AI adoption is at a turning point. As cost pressures rise, companies are prioritising an increased return on investment (ROI) from their AI initiatives. Although widespread adoption of AI goes back just a few years, companies have learned a lot in that short time. Many pumped substantial resources into AI projects following the pandemic but, by the end of 2024, surveys showed that less than half of companies reported positive ROI from their investments.[1]
Gartner seems right when it says generative AI is sliding into the trough of disillusionment.[2] To change the situation, companies should look for opportunities to provide immediate value rather than broadly implementing AI and generative AI solutions. Starting small and scaling gradually is the answer. But how should a company start building an AI strategy that delivers value in the short term and grows returns over the long haul?
Start with high impact projects
Your first step should be aligning your AI objectives with your business goals. Ask yourself where AI could address specific pain points across your operation. For example, it could automate repetitive tasks to lower error rates and reduce customer churn through better customer support or personalisation. Refine your list of AI projects by identifying the ones that offer quick wins and significant results. Examples of high-impact projects include predictive lead scoring and intelligent customer support chatbots. AI-powered chatbots can save up to 30% in customer support costs.[3]
Get your data foundation in order
With your AI projects lined up, you should start thinking about your data foundation and whether it’s fit for purpose. Your data should be accurate, consistent, complete and timely. Moreover, there should be enough of it to support the specific business objectives you have in mind. Indeed, in the last few years, it’s become clear that data is a weak point for many companies, with one survey showing that 78% of companies lack the data foundation for generative AI projects.[4]
While a poor data foundation has slowed down many AI initiatives, other factors have also affected time to value. These include integration issues across AI models, frameworks and software, as well as essential data management. Moreover, synchronising data between on-premises and cloud infrastructures, which many companies have included in their strategies, has proven problematic, slowing project completion.
Invest in your talent and company culture
Your strategy should also reflect your organisation’s AI expertise and culture. Given the rapid development of AI coupled with IT workloads, it’s possible your IT team hasn’t had a chance to develop its AI talents. Bridging these skill gaps will take a determined effort because, according to current findings, 48% of tech leaders say the biggest barrier to AI success is a lack of staff with AI skills.[5]
Consider the broader culture of your business and whether employees are ready to embrace AI. Be prepared for resistance unless you build training and open communication into your strategy. The fact is that fear of job displacement is real among many who worry AI will make their roles obsolete. And, as with change in general, a lack of familiarity with AI breeds uncertainty and scepticism, making it critical to have a dialogue with staff about how the technology is there to help and not to replace them.
Plan for ethical and responsible AI
Your commitment to employee support should be part of a broader ethics and responsibility plan for AI. Your strategy should include your goal to use AI in ways that prioritise personal information privacy, fairness and transparency. Establish mechanisms to address bias issues and ensure regular audits to identify and correct biased outcomes.
Since trust is essential, prioritise the safeguarding and management data used by any AI model. Not only must your data be secure, but you should also have contingency plans to prevent unintended harm and ensure that data remains safe during system attacks. Additionally, for compliance, you need to enforce data privacy regulations and grant individuals control over how their data is collected and used.
An end-to-end enterprise solution
Increasingly, companies are building their strategies around end-to-end enterprise AI solutions that enable them to fast-track and maximise ROI. They are unlocking the potential for AI to lower costs, increase revenues and reduce customer churn through partnerships with organisations such as Dell Technologies and NVIDIA.
The Dell AI Factory with NVIDIA is a simplified, tailored and trusted solution that empowers businesses to gain more value from AI at speed. Through a combination of technology and services supported by Dell Technologies and NVIDIA, you can build your AI plan, implement practical solutions, drive behavioural change, ensure ethical and responsible AI, and exceed your business objectives.
The industry’s first and only end-to-end enterprise AI solution,* the Dell AI Factory with NVIDIA is designed to speed AI adoption by delivering integrated Dell and NVIDIA capabilities to accelerate AI-powered use cases, combine data and workflows, and enable you to create your AI journey for repeatable, scalable outcomes.
Discover more about our AI story and learn how the Dell AI Factory with NVIDIA can advance your AI strategy.
* Based on Dell analysis, July 2024. Dell Technologies offers solutions with NVIDIA hardware and software engineered to support AI workloads from PCs with AI-powered features and workstations to servers for high performance computing, data storage, cloud native software-defined infrastructure, networking switches, data protection, HCI and services.
[1] https://newsroom.ibm.com/2024-12-19-IBM-Study-More-Companies-Turning-to-Open-Source-AI-Tools-to-Unlock-ROI
[2] https://www.gartner.com/en/documents/5505695
[3] https://www.nimblework.com/blog/chatbots-for-customer-support/
[4] https://www.techmonitor.ai/digital-economy/ai-and-automation/survey-reveals-78-of-businesses-unprepared-for-gen-ai-due-to-poor-data-foundations?cf-view
[5] https://www.hrdive.com/news/tech-talent-is-hard-to-find/715784/
