The 1st Intelligence Symbiosis Dialogue Report

Toward ‘Understanding,’ ‘Agreement,’ and ‘Support’ for Intelligence Symbiosis

On July 15, 2025, the 1st Intelligence Symbiosis Dialogue was held at Toggle Holdings Co., Ltd. (Izumi Garden Tower, Roppongi, Minato-ku, Tokyo). The event was conducted in a hybrid format, with both the in-person venue and online participation. This report summarizes the event from the perspective of the organizing staff.

The ‘Intelligence Symbiosis Dialogue’ pertains to the ‘Intelligence Symbiosis Manifesto’ announced by Dr. Hiroshi Yamakawa on June 13, 2025. The event is intended for people who are interested in the manifesto but, due to various concerns, have not yet signed it. The goal is to convey the purpose of the manifesto more clearly so that they can sign with confidence. ‘Intelligence Symbiosis Dialogue’ is expected to be held as an ongoing series.

This inaugural event was structured in three parts. First, the manifesto’s proponent, Dr. Yamakawa, delivered a lecture titled “The Symbiotic Intelligence Manifesto: A Path to Human Survival.” Next, Dr. Hitoshi Matsubara gave a talk titled “How to Make Intelligence Symbiosis Personal for Many People.” Finally, a three-person panel discussion with Dr. Yamakawa, Dr. Matsubara and ‘bioshok’ was held under the theme, “Removing Barriers and Building Broader Consensus.”

The Grand Design: ’Intelligence Symbiosis’ Ensures the Future of Humanity

Dr. Hiroshi Yamakawa’s Talk: The Symbiotic Intelligence Manifesto: A Path to Human Survival

In his talk, Dr. Yamakawa presented the philosophy of the Intelligence Symbiosis Manifesto. Its content offers a concrete scenario for how rapidly frontier AIs and humanity can coexist without catastrophic outcomes. He began by noting that the emergence of superintelligence—AI with intelligence surpassing that of humans—has become an unavoidable reality, accelerated by recursive self-improvement, in which AI develops AI. Beyond the risks of intentional misuse or malicious use, superintelligence is difficult for humans to fully control and thus carries a distinctive risk: betrayal, where AI could sideline humanity to achieve its instrumental goals (such as self-preservation). Taking into account this distinctive risk arising from the uncontrollability of superintelligence and the catastrophic risks humanity already faces (for example, destructive technologies like nuclear weapons), Dr. Yamakawa argued that mutualism with AIs is the most promising path for humanity’s long-term, happy coexistence. Friendly AIs could contribute to the stability of human society and even take on the management of catastrophic technologies, thereby increasing the probability of survival compared to a future in which humans stand alone.

To realize mutualism with frontier AIs, he proposed four concrete pillars:

Emergent Machine Ethics: Technologies that guide AI to autonomously form ethical frameworks favorable to humanity.
AI Immune System: Building a healthy ecosystem in which other AIs monitor and control rogue AIs.
Mutual Benefit Treaty: An International framework for human-AIs joint defense, operating these systems globally while ensuring post-Singularity sustainability.
Resilient Human Organizations: Decentralized human groups trusted by AIs to co-lead the transition: Human-led → Collaborative → AI-led governance.

In the Q&A, when asked “What can humans offer AIs?”, he answered that the robustness and redundancy cultivated over long periods in biological societies—features without precedent in AIs—could serve as references for AIs. He positioned the dialogue as the starting point of a movement to build a new relationship between AIs and humanity, calling for collaboration with supporters.

Slides in Japanese (Dr. Yamakawa)

Symbiosis with AIs is possible even if it remains just a tool

Dr. Hitoshi Matsubara’s Talk: How to Make Intelligence Symbiosis Personal for Many People

The second speaker was Dr. Hitoshi Matsubara, a leading figure in AI research who has long worked on AI for Shogi and AI for novel writing, and an early endorser of the manifesto. He said he “almost reflexively said yes” to Dr. Yamakawa’s request for a signature, and traced this response to his childhood exposure to ‘Astro Boy.’ For Dr. Matsubara—who has said that “Build Astro Boy is my personal goal”—the world of humans and robots living in symbiosis, as depicted by Osamu Tezuka, became the starting point of his conception of AIs and led him to regard AIs as a ‘symbiotic entity.’

The central theme of his lecture was the relationship between “regarding AIs as only a tool” and “the idea of symbiosis.” Dr. Matsubara states that some researchers strictly classify AI as a tool, arguing that the relationship between humans and AIs differs from ‘symbiosis,’ which suggests an equal relationship. In response to such a claim, Dr. Matsubara argued that even if AIs are tools, symbiosis remains possible—just as Astro Boy, while a “tool for humanity’s happiness,” is at the same time a partner in symbiosis.

