JoinAI：From World Models to Utopia

Empowering Everyone to Build Their Own Digital Utopia

If you ask an entrepreneur about the reason why they started their business, the answers would be boundless.

In this world, some chase trends, others aim for the next big thing, and there are even those who might start a business simply because it's easy to secure funding. But there are also some "fools" who embark on the path of creation solely to realize a dream that has lingered in their hearts for years. We, the founding team of JoinAI, belong to this last category.

Our dream might sound a little bit crazy: to empower every individual with the ability to construct their own digital utopia. Hmm, if you're thinking about the film "Ready Player One" right now, I'm afraid that I'll have to ask you to pause for a moment.

More Than Just Another "Virtual World"

What we aim to create is not a fantastic metaverse game like the Oasis from "Ready Player One," or as miHoYo describes, a virtual world where everyone would want to live. If we must draw a comparison, it's closer to a world-building tool with capabilities similar to the Matrix from "The Matrix" movies. (It's important to note: we're only comparing the "world-creation ability" aspect of the Matrix here, not the parts about manipulating consciousness or its malevolent nature.)

The fundamental difference between a "virtual world" and the "real world" lies in our perception. The essence of human perception of the world stems from our multidimensional awareness of the "continuity of time and space," as well as our cognition of interactions and feedback with "time" and "space." The "world" possesses a randomness and diversity that a "virtual world" cannot replicate or imitate.

To put it more simply, unlike a virtual world which can be precisely defined and governed by finite rules, the 'real world' cannot be replicated with such 1:1 exactness. The 'real world' does not allow for completely identical people, identical buildings, or identical events. This inherent complexity and the physical properties of the world are deeply rooted in our subconscious, forming a clear boundary between what is 'virtual' and what is 'real.

Before the advent of GEN-AI, or more precisely, the "world models," such creation was impossible. Even the best engines lack the capability to simultaneously recreate the "world creation" under the rules of "spatial + temporal continuity" x "random events" x "physical properties" x "X" in the "real world." Without this, a true sense of realism is unattainable.

Take our Terra series model, which we are currently developing, for example. It can create a world where randomly inserted events affect the trajectories of internal objects, and these changes are both random and conform to physical properties. Of course, while Terra shows the embryonic form of this capability, we still have a long way to go before reaching the "Matrix" we envision.

Why Not Make a Game Engine?

You might ask, why pursue such a distant goal? Why not just build a game engine based on existing technology and use it to create an unprecedented metaverse game? Wouldn't that be more practical?

The answer is simple: our purpose cannot be fulfilled by a metaverse virtual world or game.

Our idea didn't come from all of us happening to watch the famous "The Matrix." In fact, it's because our team shares a common trait: we've all loved traveling since childhood. To this day, our footprints may have covered the globe. We've seen the most technologically advanced and wealthy places, and we've also been to the most chaotic and desperate places. We understand from the bottom of our hearts: this lovely world is unfair.

This world is always full of "things not going as wished" and "powerlessness to change." As small individuals, we often can only seek inner peace and courage to move forward on the basis of unconditionally accepting reality.

We unanimously, and perhaps naively, thought that if reality is truly helpless and desperate, perhaps we could at least have a perfect and fulfilling dream to inspire some courage . We consulted numerous social and psychological research results, which show:

A sufficiently good experience, even if it's a dream, has the power to heal the human psyche.

It is this understanding that drives us to create a new option: a place where every wounded soul can find redemption, where every difficult dream can gain strength, even if it's just a dream that's real and beautiful enough. What we hope to deliver to people is not just a fixed virtual world or game scene, but an ability—the ability for everyone to create and experience their ideal world/dream.

Imagine, this ability would allow you to freely:

• Define and experience your "perfect world," living a life you've always dreamed of. Perhaps you'll discover that what you have now is the best, or you'll find renewed courage to face reality.
• Create a safe, tranquil private space without needing to purchase any real estate, giving your mind and body a haven to recharge.
• Design experimental scenarios for gamified education, allowing your students to learn through immersive experiences.
• Simulate potential future situations, improving your ability to cope and achieve true "risk management" rather than mere theoretical exercises.
• Reshape past regrets, explore "what if" possibilities, and see if "if you could really do it over" you might be able to "forgive yourself."

Every Coin Has Two Sides, And So Does Everything Else

Certainly, nothing in this world exists with only benefits and no drawbacks. Delivering such capabilities to every individual/entity/corporation/nation inevitably comes with potential risks and problems.

