Seven artificial intelligence predictions for 2023

The potential effects of AI are diverse, as are the forecasts for everything from sentient to generative and responsible AI, collaboration, and automation. What will be important to IT leaders in 2023?

Machine learning will assist in the correction of AI bias. By leveraging information about that specific person, systems that 'know the customer' will also reduce bias in conversational AI. And that's just the beginning. Let's look at some other important trends.

AI investments to get clarity

CTOs should not purchase AI simply because it is AI or because it is the latest and greatest technology. Instead, CTOs should consider the possibilities that AI provides. How will it function in their particular organization? What changes will it make to business processes? That is crucial. Previously, you could say, 'We're implementing AI or digital transformation,' and get a blank check, but that's no longer the case. Organizations want to see results and be able to quantify their impact. A CTO must do more than declare that AI is the future to justify the budget they demand.

Groundbreaking impacts

Artificial intelligence will provide tremendous advances in treating medical conditions in the coming years. Consider Dr. David Baker, the 2021 Breakthrough Prize winner. Dr. Baker used AI to create entirely new proteins. This groundbreaking technology will have far-reaching implications in the life sciences, potentially leading to the development of life-saving medical treatments for diseases such as Alzheimer's and Parkinson's.

Keep an eye out for the quantum and quantum-inspired computing crossover from fundamental physics into informatics. While I'm not expecting a practical quantum computer, we will see a crossover. One of the more intriguing examples is the use of AI to tune quantum computers! The combination of advanced mathematics and informatics will unleash a new generation of engineers uniquely positioned to capitalize on the AI wave.

Artificial and human intelligence

While AI will be increasingly used to improve our collective user experience at scale, it will be balanced by appropriate human intervention. Humans using AI insights will be a more effective combination overall than either doing it alone. Depending on the industry and the function's criticality, this balance will be struck differently. According to new research, radiologists assisted by AI screen for breast cancer more successfully than when they work alone. The same AI produces more accurate results when used with a radiologist than when used alone.

Responsible and generative AI capabilities

Major trends are expected in responsible and generative AI in 2023. Responsible or ethical AI has been a hot topic for some time, but it will move from concept to practice in the coming year. Smarter technology and emerging legal frameworks for AI are also positive steps. For example, the AI Act is a proposed, first-of-its-kind European law designed to govern the risk of AI use cases. The AI Act, which aims to become law next Spring, could become a baseline standard for responsible AI, similar to GDPR for data usage. This will have an impact on AI companies all over the world.

Generative AI will also make significant progress over the next year. Recent models can easily generate realistic images and drawings from natural language descriptions. Such capabilities are now transitioning from cool functionality to actual business use cases. Dozens of companies sell products that will write you essays, ad copy, or love letters. Rather than searching through stock photography, you can type a query and receive a newly generated image. And this is only the beginning; we're only scratching the surface of generative voice and video applications, so it'll be interesting to see what innovations and use cases emerge in the coming year.

Better collaboration between IT and business teams

As businesses prepare for increased economic volatility in 2023, they will face increased pressure to not only do more with less but also to demonstrate AI's business value from the start. Even though IT leaders are aware of the advantages AI offers in terms of better automation, insights, and efficiency, a stronger level of business-IT collaboration is still needed to ensure the technology is meeting the needs of the business.

Another trend we're seeing is a continued push to integrate AI throughout the organization. Several software and hardware solutions, including data models and AI chips, are vying for a piece of the lucrative AI market.

Efficiency and output

The current debate over whether AI will become sentient and pose a threat to humanity vastly overestimates its current capabilities. AI already completes many tasks that would take humans thousands of hours to complete: beating chess grandmasters, identifying fractures in X-rays, choosing the fastest route for a delivery truck, and so on. However, AI does not 'understand' how it completes these tasks. It cannot, like a human, explain why one chess move is more strategic than another; it simply knows. However, AI can handle many tasks inside and outside the workplace.

To make the most of AI, we need to understand why it can do so much despite lacking human-like intelligence. Can AI, for example, perform many of the tasks that people perform in the legal industry, where lawyers are still paid in six-minute increments? Assigning more tasks to AI will significantly improve team efficiency and output.

AI-driven automation

Everyone recognizes the importance of automation, and almost anything can be automated in today's software-defined world. On the other hand, the decision point or trigger for automation remains one of the more difficult elements. This is where AI will increasingly come into play: AI can make more intelligent, less brittle decisions than traditional 'if-this-then-that' rules in automation.