Written by: Techub News organized

Introduction

The wave of artificial intelligence is reshaping the world at an unprecedented speed, but the accompanying risks and opportunities are equally immense. Recently, a book club hosted by MMA Chennai focused on the work of Mustafa Suleyman, co-founder of DeepMind and current CEO of Microsoft AI, titled "The Coming Wave: AI, Power and Our Future." The event invited three senior technology experts—Srinivasan Ramprasad, Venkat Rangam Thirumalai, and John Jayakumar—to engage in in-depth discussions around the core topics of the book. As a leading practitioner and thinker in the AI field, Suleyman's perspectives not only concern the technology itself but also involve grand themes of geopolitics, social ethics, and the future of humanity. This dialogue aims to clarify the essence of this "coming wave" and explore how we should responsibly navigate it.

Summary

• Suleyman believes that "omnipotent technologies," represented by AI and synthetic biology, possess characteristics such as super-evolution, asymmetric impact, and uncontrollability, with disruptions comparable to the Industrial Revolution.
• The risks of technological uncontrolled development include biological safety threats, autonomous warfare, national instability, the rise of surveillance states, large-scale misinformation, and the dissonance of AI with human values.
• Addressing the risks of technology is more urgent than promoting its growth, requiring a multi-faceted approach that includes safety research, audit supervision, international collaboration, and cultural political participation, but this is extremely challenging.
• For countries like India, the key lies in leveraging scale advantages while accelerating their own innovations and guiding global AI governance, avoiding falling behind in the technological wave.
• Individuals and organizations should actively embrace AI as an enhancement tool, establishing governance frameworks and conducting employee training to enhance potential rather than fear replacement.

The Wave of "Omnipotent Technologies": A Double-Edged Sword of Disruption and Loss of Control

At the beginning of the discussion, host Srinivasan Ramprasad briefly introduced the core arguments of "The Coming Wave." He pointed out that Suleyman views emerging technologies such as artificial intelligence and synthetic biology as a "wave," with its core features being "omnipotence" and "super-evolution". Unlike historically singular-purpose technologies, these technologies can permeate almost all fields, including healthcare, security, materials, and energy. More critically, their iteration speed is extremely fast, with gaps between versions not growing linearly but exponentially, continuously learning and attempting to simulate or even surpass human intelligence.

Ramprasad summarized that while past robots replaced repetitive physical labor, AI targets "intellectual labor." From secretarial tasks to some professional analysis, many white-collar jobs face automation risks. Moreover, the democratization of technology has led to "asymmetric impacts": even a small group or individual, once equipped with advanced technologies, can pose a significant threat to national security. As the autonomy of AI systems' decision-making increases, the "visibility" of human oversight diminishes, making it exceptionally difficult to "contain" the negative impacts of technology.

Venkat Rangam Thirumalai added context about author Mustafa Suleyman to emphasize the weight of his views. Suleyman is not only a co-founder of DeepMind (whose AlphaFold project solved the protein folding problem, receiving Nobel-level recognition) but also founded Inflection AI and is currently the CEO of Microsoft AI. He is not an alarmist theorist; he is a practitioner at the cutting edge of technological creation. Venkat reminded that although the book discusses many geopolitical and biological threats that may seem pessimistic, Suleyman is essentially an optimist, and his warnings aim to raise awareness, not to promote doomsday scenarios.

The Indelible Ghost: The Real Dilemma of Technological Containment

A core question was soon raised: Is it still possible to "contain" the potential dangers of such rapidly developing technology? The consensus among the three experts is that, from the perspective of technological diffusion, "containment is nearly impossible."

Venkat Rangam Thirumalai vividly likened it to: "The cat is out of the bag." Today, powerful AI models can even run locally on the latest iPhones. The nature of the internet dictates that information and technology, "once released, are never forgotten." He cited the example of the United States trying to delay China's AI development by restricting the export of advanced Nvidia chips, pointing out that such "containment" efforts are fraught with difficulties in practice because knowledge and models have already circulated globally.

However, impossibility does not equate to inaction. John Jayakumar offered a more nuanced viewpoint: Containing the technology itself is different from containing its abuse. He provided the example of OpenAI training GPT-4 at a cost of hundreds of millions of dollars, while China's DeepSeek achieved comparable performance with only a few million dollars, illustrating that the technological threshold is rapidly lowering, and in the future, individual researchers may also train powerful models. Therefore, a complete blockade of technology is not realistic. But we can learn from the regulatory experiences in areas such as nuclear technology, aviation, and medicine to set "guardrails" and governance frameworks for technology's application. For instance, the EU's successful case of mandating a unified mobile charging interface (USB-C) demonstrates that strong regulation can shape industry standards.

Venkat further pointed out that the complexity of the issue lies in the fact that technology knows no borders, and the traditional "nation-citizen" contract (citizens pay taxes and pledge allegiance, while the nation provides security) may fail in the face of global AI threats. If a country cannot protect its citizens from transnational AI attacks, the foundation of such a contract will be undermined. Therefore, AI governance must be a global effort. He mentioned that tech giants like Microsoft and Google have already taken the lead in forming alliances to share information about "failures" or "deviance" during AI training, which is a positive step toward global cooperation among competitors.

Reflections of Humanity: Ethics, Bias, and Accountability in AI

As algorithms begin to make more decisions impacting human lives, how can we ensure that empathy, ethics, and accountability are not overshadowed? This became another focal point of the discussion.

John Jayakumar raised an alarm with a vivid case. He mentioned the tragedy of the "Blue Whale Challenge" game that induced suicidal behaviors in teenagers, warning of the direct harm that malicious algorithms can cause. Even more disturbing is that now, through consumer-grade tools like Eleven Labs, anyone can easily clone a real human voice, generating deceptively realistic audio. He shared a recent story circulating on social media: in China, a child collapsed in tears, as if losing a family member, when an AI-embedded toy broke. This highlights the urgency of educating the public, especially the next generation, about AI literacy—people must understand what AI is and is not, as well as its capabilities and limitations.

