Is 2025 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, all technology leaders know that they cannot predict every advancement on the horizon, although they recognize their responsibility to plan as much for the future as possible. Across industries, leaders need to take the plunge and invest in technologies like AI tools to future-proof their businesses. But without the right adoption strategy and plan, you can find yourself adrift without a clear idea of where you’ll go next.
Walking this tightrope takes 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 your business needs. This could involve choosing platform-agnostic solutions to avoid vendor lock-in, adopting open source to benefit from flexibility and transparency, adopting hybrid and multi strategies. -cloud to ensure the best environment for your AI workload or to focus on right-sizing your AI solutions. .
CTO at Dell Technologies UK.
Pillars for practical implementation of GenAI
Partnering with technology providers can ensure customers harness the power of AI, tackling the complexity, risks and costs of immersing and supporting AI, today today 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 implementation that aligns with the unique characteristics of a business. needs, risk tolerance and long-term vision. In short, they can help ensure a strategy that is not only forward-thinking but also pragmatic and sustainable.
We can do this for our customers through lessons learned during our own AI journey. By implementing AI within our own operations, we have gained first-hand experience of its challenges and opportunities, allowing us to deeply understand what works and what doesn’t in real-world business contexts. Our “zero customer” approach, in which we become our first and best customer, ensures that our AI solutions are not just theoretical concepts, but are grounded in practicalities, refined through real-world experience and ready to deliver tangible results to our clients.
With this practicality in mind, we’ve developed these five guiding principles to help you more quickly and effectively deploy AI technologies that will serve your business today and prepare you for your business tomorrow. These pillars of practical GenAI implementation are a testament to our own journey and commitment to helping customers simplify complex technologies.
1. Business data is your differentiator
Never lose sight of the fact that your data is a gold mine of information and that unlike your competitors, you have exclusive access to it. You have a treasure trove of customer, operational and market data – information that reflects your company’s unique journey and expertise. This data is the secret to success in the AI race.
By leveraging pre-trained models and customizing them with your proprietary data, your differentiator, you can gain a competitive advantage through deeper customer insights (AI can analyze your customer data to uncover hidden patterns and predict future behavior), proactive risk management (AI can detect fraudulent transactions in real time by analyzing customer trends and reporting anomalies), and improved decision-making (AI can analyze large amounts of data for identify trends, forecast demand and optimize pricing strategies – giving you the insights you need to make smarter, faster decisions).
2. Respect data gravity
Although data can be a treasure, it is never united into a single treasure. Data is highly distributed, with most residing on-premises and more than 50% of enterprise data generated at the edge.
For data to be effective, it must be in close proximity to the applications and services that depend on it for efficient processing and analysis. It’s better to give in to “data gravity” and bring AI to the data (the majority of which is on-premises) rather than moving the company’s data to available IT resources. Most organizations find it more effective and efficient to train and run AI models on-premises to minimize latency, reduce costs, and improve security. To transform data into actionable insights using AI, often in real time, a combination of on-premises, edge, and cloud deployments is essential. For this reason, 66% of UK decision-makers prefer to develop an on-premises or hybrid approach to the use and procurement of AI.
3. Scale the size of your AI infrastructure
There is no one-size-fits-all approach to AI. I’ve seen customers across multiple industries, in organizations of varying sizes, implement their AI in many ways: locally on devices and at the edge to massive hyperscale data centers. Not all models are large, and not all AI workloads run in a data center. Or in the cloud. To avoid massive 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 and modular architecture
It is equally important to keep in mind that the AI landscape is constantly evolving and no one can predict how it will evolve in the future. This means that a rigid, closed system can quickly become obsolete. Therefore, maintaining an open and modular architecture will be crucial to help businesses adapt to rapid changes in AI technologies and avoid getting locked into outdated or rigid architectures.
AI/GenAI workloads are a new class of workloads – requiring a new class of open, modern innovation spanning the entire AI domain: data layers and lakes, compute, networking, storage , data protection and AI software applications. But it is entirely plausible, even likely, that new GPU infrastructures, algorithmic infrastructures or inventions will emerge in the future, which would force companies to adapt. The worst mistake you can make today is to gamble and commit to a closed, proprietary, one-dimensional, and inflexible AI system.
Open standards AI tools provide flexibility, transparency, and a vibrant community for support and innovation. By integrating open standards solutions into their AI strategy, companies can avoid being beholden to a single vendor and customize tools to meet their specific needs.
5. Forge a thriving AI ecosystem
No single vendor can solve all AI challenges; collaboration is essential. AI is a collection of many technologies, intellectual capabilities and services that businesses will need to integrate with each other to succeed. 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 to open source leaders like Hugging Face.
Open ecosystems provide equal opportunity across the technology ecosystem, support the creation of new GenAI advances, 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, fueling a global “innovation engine” across all industry sectors, from individual developers and startups to to the public sector and businesses.
A real-world approach for real results
Successfully navigating a new landscape almost always requires a pragmatic approach that balances enthusiasm with realism, preparation, and careful execution. Being able to take advantage of new technologies requires the creation of strategic roadmaps, and when it comes to AI, the needs for preparation, quality and storage of the data that powers it take on increased importance. Don’t get caught up in feeling like you need to transform yourself into an AI powerhouse overnight. Start by identifying a specific, achievable goal that can generate ROI for the business, and strengthen the path to success with a clear vision and the right partnerships.
We have 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 technology industry. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you’re interested in contributing, find out more here:
