Little more than a year after OpenAI released its groundbreaking tool, ChatGPT, enterprises, services, and public institutions are racing to capitalize on the potential of generative AI with new applications and services. Gartner predicts that by 2027, generative AI tools will be used to explain and modernize legacy business applications, reducing costs by 70%.
While generative AI has clearly taken the front seat in optimizing business operations, there are three classic AI disciplines that are the top contenders to help businesses effectively accelerate data utilization. The trifecta of AI workloads for operational excellence include conversational AI, computer vision and digital twins.
Amid the generative AI boom, industries are leveraging the power of conversational AI capabilities to advance their products and services. These services can help people easily communicate with a wide variety of devices and systems, including their cars, phones, service kiosks and more.
Conversational AI applications can help streamline customer and patient management, reduce wait times and drive improved productivity, accuracy and overall satisfaction in high-traffic environments.
These examples showcase how industry leaders are leveraging conversational AI today to massively expand access to their global customer base:
Customizing speech applications is beneficial for improving customer and employee experiences and relationships. Providing differentiated, end-to-end services to clients through digital, data and cloud platforms is top of mind for enterprises everywhere.
For example, NCS, a technology partner of the Singapore government, designed Breeze, a local navigation app with a native Singaporean voice that can pronounce local phrases and street names in Mandarin, Hokkien, Malay or Tamil.
Applying AI to voice recognition increases the scale and speed of customer satisfaction and service. For example, Artisight, an operations company located in the Midwest, offers an automated smart hospital service that reduces patient wait times by up to 50% through a voice-enabled check-in process. It also eliminates data entry errors and scales workflows to help top hospitals coordinate care for over 1,200 patients in a day.
Cities, retailers, transportation companies, and manufacturers are embracing the fundamentals of computer vision, which helps to automate processes for important use cases, such as detecting manufacturing concerns and increasing safety for factory workers.
Computer vision is a broad term for deep neural networks that develop human-like vision capabilities. The three most commonly recognized areas of computer vision are:
Computer vision is being widely embraced across these industries:
Computer vision can help companies detect unsafe working conditions, faulty equipment or hazardous materials to mitigate the nearly 340 million occupational accidents that occur every year.
Taipei-based manufacturer Pegatron uses computer vision to control its production of 300 products and over 5,000 parts a day. Its factory is able to detect defects, achieving 99.8% accuracy with its automated optical inspection systems.
Retailers need computer vision for sales optimization, customer satisfaction, reducing theft, and inventory management such as adding robots for stocking shelves.
Kroger, the nation’s largest grocer, is reimagining the shopping experience using AI-enabled applications and services. With computer vision and analytics, Kroger can detect early signs that produce is no longer fresh, as well as optimize store efficiency and processes. Developers can customize and extend AI workflows to provide analytical insights such as store traffic trends, number of customers with shopping baskets and aisle occupancy.
Whether you live in the dense woodlands of California or urban areas affected by wildfire smoke from hundreds of miles away, people look for answers to the millions of acres burned every year.
Chooch, a Silicon Valley-based startup, offers a generative AI tool that creates descriptions of images that can discern when smoke is present. It’s able to greatly reduce false positives and it gives California firefighters a dashboard on their smartphones and PCs, populated with real-time alerts for quicker response times.
More and more customers across industries are digitizing their industrial datasets and systems, creating digital representations of physical objects such as homes, factories, computers or wind turbines.
These digital twins allow for complex simulations, reducing risks and encouraging collaborative work. The ability to simulate thousands of “what-if” scenarios can save significant time, budget and resources.
Industries are paving the way in digital twins usage in the below areas.
Labeled the machines of the future, robots have been used in factories ever since the 1960s. Building on its use of robotics in its logistics over many years, BMW recently began using AI, digital twin simulation and high-performance server technology to enhance its robots’ abilities to recognize obstacles such as forklifts, tugger trains and people.
The technology helps BMW’s 57,000 factory workers more safely and efficiently share workspaces with robots in factory lines, where up to 10 different car models can be produced in 2,100 possible ways.
According to the American Council for an Energy-Efficient Economy, energy efficiency can make big reductions in greenhouse gas emissions and other pollution while saving consumers money and achieving other benefits.
Siemens Gamesa Renewable Energy creates physics-informed digital twins of wind farms to produce electricity for schools, homes, hospitals and factories. Its thousands of turbines generate over 100 gigawatts of wind power – enough to power nearly 87 million homes annually. AI can help Siemens quickly optimize wind farm layouts and increase energy production while reducing loads and operating costs.
Telcos plan to deploy over 17 million 5G microcells and towers worldwide by 2025. Building, managing and optimizing this new infrastructure is the industry’s next big challenge. HEAVY.AI, based in Silicon Valley, offers HeavyRF, a network planning and operations tool-based platform for creating digital twins. HeavyRF allows telecommunications companies to test radio frequency propagation scenarios in seconds, resulting in time and cost savings as base stations and microcells can be more accurately placed during installation.
Businesses can build robust AI strategies by using computer vision to boost safety, conversational AI to transform customer experiences, and digital twins to optimize operations and costs.
These foundational AI disciplines are crucial for businesses to excel on their path to mastering data usage. By harnessing them, enterprises can gain a competitive advantage, driving higher revenue, greater customer loyalty and improved productivity.
About the Author:
Himanshu Iyer, Senior Industry Manager, Manufacturing, NVIDIA
The post Traditional AI with Generative AI: The Next Wave of Enterprise Innovation appeared first on eWEEK.
