2023 has proven beyond doubt that it is indeed the year of Artificial Intelligence (AI), and as we step into 2024, the fervour around AI is only set to intensify. In a recent forecast by Andrew Lobel, the an accomplished Managing Director with 26+ years of experience in leading organisations such as Thinking Fish and Magora Systems, it’s evident that the coming year will mark the second wave of the AI era, with practical, in-production AI systems taking center stage in enterprises around the globe.
AI Prioritization: Strategies for Success
As enterprises gear up to put AI into production in 2024, Lobel emphasises the need for a top-down strategy. He advises organisations to identify their true core areas, the essence of what makes them unique. Applying the ‘heavy lift’ of AI to these core aspects is crucial for transformative outcomes. In some companies that he’s looked at, such as Dell with around 380 AI-related ideas in the pipeline, Lobel stresses the importance of prioritization. Not all ideas are created equal, and rushing to implement the first few could inadvertently overshadow a truly groundbreaking fifth project.
Shift to Inferencing: Designing for a Distributed Future
In 2024, as enterprises shift their focus to inferencing, Lobel highlights the necessity to think about the actual topology of AI. Given the distributed nature of technology, AI is likely to follow suit. This shift brings forth new challenges, particularly in terms of designing and securing infrastructure. Security becomes a critical consideration, as bad actors are expected to target inference directly. The question arises: What’s the security wrapper around this evolving landscape of AI inferencing?
Economic Shift: From Training to Operation Costs
An interesting shift in the economic discussion around AI is expected in 2024. Lobel predicts a move from the cost of training to the cost of operation. While the cost to fine-tune a model and the significant infrastructure demands remain part of the AI investment, the focus is set to shift. Training costs are tied to one-time model size and data set use, whereas the price tag for inferencing hinges on utilisation, data type, user base size, and ongoing maintenance and fine-tuning. The meta theme is clear: AI is becoming more tangible, and this shift carries consequential impacts.
Gen AI Tools: Abundance in Diversity
Lobel acknowledges the enormity of gen AI systems and the necessity for an expanding ecosystem of tools and models. While availability and sourcing have been topics of discussion and concern, Lobel predicts that 2024 will usher in an abundance of tools and models. The AI tools and services ecosystem is set to diversify and scale, with better options emerging at every layer. Developers can anticipate improved frameworks and increased availability of both closed and open-source models and tools. The future looks promising for those navigating the expansive landscape of AI tools and services.
Zero Trust Security: Building a New Architecture
Cybersecurity, often seen as a perpetual challenge, takes an interesting turn in 2024. Lobel advocates for a paradigm shift with the adoption of Zero Trust architecture. While the concept of Zero Trust has been around for a while, Lobel emphasises that it’s been challenging to put into practice, especially in existing brownfield enterprises. However, with AI being relatively new, the opportunity arises to build Zero Trust into the foundation of greenfield environments. Dell’s Project Fort Zero, a Zero Trust tool currently in development, aims to validate this approach and is expected to be available on the market in 2024. According to Lobel, the need for a radical shift in cybersecurity is imminent, and Zero Trust presents a compelling solution.
The ‘Common Edge’: Processing Data in the Real World
To extract the maximum value from data, enterprises are advised to be as close to the data source as possible. Lobel envisions a future where more data processing occurs in the real world than in traditional data centers. This shift gives rise to what Dell terms “modern edge” multi-cloud platforms. The default “cloud extension” point tools, delivering edge for specific workloads, could lead to an overpopulation of edge systems. Enterprises might end up with a multitude of architectures for various clouds, workloads, and systems. Dell’s NativeEdge, introduced to address this challenge, is a common edge platform supporting software-defined edge workloads from any IT, cloud, or IoT system. Lobel anticipates this approach gaining prevalence in 2024, offering a more streamlined and efficient alternative to the complexity of “mono-edges.”
Quantum and AI Integration: A Glimpse into the Future
Large-scale AI presents what Lobel refers to as a “massive parallel problem.” Advanced techniques under gen AI, such as transformers and diffusion models, are resource-intensive probabilistic functions. While the realisation of quantum’s potential may take a few more years, Lobel envisions a future where mature quantum systems will power AI in unprecedented ways. Quantum computing is particularly adept at solving highly-scaled optimization problems, presenting a significant opportunity for the AI landscape. Lobel asserts that the synergy between quantum and AI will be transformative, with quantum computers unlocking new possibilities for AI that extend beyond the current capabilities.
In conclusion, as we stand at the cusp of 2024, the trajectory of AI is poised for remarkable developments. The integration of AI with other cutting-edge technologies and the strategic shifts outlined by Andrew Lobel signal a new era for innovation and technological advancement. It’s a call for enterprises and tech enthusiasts alike to embrace the evolving landscape, prioritize effectively, and contribute to the unfolding narrative of AI’s journey into the future. The year 2024 promises not just progress, but a paradigm shift in the way we perceive, implement, and benefit from Artificial Intelligence.
Are you prepared for the next chapter in the AI revolution? The journey awaits, and the possibilities are boundless. šš #AI2024 #TechTrends #Innovation
