code in a different programming language to solve the same problem. Or even if it's someone who's not an expert developer, there are going to be levels of improvement for folks that are looking into application development modernization.

On the cyber-resiliency side, I think generative AI will enable organizations to really improve their level of response. And it's not so much about monitoring and alerting, but more [about] doing things with synthetic data. Imagine being able to take all of a company's network traffic and then generate a data set based on that data, with the addition of a new anomaly that [an] organization hasn't seen before and test how [its] systems would respond.

Regarding budget allocation for generative AI, did you find anything surprising about the level of commitment that came through in the report?

Leone: So, we asked two questions. The first was simply about the overarching organizational budget that has been allocated to GenAI. Forty-eight percent of organizations said, 'Yes, we have such a budget.' That budget could be coming from anywhere, but knowing a lot of technology investments come from IT's budget, we asked a follow-up question about the percent of IT's budget being allocated to GenAI.

Interestingly, just based on conversations I'm having with early adopters, a lot of budgets are coming from outside of IT. So, we're seeing a low percentage of IT's budget go to GenAI. That means there are other budgets coming from business [functions] that are going to GenAI and, in many cases, they're not even looping IT into the discussion until far too late. These are business-led initiatives.

Chart showing organizations' generative AI investment plans
Budget allocations underscore business commitment to generative AI adoption.

What does that say about IT? Will IT concede more ground to business and focus more on things like security and governance?

Leone: This may sound extreme, but I think there will be a reckoning to some extent when it comes to technical stakeholder involvement. IT needs to understand the infrastructure requirements to satisfy these GenAI initiatives. How much storage? How much compute? What are the expected levels of concurrency? How do those resources need to scale? And that's just the start. What about the data teams? Is there sensitive data? Was it used to train a model? Did someone put in sensitive data inadvertently?

I also think there's going to be a lot of opportunity for IT to extend their expertise into some new emerging areas, based on the direction that GenAI is going to take some organizations. I think there's going to be more governance opportunities. I think there's going to be more security opportunities. If I'm an IT person, I'm embracing that -- the ability to extend knowledge outside of what I'm used to doing from an IT operations standpoint.

Read the full survey results

Enterprise Strategy Group subscribers can click here for a report on the survey covered in this Q&A.

Editor's note: This interview has been edited for clarity and length.

Next Steps

10 realistic business use cases for ChatGPT

Dig Deeper on AI business strategies

Business Analytics
Data Management