Governments which recognize the importance of a Super Smart Society have been promoting initiatives to realize Society 5.0. From an R&D perspective, the most important thing to consider about these is how best to achieve the "diversity" and "change" that they require. As society changes to become more complicated and diverse, it becomes less feasible to have one rule apply to everyone, as in the 20th century. However, the way we think, the way our society works, and the way company systems are constructed are still running on 20th century ideas. We must change this.

The concept of everyone following a single rule, without exceptions, was very strong throughout the 20th century. It was believed that tolerance and promotion of diversity would incur management costs and reduce efficiency. The idea that everyone following the same rule would improve costs and efficiency still lingers in our current way of thinking. We must change that way of thinking and be flexible in tolerating diversity.

When we think about AI, no matter how deep we delve into its algorithms, or how much data we amass, there is actually rather little it can do. What's most important is its numerical results - its outcomes. The importance of its outcomes, as well as its nurturing of changes and diversity, are actually two sides of the same coin. As long as we are able to achieve desirable numerical results, it can be done by instituting diversity as necessary based on the needs of people, society, and local conditions – eliminating the necessity for everyone to follow the same rules. People and businesses each have their own backgrounds, histories, and intelligences. Fostering the diversity necessary to utilize these, is known as "outcome thinking."

When we think about what the most important outcome would be for society as a whole, we arrive at the word "happiness." The ancient Greeks and Confucius both used this as a word to represent the most extreme outcome for humans. If we could quantify this "happiness" and feed it into AI as data, we would be able to utilize the abilities of diverse people to improve the happiness of society as a whole. Search engines have been our means to increase human abilities for approximately the last 10 years. As a way to use the information available on the internet to increase our abilities, we have come to use them on a daily basis. However, this information is only a small part of the world's data. In real life, human behavior occurring in real time cannot be represented in search results. What's more, it can't be utilized as data. All the actions that happen inside a company go completely unused as data.

It will be necessary to take a very systematic approach in order to begin using these types of data. As the performers of these actions are humans, this also relates directly to happiness. This is because when placed in unhappy circumstances, humans become extremely unproductive. Although there are clues to solving this problem hiding in various places, they are difficult to see from inside unhappy circumstances, making the desire and innovation to overcome those circumstances and get to the next step all but impossible. The fact human happiness and the productivity of businesses and people are fundamentally related, is something that we in R&D see in quantitative data.

As AI changes, so too does other software. Conventional software performs its actions exactly as dictated by the logic of its code, never surpassing these actions. As such, the same input will necessarily produce the same outcome.