Will 2025 be the year we seriously move from GenAI hype to GenAI results? Recent research suggests yes, particularly for the UK, which could see its economic growth almost double over the next 15 years thanks to this cutting-edge technology.
However, every technology leader knows that they cannot predict all the advances on the horizon, even while recognizing their responsibility to plan for the future as much as possible. Across all sectors, leaders are faced with the task of taking the plunge and investing in technologies like artificial intelligence tools to future-proof their businesses. But without the right adoption strategy and plan, you can end up drifting without a clear idea of where you’re headed next.
Walking this tightrope requires a pragmatic approach, leveraging the best tools available while maintaining flexibility and control. Practical implementation of GenAI is not about rigidly committing to a single path. Rather, it’s about creating an AI ecosystem that adapts and evolves with the needs of your business. That could mean choosing platform-agnostic solutions to avoid vendor lock-in, adopting open source to benefit from flexibility and transparency, adopting hybrid and multi-cloud strategies to ensure the best environment for your AI workload, or focusing in right-sizing your AI solutions. .
CTO of Dell Technologies UK.
Pillars for the practical implementation of GenAI
Partnering with technology providers can ensure customers harness the power of AI, addressing the complexity, risk and cost of diving into and supporting AI now and in the future. By offering flexible consumption models, an end-to-end AI-optimized IT infrastructure portfolio, an open ecosystem of deep partnerships with other leading AI companies, and a commitment to open standards, they can support a GenAI deployment that aligns with the unique characteristics of a company. needs, risk tolerance and long-term vision. In short, they can help ensure a strategy that is not only cutting-edge but also pragmatic and sustainable.
We can do that for our customers thanks to the lessons learned on our own AI journey. By implementing AI within our own operations, we have gained first-hand experience of its challenges and opportunities, giving us a deep understanding of what works and what doesn’t in real-world business environments. Our “zero customer” approach, where we become our first and best customer, ensures that our AI solutions are not just theoretical concepts, but are based on practicality, refined through real-world experience and are ready to offer tangible results to our clients.
Through that practicality, we developed these five guiding principles to help you more quickly and efficiently deploy AI technologies that will serve your business today and prepare you for your business tomorrow. These pillars for practical GenAI implementation are a testament to our own journey and our commitment to helping customers simplify complex technology.
1. Business data is your differentiator
Never lose sight of the fact that your data is a goldmine of insights and, unlike your competitors, you have exclusive access to it. You have a treasure trove of operational, customer and market data – information that reflects your company’s unique journey and experience. This data is the secret to success in the AI career.
By building on pre-trained models and personalizing them with your proprietary data, your differentiator, you can gain a competitive advantage through deeper customer insights (AI can analyze your customers’ data to uncover hidden patterns and predict future behaviors), management proactive risk management (AI can detect fraudulent transactions in real time by analyzing customer patterns and flagging anomalies) and better decision making (AI can analyze large amounts of data to identify trends, forecast demand and optimize pricing strategies , providing you with the information you need to be smarter, make faster decisions).
2. Respect data gravity
Although data can be a hidden treasure, it is never all in one container. Data is highly distributed: most resides on-premises and more than 50% of enterprise data is generated at the edge.
For data to be effective, it must be close to applications and services that depend on it for efficient processing and analysis. Better to give in to “data gravity” and bring AI to the data (where most of it is local) rather than moving business data to available computing resources. Most organizations find it more effective and efficient to train and run AI models locally to minimize latency, reduce costs, and improve security. To turn data into actionable insights with AI, often in real time, a combination of on-premises, edge, and cloud deployments is vital. For this reason, 66% of UK decision makers prefer to create an on-premises or hybrid approach to AI use and procurement.
3. Right-size your AI infrastructure
There is no one-size-fits-all approach when it comes to AI. I’ve witnessed customers across multiple industries, in organizations of varying sizes, deploy their AI in countless ways, from locally on devices and at the edge to massive hyperscale data centers. Not all models are large nor do all AI workloads run in a data center. Or in the cloud. To avoid over- or under-provisioning, it will be important to tailor the AI solutions you adopt to your use case and requirements, so analyze your use cases and goals to determine the most appropriate infrastructure and model types.
4. Maintain an open modular architecture
Equally important is to keep in mind that the AI landscape is constantly evolving and that no one can predict its future course. This means that a rigid, closed system can quickly become obsolete. Maintaining an open modular architecture will therefore be crucial to helping companies adapt to rapid changes in AI technologies and avoid becoming trapped in outdated or inflexible architectures.
AI/GenAI workloads are a new class of workload that requires a new class of open, modern innovation that spans the entire AI domain: data lakes and layers, compute, networking, storage, data protection and AI software applications. But it’s entirely plausible, if not likely, that new GPU infrastructures, algorithmic infrastructures, or inventions could emerge in the future that require companies to adapt. The worst mistake you can make today is to bet and commit to a closed, proprietary, one-dimensional, and inflexible AI system.
Open standards AI tools offer flexibility, transparency, and a vibrant community of support and innovation. By integrating open standards solutions into their AI strategy, companies can avoid relying on a single vendor and customize tools to meet their specific needs.
5. Forge a thriving AI ecosystem
No vendor can solve all AI challenges alone; collaboration is key. AI is a combination of many technologies, intellectual capabilities and services, which companies will need to combine with each other to be successful. Be sure to adopt vendors that enable an open ecosystem of partners, from major AI players like Microsoft to silicon vendors like NVIDIA and Intel and open source leaders like Hugging Face.
Open ecosystems provide equal opportunity across the technology ecosystem, support the creation of new GenAI advancements, and provide customers with greater access to innovation and flexibility. Access to open models and technologies can accelerate progress and solve problems around the world, driving a global “innovation engine” in all corners of the industry, from individual developers and startups to the public sector and business organizations.
A real-world approach to real-world results
Successfully navigating a new landscape almost always requires a pragmatic approach that balances excitement with realism, preparation, and careful execution. Being able to derive value from new technologies requires the creation of strategic roadmaps, and when it comes to AI, the preparation, quality and storage needs of the data that feed it have added importance. Don’t get caught up in the feeling that you need to transform into an AI powerhouse overnight. Start by identifying a specific, achievable goal that has the ability to generate a business return on investment and strengthen the path to success with a clear vision and the right partnerships.
We’ve compiled a list of the best free cloud storage.
This article was produced as part of TechRadarPro’s Expert Insights channel, where we feature the best and brightest minds in today’s tech industry. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing, find out more here:
