Info-Tech has released the Artificial Intelligence (AI) Trends 2023 Report. It details an overview of the AI trends that continue to drive innovation and new opportunities for organizations throughout the year.
report written by Irina Sedenko and Anuradha Ganesh Categorized into eight trends: AI Design, Event-Based Insights, Synthetic Data, AI at the Edge, AI in Science and Technology, AI Inference, Digital Twins, Combinatorial Optimization
Trend highlights include:
AI design: Info-Tech reports that the design of AI systems will change as technology spreads. Designing a sustainable AI system should consider several aspects, including the system’s business applications, data, software and hardware, governance, privacy, and security. According to the report, an AI system design approach should cover all stages of the AI lifecycle, from design to maintenance. It also needs to support and enable iterative development of AI systems.
Leveraging a variety of tools and technologies to develop, deploy, and monitor AI systems requires consideration of software and hardware needs in AI system design.
AI in science and engineering:
This report details the impact AI has and will continue to have on the fields of science and engineering. AI can help sequence genomes to identify variants in her DNA in individuals that exhibit inherited disorders. It enables researchers to model and compute complex physical processes, predict the origin of cosmic structures, and understand planetary ecosystems to advance climate research. AI could advance drug discovery.
“AI will play a bigger role in science, enabling scientists to innovate faster,” reports Info-Tech. Continuing to contribute more to science by assisting scientists with research that helps find new insights, generalize scientific concepts, and transfer them between fields of scientific research.
Using synthetic data and combining physical and machine learning models and other AI/ML advances will accelerate the use of AI in science and engineering, the report adds.
Synthetic data:
Synthetic data is artificially generated data that mimics the structure of real data. It is used to train machine learning models when there is not enough real data or the existing data does not meet a particular need. Synthetic data allows users to remove contextual bias from data sets containing personal data, prevent privacy concerns, and ensure compliance with privacy laws and regulations.
Synthetic data is now used in language systems, Info-Tech reports, self-driving car training, fraud detection improvement, and clinical research. In the future, synthetic data has the ability to grow across all industries and AI applications by enabling access to data for every scenario, technology, and business need.
Digital twin:
A digital twin (DT) is a virtual replica of physical objects, devices, people, places, processes, and systems. DT and AI technologies have enabled organizations to design and digitally test equipment such as aircraft engines and wind turbines before they are physically manufactured, reducing costs and making the entire process more efficient.
Info-Tech says the future of this technology will include enabling autonomous operation of DTs. Advanced DTs must be autonomous as they can self-replicate as they move to multiple devices. “Such DT autonomous behavior will have implications for the growth and further progress of AI,” the report adds.
Edge AI:
Edge AI integrates AI into edge computing devices for more seamless data processing and smarter automation.
Key benefits of Edge AI include real-time data processing capabilities that reduce latency and enable near-real-time analytics and insights, and cost and bandwidth savings because data does not need to be transferred to the cloud for computing. Includes reduced requirements. It also helps improve automation by training machines to perform automated tasks.
Challenges and solutions
Info-Tech’s report also describes the challenges that have slowed the adoption of AI.
Some of these include data quality issues such as lack of harmonized systems and harmonized data. The report notes that the lack of tools and technology to operationalize models created by data scientists is also slowing AI adoption. Additionally, there is a general lack of understanding of AI use cases, including how AI and machine learning (ML) can be applied to solve organizational problems.
Solutions that accelerate AI adoption include improved data management capabilities such as data governance and data initiatives. Info-Tech has also found that increasing the availability of cloud platforms can help improve the operational capabilities of machine learning.
