OUEN Corporation's Dream: "We want to surprise with technology"

Our company endorses April Dream, which aims to make April 1st a day to announce dreams. This press release is OUEN Corporation's dream.

Approximately one year since the highly acclaimed press release of Zero-Learning AI.

Approximately one year has passed since we released Zero-Learning AI as a new feature of our integrated generative AI inspection software, Inspection Designer. Thanks to the great response, we have engaged in many dialogues with various people.

Sites that considered and adopted it

• Inspection during the prototype stage of industrial products. Zero-Learning AI can operate even in phases where it is difficult to collect defective samples.

• Appearance inspection of consumer products with strict appearance quality but short lifecycles. It overcomes the conventional barrier of high learning costs with each product changeover.

• Inspection of food ingredients where the condition of good products varies by origin and season, and it is difficult to anticipate what defects might be mixed in. It can handle sites where defect patterns are unpredictable.

Feedback received from the field

• "I've been searching for a product like this for several years. I finally found it."

• "This AI is easy to operate on-site."

• "It's amazing that it can detect so much with such short learning times."

• "It's an AI that combines the best aspects of both good product learning and defective product learning."

While sincerely accepting these voices, we would like to reconsider our "dream".

One question that drives us

OUEN Corporation defines its cherished purpose as follows:

"To create new experiences that exceed expectations with technology, ideas, and human capabilities."

This statement might sound nice. But for us, it is the criterion for our daily work decisions. "Does this proposal exceed the customer's expectations?" "Is this explanation honest about the facts?" "Can this technology truly be used on-site?"—We move forward every day by continuously asking these questions.

So, why did we come to ask such questions? Behind it lies a certain "lack of freedom" that we have repeatedly witnessed in manufacturing sites.

Technological disparity creates a lack of freedom

Technology can be presented in a "conveniently selective way" to those who do not understand its mechanisms. It's easier and even more efficient in the short term to line up pleasant words, hide what cannot be done, and secure contracts while fostering excessive expectations.

However, what happens as a result?

"It didn't work as expected after introduction," "Ultimately, inspections are still done by human eyes," "AI is useless"—such voices begin to be heard from the field. This is not a failure of technology. This is distrust born from a lack of integrity.

There is a structural information gap between the provider and the user. Users are shown only the convenient aspects of technology and regret their purchase. We have witnessed such situations repeatedly. And that accumulation quietly erodes trust in technology itself.

Technological disparity is not merely a difference in knowledge. As a result, it deprives people of freedom of choice. If you don't understand correctly, you can't choose correctly. If you can't choose, you can't gain the power you should have. We believe that a world where the size of the weapon you can use is determined by whether you understand technology is not fair.

Our chosen approach: Being honest

That's why we value one thing when facing customers and the field:

What can be done, can be done. What cannot be done, cannot be done. But we make every effort to eliminate what cannot be done.

We communicate this directly. We don't create excessive expectations. And then, we do our best to make them happy.

This might be a humble stance in a sense. It might be easier to close a deal by asserting, "AI can solve any inspection." But we don't choose that path.

Because we believe that a "surprising" experience can only come from an honest relationship.

A "surprise" is born only when someone who introduced a solution with the correct expectations achieves results beyond what was anticipated. That experience will not come to someone who was shown an exaggerated dream and then disappointed. When the person who chose the solution and the person who actually uses it are pleasantly surprised. We truly believe that this continuous cycle will change one company's site, foster that company's competitiveness, and eventually become a force that moves society.

That's why refining solutions and accurately communicating them are equally important tasks for us.

The reality facing Japanese manufacturing and the hope beyond it

Manufacturing supported Japan's rapid economic growth. That is an undeniable fact. However, today, the business environment for domestic manufacturing is becoming increasingly severe due to the dual headwinds of globalization and a declining productive population.

Nevertheless, we see hope here.

The reality of "having to produce with fewer people" is, conversely, a positive hurdle for labor-saving and automation. History proves that challenging environments can be the seeds of innovation.

Furthermore, what Japanese manufacturing boasts to the world is meticulous quality control and the know-how of skilled workers. However, behind that strength lies a certain difficulty. High-mix, low-volume, and highly fluid production forms have resisted simple automation. Each time the product type changes, the setup changes, inspection standards change, and processes arise that must rely on human eyes and experience. This is the reality of Japanese sites.

The difficulty of the solutions that can respond to this increases accordingly. However, if we can create something that can adequately respond here—a solution that solves this difficult problem domestically should also be applicable to manufacturing sites overseas in the future. We view the "difficulty" of Japanese manufacturing as, at the same time, a breakthrough to the world.

Zero-Learning AI: The first, and a significant step

The "Zero-Learning AI" introduced at the beginning is the embodiment of that vision in technology.

Structural challenges faced by conventional AI

For a long time, the world of AI visual inspection has faced the challenge of "requiring large amounts of training data."

When trying to introduce conventional AI visual inspection systems to high-mix, low-volume sites, new data had to be collected and relearned each time the product type changed. The cost and time involved became a barrier to adoption. This is one reason why, even when told "it's theoretically possible," it often wasn't practical on-site.

Rewriting common sense with proprietary technology combining generative AI

We tackled this structural challenge head-on with a unique approach that combines generative AI technology.

Zero-Learning AI, literally, can perform visual inspections without prior training data. Even if the product type changes, there is no need to relearn from scratch. This rewrites the common sense of AI visual inspection, which states "large amounts of learning are required," with the power of technology.

In high-mix, low-volume production sites, visual inspection is automated. Quality control that previously only skilled workers could judge is handled by the system. People can concentrate on more creative tasks. —That is the near future we envision with Zero-Learning AI.

And this is just the first step. We will continue to create "surprising" solutions, communicate them honestly, and nurture them on-site.

Our envisioned future

Short-term goals

To create a world where visual inspection in high-mix, low-volume production is automated by Zero-Learning AI.

We will eliminate the barrier of learning costs and deliver inspection lines that can adapt to any product change to more sites.

Medium-to long-term goals

A future where technological disparities are eliminated, and everyone can freely choose optimal technology.

Instead of people struggling to adapt to technology, technology adapts to people. "People can naturally enjoy the benefits of technology without being conscious of it"—we are seriously aiming for a future where such an approach becomes the standard for manufacturing.

To achieve this, we will continue two things: refining our solutions, and continuing to communicate them in a way that more people can correctly understand and choose. These two are the wheels of a car.

Technology adapts to people. Honesty fosters trust. Surprising experiences improve sites and improve society.

We, OUEN, will continue to stand at the forefront of this.

About OUEN Corporation

Our company is committed to unique AI development, providing innovative technologies such as large language models and image detection to manufacturing companies. Our mission is to actively and sincerely address the challenges of customers who are serious about problem-solving, and to develop together, revitalizing Japanese manufacturing with Japan-originated technology.

Contact Us

For more information on Inspection Designer or inquiries about its introduction, please visit the product website below:

https://ouenx.com/inspection_designer

For inquiries regarding interviews, please contact us using the inquiry form below:

https://ouenx.com/contact/

Company Information

Company Name: OUEN Corporation

Established: August 2023
Address: 1-21-1 Dogenzaka, Shibuya-ku, Tokyo

Business Activities: Research and development, sales of generative AI/AI software packages, consulting, corporate training
Website: https://ouenx.com