Industry executives and experts share their predictions for 2024. Read them in this 16th annual VMblog.com series exclusive.
AI and Edge Computing Will be Key Influences in 2024
By Bjorn Andersson, Senior Director, Global Digital Innovation Marketing and Strategy, Hitachi Vantara
The year ahead promises more practical AI applications that are more accessible to all industries. In addition, we’ll see a rapid increase in Internet of Things (IoT) devices, coupled with AI and edge computing, which will inform how companies think about their data storage, management, and analysis. Generative AI will continue accelerating the need for quantum computing, opening up more opportunities for players throughout the space.
The honeymoon is over – let’s put generative AI to work in real-life applications
AI has been an evolving field for over half a century, with AI technologies supporting many of the services we use today. Platforms like Amazon for shopping, Netflix for entertainment, and virtual assistants like Alexa and Siri have become integral parts of our daily lives. However, when ChatGPT burst onto the scene in late 2022, it rapidly gained millions of users and continued evolving at an unprecedented speed.
Generative AI now stands at the peak of Gartner’s hype cycle for emerging technologies. It has proven helpful in many applications, including content generation (text, video, and even music). It has helped professionals across various industries, such as technicians servicing machinery, programmers writing code, and marketers conducting competitive research. And the human factor remains essential when applying AI and large language models (LLMs) to real-life applications.
In 2024, there will be an increased focus on integrating domain expertise into applications using generative AI and LLMs to develop specific answers that can be applied to real-life situations. Companies must provide the necessary domain expertise and context to make the application more specific and significantly more accurate, which will prove crucial as the technology matures. In addition, there will be a growing demand for explainability and traceability of AI-generated outputs, with AI pointing out how it produced a particular answer, referencing source literature, for example, or sending generated source code through an independent system to provide a detailed explanation of its functionality.
2024: The year of digital technology for all
In 2024, we’ll witness the broader democratization of digital technology, including AI and IoT, ushering in a new generation of use cases we can’t predict today. Democratizing implies that the technology is accessible, easy to use, and affordable. ChatGPT represents a quantum leap in this direction. Its no-code approach empowers domain experts, even those without coding skills, to create applications that were previously unattainable.
Democratization of technology also paves the way for a collaborative and interconnected digital community, where the collective of diverse minds fuels progress and innovation. Right now, few companies are equipped with the internal expertise and capabilities required to create and sustain end-to-end solutions spanning the physical and virtual realms. To truly achieve digital technology for all, we must adhere to the teachings of software development experts Ron Goldman and Richard Gabriel: “Innovation happens everywhere, but there is simply more elsewhere than here.”
Plentiful IoT devices, along with AI, will blur the lines between real and virtual
Analysts offer different predictions for the number of new IoT devices that will be deployed in 2024. However, even the most conservative estimates project figures in the billions. It’s safe to say that several billion new IoT devices will be deployed, connected, managed, and actively producing massive amounts of data next year.
The expansion of IoT devices enables an unprecedented capacity to produce and ingest data about the real, physical world through the combination of many sensors and from diverse sensor types. In this, AI assumes a key role. It is well suited to interpret and extract insights from the complex web of data from various sensor types in real time.
Edge computing softens the impact of the continued IoT data explosion
Edge computing is the key to managing the vast data generated by the IoT explosion. Until now, companies have focused on near-term, exponential data growth. However, in 2024, there will be an exponential shift in data management. With the integration of edge AI and edge analytics plus more capabilities in edge devices, companies will be better equipped to address the challenges that arise from the surge in IoT data, making it more efficient and manageable.
Achieving this goal involves filtering unwanted data from cameras and sensors at the edge and refining it to reduce its size before it is processed and stored.
Gen AI will introduce perceived empathy in AI applications
In 2024, generative AI will teach empathy to AI applications, a well-known characteristic missing from virtual assistants and other AI we see today. Humans show empathy by attentive listening and responsiveness to real-time cues through hearing and seeing. Those real-time interactions are often lacking in AI applications.
Gen AI will bridge this gap by incorporating emotional recognition capabilities – or technology that understands a user’s emotional state or context – and LLMs to identify emotions and situations to generate real-life responses. This will allow AI applications to respond to real-time data, emotions, and nuanced contextual cues, mimicking empathy for a more human-like persona.
Cost and shortages pave the way for alternative AI accelerators
In 2024, the landscape of AI accelerators will transform significantly, impacting the efficiency and accuracy of computing tasks. Currently, the limited availability of accelerators has led to extended lead times and increased costs, forcing the makers, as McKinsey puts it, to create and train an LLM from scratch, with investments reaching the hundreds of millions. Alternative, specialized technologies are crucial and will create options for “Makers” and prompt other companies to redefine their roles.
Businesses will realize that they don’t all need to create their own LLMs from scratch, or become “Makers.” McKinsey’s framework says that they can also be “Takers,” utilizing existing services like ChatGPT or Bard via APIs, or they can be “Shapers” and usher in proprietary capabilities to existing public LLMs and shape them to their needs. Thus, an alternative hardware accelerator market will introduce a new wave of startups that give rise to these archetypes.
A VHS-Betamax-style war is brewing for quantum computing
It’s been a long time coming, but quantum computing shows signs of nearing a tipping point of market domination, reaching an estimated market size of $125 billion by 2030.
However, it’s not quite ready for 2024 market domination. Quantum computing is still too nascent. But that doesn’t mean the industry should ignore its potential and miss the early signs. As Ernest Hemingway puts it, it will happen “[g]radually, then suddenly.”
This is because of the unique ability of quantum bits, or qubits, to represent both 1 and 0 at any point in time, enabling massive amounts of parallel computations far beyond what a classical computer can do.
The noteworthy types of quantum computers have varied potential. The most advanced type is superconductive, which implements a qubit using superconductive circuits cooled by advanced and expensive cooling systems. Ion trap or cold atom types use electrons in fixed atoms to implement qubits. There’s also research around using light to create photonic quantum computers. The silicon type leverages silicon semiconductor technology that is well established and already proven at the nanometer scale. It may not be the absolute best performer, but it has the potential to scale more efficiently and is more cost-effective. It could win the race akin to VHS’s victory over Betamax.
Looking ahead, AI and edge computing will continue to evolve greatly, especially in 2024. Integrating gen AI and LLMs into real-world business applications and data management processes will be a priority for IT leaders. Companies that recognize the data needs accompanying this emerging technology will look for energy-efficient, cost-effective, and trustworthy models that they can adapt to deploy across their business sectors.
##
ABOUT THE AUTHOR
With an engineering and computer sciences background and his start at Sun Microsystems, Bjorn Andersson has worked in technology development, product management and marketing for more than 25 years. His interests include sustainability, digital transformation, data analytics, visualization, high performance computing and the Internet of Things (IoT). Today, he leads strategy and marketing for selected industry solutions practices at Hitachi Vantara.
