Trends
Latest trends and topline developments in generative AI
Increased adoption of private AI tools Two-thirds of business leaders have introduced private, secure generative AI tools in their organizations.
Focus on expert training72% of organizations are conducting employee training on generative AI concepts, but such training can be ad hoc and requires proper guidance.Productivity requires engagementA significant portion of leaders (78% for content generation, 54% for productivity and 65% for customer engagement) are using generative AI to enhance business functions. Security remains a priority38% of leaders identify security as their top challenge and have concerns about unauthorized AI interactions (such as prompt injections).
Data quality matters Although only 23% of business leaders currently see data readiness and quality as a barrier to generative AI, it still ranks among the most significant hurdles to generative AI adoption.
AI Center of Excellence (CoE) Only 47% of business leaders have launched an AI CoE, leaving a window of opportunity for organizations to gain industry recognition for demonstrating generative AI leadership.
A general mood of unfocused exuberance has shifted toward a paradigm of practical execution.
The focus in the generative AI business landscape is no longer on “unlocking new capabilities” in a loosely defined sense. Rather, we see a focus on the specialization of generative AI as more organizations discover custom use cases.
Insight’s National AI Practice Manager, Meagan Gentry, advises that “it’s important to get the right experts in the room to make sure everyone across the stakeholder chain understands what generative AI is capable of.”
Indeed, two-thirds of business leaders we’ve surveyed have introduced private, secure generative AI tools to their workforce.
As these tools become incorporated into day-to-day operations, training — both of the AI model and the employees using it — remains a key component of adoption.
Approximately 72% of companies are conducting employee training on generative AI concepts. However, the quality of that training is likely to vary in the absence of established standards and principles.
Increased significantly
Increased somewhat
Remained the same
Decreased significantly
Decreased somewhat
Not applicable/we do not use generative AI tools
It’s hard to fault anyone for subpar training. As Juan Orlandini, CTO of Insight North America and Distinguished Engineer says, “This stuff is being made up on the fly, and it’s changing so rapidly, if a user asks for a manual for how to use certain AI-enabled capabilities, it might not even exist.”
This challenge highlights the importance of having a dedicated team of experts who can clearly articulate the nuances of AI applications and guide users in using these tools effectively within their unique organizational contexts.
CTO of Product Innovation at Insight, Amol Ajgaonkar, believes “there is a fundamental shift in how we interact with computers, and training needs to be continually updated and modified to keep pace with this change.” Ajgaonkar advises that businesses must understand their user groups better and tailor training accordingly. This involves distinguishing between technical users who need in-depth knowledge and new users who require more basic, practical training.
The payoff will be in productivity. We expect that power users of generative AI within an organization will assume a more prominent role in shaping how employee training is conducted. This will also determine how an organization applies these insights to their strategic decision-making processes.
And those processes are already well underway. 78% of business leaders use generative AI for content generation for internal (e.g., email writing) and external (e.g., marketing collateral) purposes. Others use generative AI tools for data analysis and visualization, such as interpreting customer data patterns and optimizing logistics through predictive analysis.
We expect that power users of generative AI within an organization will assume a more prominent role in shaping how employee training is conducted. This will also determine how an organization applies these insights to strategic decision-making processes.
Security consistently ranks as a major hurdle for generative AI adoption, with 38% of leaders citing it as their top challenge. Given the increasing sophistication in AI-powered cyberattacks, it is highly probable that more leaders will prioritize security in proportion to their knowledge of the damage that is possible. “Security concerns in generative AI, especially prompt injections, are critical,” says Carm Taglienti, Insight Chief Data Officer and Distinguished Engineer.
“This risk involves unauthorized interception and alteration of AI prompts, leading to unexpected and potentially harmful outputs from the AI models.
Such vulnerabilities require serious attention to ensure the integrity and safety of AI interactions.” The industry is responding with more sophisticated defensive measures. Advanced encryption, real-time monitoring systems and AI-driven threat detection protocols are becoming standard practices.
However, only 15% of business leaders consider implementation costs as an obstacle to bringing generative AI into their organization. A shift in prioritization toward security assurance and the long-term benefits of secure AI integration may explain the emphasis on innovation and adoption over cost concerns.
Despite the growing popularity of productivity-oriented use cases, there is a clear sense of cautious optimism with only 25% who see it impacting bottom-line growth. However, the prevailing belief that the greatest outcome of generative AI is productivity suggests an emphasis on its cost-reducing potential rather than a direct mechanism to boost revenue. Taglienti explains that, for business leaders, “it’s about more than just deploying AI; it’s about understanding its business impact and setting the right expectations.”
“Assessing the business value of AI models critically hinges on an organization’s ability to quantify its ROI,” adds Meagan Gentry, Insight’s National AI Practice Manager. “At a minimum, that means knowing whether performance expectations are being met throughout the operational lifespan.”Such expectations, however, can fall short in practice. Common issues such as technical debt make it difficult to fully exploit the capabilities of generative AI. This speaks to the continued importance of modernizing data platforms to deploy AI capabilities effectively.
Improved productivity 54%
Improved customer experience (e.g., improved website or app layouts) 42%
Improved customer service (e.g., chatbot) 40%
Reduced human error 37%
Automation of manual processes 35%
Discovery and innovation 30%
Revenue growth for product or service lines 25%
Summary of complex information 24%
Mike Gaumond, SVP of Strategy at Insight, explains, “Improving the quality of data is crucial for the successful implementation of generative AI, and businesses need to prioritize this.” Those who succeed on this front can expect to have much shorter lead times for their generative AI initiatives.
In light of these trends and developments, there is one unchanging variable. The journey of integrating generative AI into business operations requires a comprehensive strategy — not merely for rapid technological adoption — but to optimize the most important aspect of this technology: The humans using it to drive innovation.
Improving the quality of data is crucial for the successful implementation of generative AI, and businesses need to prioritize this.
