From individuals using ChatGPT for fun or function to enterprises applying artificial intelligence (AI) to automate tasks, the use of AI tools has exploded over the last few years. AI tools are becoming commonplace in today’s business.
Getty
According to IBM, 43% of CEOs are using AI to inform strategic decisions, and 75% of CEOs believe having the most advanced generative AI is the key to success.
The possibilities offered by AI are staggering, with almost limitless global potential. But to achieve that promise, operators of critical enterprise and government infrastructure, like transit systems, hospitals, and financial institutions, need to start with the right technology foundation.
The limitless potential of AI
Commercialization of AI has made massive leaps forward in recent years and 2024 stands to mark another year of advancement. At the same time, innovations in quantum computing and digital twins for industrial environments are closely coupled with applications of AI. Together, these technologies are poised to help industries achieve unprecedented levels of productivity. AI alone is expected to support the creation of 97 million jobs by 2025 and contribute upwards of $4.4 trillion to the world economy every year. Goldman Sachs estimates that AI could raise the global GDP by 7% over the next decade.
Many of these gains will come from increased efficiency through the mainstream application of predictive analytics and other operational optimizations. It is claimed that AI could augment or automate up to 70% of work tasks. That could boost overall productivity by up to 3.3% and give workers time to focus on higher-value, more strategic growth tasks. But it’s not all about automation. AI can also support critical thinking and problem-solving, offering value to nearly every industry.
All of these capabilities help businesses do more with less, which is critical in the face of an aging workforce.
AI needs a strong technical foundation
As exciting as the potential of AI is, enterprises and governments need to consider what’s required to maximize the benefits, and start taking steps now to put those critical foundations in place.
Data is at the heart of AI. Processing massive lakes of relevant data requires significant compute resources. As the use of AI expands, so do the data lakes demanding more and more computing power across distributed (edge) and central locations, and relying on routing, transport, and data center infrastructures and architectures that can perform flawlessly and efficiently.
As my colleague Hardik Gohil explained in a recent interview, this often involves multiple Data Processing Units (DPUs), Central Processing Units (CPUs), and Graphics Processing Units (GPUs) all working together to process data and transfer it across workflows and between sites. Transferring data alone accounts for 25% of processing time. That means “networks must be efficient, lossless and have higher capacity because any bottleneck in a sub-optimal network would lead to a substantial impact on job completion time.” The bottom line is... your AI objectives may not be met otherwise.
Key elements of AI infrastructure that need to be in place to support your business applications integrated with AI in the future include:
- Computational power and accelerators – Massive amounts of data gathered from your operations (IoT data, performance data, and more) coupled and interpreted to make intelligent AI decisions require massive computing power.
- Scalable data storage and management – Data systems and storage must be able to expand rapidly as data demands grow and as data is harvested for decision-making.
- Cloud computing platforms – These systems offer access to virtually limitless data resources for training AI models, analyzing huge data sets, and extracting actionable insights.
- Fast, reliable network infrastructure, especially for cloud services – Cloud services depend on the ability to move data quickly, securely, and reliably. Modernized routing and optical transport networks deliver the capacity and performance needed to support the AI era.
- Data security and integrity systems, including robust protocols and governance frameworks – With more data in use and transit, it’s more important than ever to keep it safe and look to quantum-safe security to help future-proof data integrity.
- APIs and integration with existing systems – Introducing AI doesn’t mean replacing every system and component, so interoperability and compatibility with existing systems is a must.
- Monitoring and automation tools for system maintenance – As system complexity grows, it’s vital to automate monitoring and issues handling to ensure everything is running at the right performance levels.
- Compliance – Companies must ensure they remain fully in compliance with laws and regulations governing IT systems, data handling, and more.
- Scalability – Systems must be flexible to accommodate continuously growing and shifting application, bandwidth and performance requirements.
- Expertise and training – Continually evolving technology requires ongoing upskilling and reskilling. Choosing trusted partners and experts who are knowledgeable on AI, your operations and their unique challenges, as well as the foundational technology elements can accelerate your AI journey.
What well-implemented AI could look like
Enterprises and governments will need to make necessary investments to build the right foundations. With those elements in place, AI can help make our businesses and communities more productive and efficient.
Power utilities can integrate AI into their operational systems that intelligently shift electricity resources to better meet demand and detect anomalies in the grid before they become outages. AI tools can help better predict potential service-effecting scenarios to proactively avoid or minimize service disruptions. AI-enabled systems also can optimize the grid for maximum efficiency and sustainability.
In the finance and banking sectors, AI-enabled operations can improve transaction security, deliver advanced fraud protection, enhance analytics and forecasting, upgrade customer service, and boost regulatory compliance.
The healthcare industry is responsible for about 30% of the data generated worldwide, creating an opportunity to use AI to improve patient care, advance medical research, and optimize operations for efficiency, all while preserving the security of patient data.
Transportation operations like railways, aviation and logistics companies can use AI to make more accurate weather predictions so they can plan accordingly, improve capacity planning, and optimize routing to get people and goods to where they need to be faster and more efficiently.
Governments and first response teams can use AI paired with advanced sensing technologies to better predict and handle climate change impacts like forest fires, floods and other natural disasters. With training augmented by AI and industrial metaverse simulations, as well as enhanced situational awareness tools, agencies can respond more effectively and save more lives.
It’s time to build the foundations
The usefulness of AI depends on speed, capacity and intelligence to process mass amounts of data from various sources and meet substantial real-time data transfer requirements. It will take network infrastructure that’s resilient and robust and provides fast, low-latency connectivity, along with cloud systems that are automated, open and flexible. Indeed, each of these business-critical, mission-critical and society-critical outcomes are heavily network-dependent, and rely on a robust, performant and secure underlying network infrastructure that includes data centers as well as the routers and optical transmission equipment that seamlessly interconnects them.
With these foundations in place, AI can live up to its limitless potential, improving operations and productivity exponentially. It can enhance business operations, deliver the optimized augmented work environments we envision, and help us solve some of society’s most difficult challenges. The potential is clear, and now is the time to modernize the network infrastructure that will accelerate the future.