First, there's the issue of the psychological impact of virtual worlds.

Many psychological and social studies have pointed out a common problem: virtual worlds can provide new ways and possibilities for "self-expression and exploration" through freely chosen identity attributes or appearances, but they can also lead to confusion or conflict between "real identity" and "virtual identity." If this is true for "virtual worlds," it's even more so for the more realistic "simulated worlds" created by the "matrix." We don't want to see stories like those in the movie "Inception" happen: where anyone mistakes dreams for reality and reality for dreams, leading to tragedy. To prevent this from happening, we would surely spare no effort.

Another more prominent issue is privacy and morality.

Regarding morality, although we practice self-discipline and advocate virtue, we must admit that ethics in the broader sense of the real world are enforced through state actions (law enforcement agencies) and laws. However, the virtual world itself does not have these agencies and boundaries, especially when individuals are the rule-makers.

As for privacy, regardless of who creates this virtual world, privacy issues are paramount whenever there's interaction involving data created by humans or data representing/interpretable by humans. We advocate freedom and hope to give people more choices, but we don't believe that absolute freedom without boundaries is praiseworthy. Pure freedom would only fill this world with violence and despair.

Moreover, anything created by AI models inevitably raises concerns about AI panic.v

Technology is always driving changes in the world, and new technologies that achieve commercial scale are indeed constantly changing the world. But we're not afraid of this. History has proven that we humans are far wiser and more resilient than we think. Even though it might sound pessimistic coming from AI practitioners, we must admit that despite AI doing many amazing things and potentially doing much more, its emerging "intelligence" under scaling laws is not equivalent to "intelligence" in the physical and chemical senses.

We're not so arrogant as to think we can change the world, but we do somewhat stubbornly believe that what we create/develop must aim to make people's lives better/feel better. We do not want to see our creations bring catastrophic impacts to human society.

We are not indifferent to these potential problems. We are actively exploring solutions and welcome global partners to join us in addressing these challenges. We firmly believe that collective wisdom will eventually find a balance.

Indeed, discussing potential risks without considering practical implementation paths may seem like undue worrying.

Although even our research on Terra can attest that our dreams can become goals, achieving them is not easy. If we compare our current starting point to standing on solid ground, then our goal is like a towering mountain in the distance. Unlike ordinary virtual worlds or metaverse concepts, we are pursuing a system that can truly simulate the complexity and randomness of the real world. This requires breakthrough AI technologies, especially in the fields of world models and generative AI.

As mentioned earlier, the current Terra is far from enough. We might need to develop until Terra-X, or even a completely new model architecture. This requires massive financial investment and vast amounts of data. As pragmatic idealists, we believe it's naive and irresponsible to simply expect others to fund our ideals.

Ways to Reflect Ideals in Reality

To realize "our Matrix," we must find a self-reliant, down-to-earth path to realization.

Based on this understanding, we started our current entrepreneurial project. On the surface, we now focus on synthetic data and AI model development, which may seem far from our ultimate goal. However, these projects are actually the necessary paths to our dream. They are like the bridges or spacecraft we are building to cross the vast chasm to Utopia.

Through these projects, we are not only accumulating key technologies but also building a dual flywheel of data and funding. Each breakthrough, each step forward, brings us closer to our goal. We believe that this pragmatic approach will support our long-term dreams and, at the same time, make substantial contributions to the entire AI industry.

At JoinAI, we are not merely chasing a distant dream. We are step by step turning our goal into reality.From the research on synthetic data to the development of the Terra model, each advancement is a crucial milestone. We know the road ahead is still long, but the solid step has been taken.

We invite everyone with a dream to join our ranks.

Whether you are a technical expert, artist, philosopher, or just a dreamer, you can find your place in this grand blueprint. We need talents from various backgrounds to explore the boundaries of technology, tackle ethical challenges, design interaction methods for the future, and improve people's lives through technology and products.

We also call on policymakers and ethicists to join this discussion. In this rapidly developing field, we need forward-looking policies and ethical frameworks to guide the healthy development of technology.

For potential investors and partners, what we offer is not just a business opportunity, but also the possibility of shaping the future. We are pioneering a brand new technological field, whose potential is not limited to synthetic data, even far surpassing traditional virtual reality or gaming industries.

JoinAI believes that with everyone's efforts, humanity can surely create a better world, a world where everyone can find their own Utopia. This is not just a dream, it is the goal that we are actively working to achieve.

Let's work together, using better data and better AI, to create a more ideal world for everyone, forever.