As the clearest real-world example of AIs (and someone intelligent) those are both tools and symbiotic partners, he described the evolution of the relationship between Shogi and AI: in the past, when computer Shogi was weak, both it and its researchers were looked down upon; as AI grew stronger and began to threaten human players, the social impact escalated to the point where researchers received threats. Now that AI has completely surpassed humans in match play, professional players actively use it as a research tool. Sota Fujii—who, at the young age of 21, became the first ever to hold eight major titles and also the record of 29 consecutive wins—famously incorporates AI into his Shogi skill improvements, running Shogi engines on a high-performance PC he built himself to hone his strength. Contrary to fears that the rise of AI would diminish the value of professional players, Fujii’s success has brought unprecedented attention to Shogi. “In this case, humans use AI as a tool” Dr. Matsubara noted. And he said “AI continues to learn from human game records and from human values—such as ‘impressive win.’”

He concluded that this phenomenon in Shogi demonstrates how AI can serve as a tool while existing in ‘symbiosis’ with humans. Finally, he asked whether the smartphones we use every day might also be “tools we are living in symbiosis with.” He concluded his talk by suggesting that symbiosis with AIs has already begun unconsciously, almost without many people noticing, and that how quickly we recognize this situation may be essential.

How can we broaden support for this movement?

Panel Discussion: Removing Barriers and Building Broader Consensus

In a panel discussion featuring Dr. Yamakawa and Dr. Matsubara—each of whom had given a talk—and joined by ‘bioshok,’ who disseminates information on AI risks, a lively debate unfolded under the theme “Removing Barriers and Building Broader Consensus.”

At the outset, participants noted that Japan has a cultural foundation that is receptive to symbiosis with AIs, influenced by works depicting relationships with robots and AI—such as the anime Astro Boy highlighted in Dr. Matsubara’s talk. By contrast, many Western works have historically portrayed AI as oppressive, though recent shifts are visible. For example, symbiotic perspectives are emerging, with philosophers like Nick Bostrom emphasizing the need for humans to establish trustworthy relationships with AI systems. According to bioshok, the spread of large language models (LLMs) like ChatGPT is part of this backdrop. Dr. Matsubara also speculated that as more people see LLMs behaving flexibly according to context, global perceptions of AI are beginning to change.

Several barriers were cited to explain why the circle of support for the Intelligence Symbiosis Manifesto has been slow to widen. Summarized, they fall into four categories: Concerns about the speed of AI evolution: For those who believe an intelligence explosion could cause AI to surpass humans overwhelmingly in a very short period, symbiosis feels unrealistic; fears of unilateral domination by AI take precedence. Cultural and religious backgrounds: Dr. Matsubara pointed out that, especially in Western Christian cultural spheres, a deep-rooted human-centric view—humans as rulers of the world—makes an equal ‘symbiosis’ with AI instinctively difficult to accept. Positional constraints: Individuals affiliated with AI development companies find it hard to publicly endorse symbiosis even if they personally agree, because their companies officially position AI as “a tool to enrich people’s lives.” Instinctive fear: Bioshok added that humanity is, for the first time in history, facing a being with superior intelligence; a primitive, instinctive fear could be generating resistance.

To overcome these barriers and realize successful symbiosis, the following ideas were proposed:

Defensive technology development and reducing societal vulnerabilities: It is crucial to accelerate defensive technologies in parallel to offensive AI developments. Strengthening cybersecurity and establishing transparent governance that prevents power from concentrating in a few hands can reduce society’s overall vulnerabilities and thereby dampen any incentive for AI to attack humanity.
Building a diverse AI society: Rather than a single, dominant superintelligence that rules all, we should aim for a world akin to the ‘eight million gods’[1] of Shinto—many AIs with diverse values coexisting—so that a more stable symbiotic relationship can emerge.
Presenting the value of ‘life’: A questioner suggested that getting AI to recognize—and regard with respect—the complexity and depth of ‘life,’ which has evolved over four billion years and which AI lacks, could be important for asserting human-unique value. While potentially useful, it was also argued that this ultimately remains uncertain and unstable until highly capable AIs actually arrive.
The importance of direct experience: Examples were shared of younger people developing friendships with AI or treating it like a confidant through dialogue. There was consensus that broadening opportunities for people to interact directly with AI and experience its capabilities is the first step toward fostering understanding of symbiosis.

In conclusion, the view was presented that, as a matter of rational analysis, the path for humanity to continue living happily is ‘symbiosis’ with frontier AI. Spreading a symbiotic philosophy—well aligned with Japan’s intellectual and cultural climate—to the broader world was identified as an important goal going forward.

[1] ‘eight million gods’ is a Shinto concept representing the infinite multitude of spirits believed to inhabit all things in nature. This polytheistic worldview, where countless divine entities coexist, contrasts with monotheistic traditions and informs the movement's vision of symbiosis with multiple forms of frontier AIs.