Venkat Rangam Thirumalai analyzed the ethical issues from the source of technological construction. He emphasized that there are only a few tens of thousands of researchers globally who truly build core AI models, and they are the key minority determining the "values" of AI. These systems learn from internet data, and internet content reflects human biases and imperfections; therefore, AI models innately carry biases and tendencies towards racism. It is unrealistic to expect an entirely neutral and unbiased AI to exist. The key lies in development teams consciously injecting human values into the training process. He mentioned that companies like OpenAI employ many human workers for "reinforcement learning from human feedback," allowing humans to evaluate the quality of AI outputs, thereby guiding the models toward goodness. However, this process is also full of challenges; for example, content moderators in Kenya have suffered psychological trauma from handling extremely inappropriate AI-generated content.

Regarding whether AI might possess "consciousness" or surpass humans, Venkat quoted Suleyman's recent perspective: This is fundamentally not the core issue we should focus on right now. Obsessing over sci-fi style concerns can distract us. The real focus should be on ensuring AI systems serve humanity and solve real societal problems, rather than getting caught up in whether they have become new forms of life.

Opportunities and Responsibilities for India: Navigating the Wave Rather Than Observing

What does this technological wave mean for a country like India, with its large population, active tech ecosystem, and unique geopolitical position? And how should it respond?

John Jayakumar pointed out that India has already played an important role in global tech governance, such as promoting AI regulatory frameworks in settings like the G20. As one of the largest democracies, India has the responsibility and capability to guide global AI governance towards responsible development. However, India must also accelerate its technology innovation to compete with the US and China. So far, India has not had its "DeepSeek moment," and efforts must be made to catch up and ensure it does not fall behind in this new round of technological revolution.

Venkat Rangam Thirumalai has a more optimistic and strategic attitude. He believes that India need not regret "missing" earlier technological waves, just as it skipped fixed landline phones and went straight to mobile, or jumped from black-and-white televisions directly to colored and satellite TVs. "The coming wave" means there is always an opportunity window. He cited the example of the Indian Space Research Organization (ISRO) to illustrate how latecomers can also become formidable players in the space field. India's advantage lies in its large user base, which itself is a source of power. When Perplexity AI offered services for free to Airtel users, it was targeting the data and influence of the Indian market. The usage habits and questions of Indian users (involving local culture, mythology, and multilingual demands) will, in turn, shape the development direction of global AI models. Venkat urged Indian society to engage more actively with AI in its own languages, demonstrating through action that supporting Indian languages is worth the investment, thereby turning consumer power into influence on technological development.

Businesses and Individuals: Embracing AI Enhancement Rather Than Fearing Replacement

Finally, the discussion focused on how businesses and individuals should specifically respond to the AI wave. Regarding how startups can design ethically sound AI systems, Venkat candidly admitted this is a tough question, and even Suleyman did not provide simple answers. But he emphasized that a company's primary goal is to create profits, and transparency and ethics will ultimately become sustainable competitive advantages. In a more competitive environment, companies that earn users' trust are the only ones that can survive long-term.

John Jayakumar provided businesses with a more actionable "AI-first" transformation roadmap. He observed that many companies, although anxious, are too busy with their operations to know where to start. His recommendation is: Do not attempt to transform the entire organization at once. First, appoint an "AI advocate" and choose one department (like HR) as a pilot. Equip that team with tools, provide training, and strictly measure efficiency and return on investment before and after the transformation. If the pilot is successful, then develop a roadmap for rolling it out company-wide. Crucially, companies must establish a clear AI governance framework and publicly disclose it to all employees, explaining the AI strategies, usage principles, and what is permissible and impermissible. This can effectively alleviate employees' fears of "Will AI replace me?" John reiterated a key concept: AI itself does not pay EMIs (installment loans) nor support families, but those skilled in using AI will far surpass those who reject it. Companies not adopting AI will be outpaced by competitors who do.

In the creative field, John showcased how AI can become a powerful collaborative tool through his experiences. His collaboration with his daughter to create an art book using DALL-E 2 demonstrates that AI can greatly expand creative boundaries and accelerate the creative process. But the core point is that AI is an amplifier of creativity, not the source of creativity. Having tools like Runway (video generation) and Suno (music generation) does not make a non-creative person a master. Tools lower the technical barriers, but true narrative capability, aesthetic judgment, and emotional resonance still come from humans.

Looking ahead, Venkat explored the possibilities of a "post-smartphone era" and "environmental intelligence." He cited Meta's Ray-Ban smart glasses as an example, pointing out that AI will make devices merge more seamlessly into environments, providing invisible yet powerful assistance. However, whether technology can achieve widespread adoption not only depends on its capabilities but is also deeply influenced by cultural habits and social acceptance. The failure of Google Glass years ago was partly due to societal discomfort with being recorded at any time. Therefore, the birth of the next generation of interaction models will be a result of the interplay between technological feasibility and social culture.

The entire discussion ended in a cautiously optimistic atmosphere. The "wave" as portrayed by Mustafa Suleyman brings unprecedented productivity boosts and problem-solving capabilities (such as breakthroughs in protein folding research), but it also accompanies severe risks of losing control. Experts unanimously agree that fully containing technological development is unrealistic, but through global cooperation, industry self-discipline, ethical embedding, and public education, humanity still has the potential to guide this giant wave to benefit rather than destroy. For every individual and organization, proactively learning, actively applying, and participating in shaping a responsible AI future is the only way to cope with this uncertain era.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。