Charting the roadmap for the era of pervasive AI

AI is the defining technology shaping the next generation of computing as a new wave of AI applications evolves at a rapid pace. From creating realistic virtual environments for gaming and entertainment to helping scientists treat and cure diseases or helping humanity better prepare for climate change, AI has the potential to solve some of the world’s most important challenges.

The explosion in generative AI and large language models (LLMs), coupled with the rapid pace of AI application innovation, is driving enormous demand for compute resources and requiring solutions that are performant, energy efficient, pervasive, and can scale from cloud to edge and endpoints.

IT leaders across the board recognise AI’s potential and are ramping up their investment, with more than two-thirds of IT leaders amassing budget for AI project implementation. There is also downward pressure where boards are pressing their CEOs to develop a strategy for incorporating AI throughout their business, even as many executives are still grappling with where to start.

However, the occasion continues to be earmarked by hesitancy and distrust. In fact, our own study finds that only half of IT leaders are confident in their companies’ ability to properly adopt AI.

As such, we can expect to see a significant majority of enterprises moving beyond the proof-of-concept stage and into the deployment phase. However, to fully realise the potential of what AI integration can bring to the organisation, it is key that leaders take measured steps and not jump in on the bandwagon without a clear vision.

To do so, leaders must first understand the immediate AI landscape and the key technologies behind it, the opportunities that lie with AI, and what can be done to tap into them, as well as the roles that individual organisations can play in advancing AI as a technology.

Demystifying AI in 2024

2023 was a banner year for GenAI. While the technology continues to offer unique opportunities and solutions, 2024 is likely to see the growing adoption of more sophisticated and personalised solutions on the enterprise front.

Also Read: India beats Singapore, US to rank highest for AI project implementation

For instance, enterprise AI customisation is on the rise, with businesses embracing tailored generative AI applications. These applications are designed to meet specific business needs by integrating proprietary data and help to ensure more accurate and relevant responses.

Proprietary models offer security-conscious organisations better oversight when it comes to the internal data used to train models. This could assuage some organisations’ worries about achieving accurate, fair and representative output using third-party models.

2024 will also witness the power of small language models. Only the very largest companies have the funds and server space to train and maintain energy-hungry models with hundreds of billions of parameters. Smaller models, meanwhile, are far less resource-intensive, but they yield better performance than training larger models on fewer data.

Multimodal AI systems will also take centre stage among IT leaders’ minds. The key strength of these systems lies in their capacity to integrate different modalities, creating a more holistic understanding of the input data. This integration goes beyond recognising individual components; it allows the AI system to interpret and respond to complex inputs that involve multiple modes of information.

What lies within the AI goldmine?

AI’s biggest impact on business stems from its ability to automate tasks and supplement employees’ abilities to further boost their productivity. The technology essentially supercharges data exploration to enable better decision-making. As a tool for future-proofing organisations, it equips them with the agility and insight to adapt to evolving markets and technologies, underpinned by its capability for continuous learning and improvement.

With these considerations in mind, it is imperative that companies starting out on their AI roadmaps pay close attention to macro industry trends for possible knock-on effects.

For instance, as larger organisations work towards bringing AI capabilities in-house, the demand for components such as the AMD Instinct MI300 Series accelerators, which can power even the most demanding AI and HPC workloads, are likely to skyrocket. Smaller businesses who are looking to emulate these strategies are at the risk of getting squeezed out as they might lack the financial capabilities to compete for resources.

In this regard, enterprises can also look towards alternatives such as AI-enabled PCs, especially if they are running smaller AI models right-sized to their needs. These PCs are powered by AMD Ryzen AI (AMD was the first to bring dedicated AI to the PC), and feature dedicated artificial intelligence hardware in their processors to not only elevate their day-to-day tasks, but also to build and deploy AI models and run them directly.

The initial conversations and experiments around AI have seen lukewarm responses from the employee populace. It is imperative that business leaders prepare their workforce by equipping them with the essential tools and resources needed to effectively engage with AI.

Building an AI-friendly environment requires comprehensive training for staff, motivating them to address daily business challenges using AI to achieve organisational goals. In the future of work, organisations must strategically navigate two essential fronts: attracting new talent while empowering the existing workforce through upskilling initiatives.

Playing the role of a proactive champion

Businesses need to recognise that AI isn’t flawless.  It comes with its own set of unique risks that many organisations are unequipped to deal with – or even recognise – due to the nature of the technology and how fast it is evolving.

Also Read: Rainy day resilience: Your guide to startup media crisis management

From lack of employee trust and unintentional biases to unethical applications and security lapses, these gaps result from the same mistakes that can sabotage any deployment of technology: inadequate planning, insufficient skill sets, lack of alignment to business goals and poor communication.

As such, business leaders should ensure that their ambitious AI strategy is informed by an equally robust governance framework that is designed to balance technological innovation with safety, ensuring AI systems do not violate human dignity or rights.

Transparent decision-making and explainability are critical for ensuring AI systems are used responsibly. AI systems make decisions all the time, from deciding which assets to generate to determining whether a certain operation process is expedited. It is essential to understand how AI systems make decisions to hold them accountable for their decisions and ensure that they make them fairly and ethically.

At the same time, another key component of a well-developed governance framework is the policies underpinning it. These approaches inform the measured and necessary approaches to achieve an open and accessible AI ecosystem, further promoting the advancement of the technology as a whole.

This means that where possible, AI teams should opt to adopt open-source tools – such as Hugging Face and PyTorch – compatible with various machine learning development platforms to benefit from the flexibility and community support. An open ecosystem promotes both competition and democratisation of the technology – key ingredients that would further accelerates the pace of innovation for the technology.

Along the same line, all enterprises should never discount partnerships with organisations to ensure each step they take is a concerted effort in the right direction. AI will be pervasive – spanning the cloud, enterprise, and end devices – and it would require two critical ingredients: a variety of compute engines optimised to deliver the right AI performance balanced with the efficiency needs of each device and an open ecosystem where developers are free to create incredible new applications that are easy to deploy.

The partner of choice should therefore be an expert with not only a comprehensive product and software portfolio, but also an active platform built off the back of open and accessible software tools to empower companies to tackle the workloads of today and the future.

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic.

Join our e27 Telegram group, FB community, or like the e27 Facebook page.

Image credit: Canva

This article was first published on May 3, 2024

The post Charting the roadmap for the era of pervasive AI appeared first on e27.

Source link


Leave a Comment