The AI hype cycle exploded in 2023 and exploded under the generation of AI and subsequent injections of funds. It brings a blind AI optimism that the organization advocates the technology without a clear understanding of its ROI and actual use cases. Some people simply follow the AI crowd and adopt the technology out of fear of being left behind. Looking back and thinking about what is going to happen in 2025, has AI expectations changed a lot? Are we still in the stage of optimism for the blind AI?
In short, no. Fortunately, we moved along the mature path. We can see the hype cycle fade away and from blind AI optimism Proven AI Optimistic – or reliable AI. Manufacturing, which has made great progress in reliable AI, is a case study of this journey, and other industries can learn from others. But before we go this path, we must address the real possibility of potentially ruptured AI bubbles.
Irrational AI is vigorous?
Blind AI optimism – or excitement about the latest, brightest AI technology, without a clear understanding of its meaning and tangible achievement – has attracted a lot of attention and capital. For example, analysts are watching Microsoft, Meta and Amazon make substantial investments in NVIDIA’s AI-powered GPUs, but there are concerns that these investments won’t bring the revenue gains these companies are looking for.
We start to see this particular AI bubble bursting whisper. MIT economist Daron Acemoglu warned that the money pouring into AI infrastructure investments may not match investors’ ROI expectations. People are excited about the promise of AI, but now they are starting to worry that it will reflect the internet bubble. Such events may trigger other investors to be more suspicious of the AI narrative and seek faster return times or reduce these investments. Disillusionment is bubbled.
There is no doubt that AI will change the way the industry works, but that won’t happen by following the shining object. Reliable AI is quantifiable and has a real impact, often behind the scenes and embedded in existing processes.
So, what is an example of reliable AI that has shown success and will stand the test of time? The manufacturing industry presents important use cases.
Measuring the success of manufacturing
A leading chemical company wants to improve the efficiency and reliability of its machines to avoid downtime and operational disruptions in undecided decisions. They invest in AI-driven predictive maintenance solutions that give their teams the insights and suggestions of machine health to proactively solve problems. They achieved a 7x return on investment in less than a year.
Similarly, one of the world’s top food and beverage companies wants to reduce product waste and optimize their plant capacity, so they have AI-enabled machine monitoring in four plants. They see a 4,000-hour increase in capacity each year, and a decrease in waste of more than 2 million pounds of products. The result was very influential, and the pilot expanded it to all its North American facilities.
These real-world examples demonstrate the measurable impact of reliable AI, which are consistent with broader industry trends. Among the recent surveys of over 700 global manufacturers, the highest areas for quantifying the impact of AI on business goals are supply chain management/optimization (41%), improving decision making through prescriptive analysis (41%), and process health/maximum change. Yield and capacity (40%).
The year’s findings reveal real progress in this journey from blind optimism to reliable results. Many respondents are now able to quantify AI’s impact on process health compared to the previous year and can measure its impact on unplanned downtime. This shows that manufacturers are becoming increasingly comfortable with using AI, which helps them achieve a deeper return on investment.
As this confidence increases, 83% of global manufacturing leaders are increasing their AI budgets – a key to business growth and effectively visualize and act on plant data. So, what about other industries that lag behind in AI success? Their expansion is not fast enough.
Slow zoom
So far, manufacturers and other industry leaders have slower scalability AI scale, which has hindered our speed of seeing meaningful results. In fact, according to a Tech.co report, nearly 7 of 10 (67%) business leaders are slowly adopting AI.
AI is a tool, not a result. To achieve the real benefits of these investments, a cultural shift has to be made – it’s not just about putting sensors on the machine. Skilled labor is already difficult to retain, or even harder to find. The U.S. population ages faster, while fewer people enter the workforce. Now is the time to advance reliable AI, as it must retain knowledge and push the industry forward.
Generative AI tools like Chatgpt are impressive, but the business world needs more. It requires specially built AI for specific and difficult problems and requires results. This is where reliable AI comes in, and manufacturing provides an impressive script.
