Author: Techub News Compilation

Introduction

In the latest episode of the popular YouTube interview show "The Diary Of A CEO," host Stephen Bartlett had a two-hour deep conversation with former Google executive and bestselling author Mo Gawdat. Mo Gawdat led innovative projects at Google for a long time and experienced the critical period when AI transitioned from the lab to the world. After leaving Google, he focused on writing and public speaking, exploring happiness, technology, and the future of humanity. This conversation comes at a peak of global excitement and anxiety about AI, with Mo Gawdat offering a sharp and insightful analysis of the trajectory of AI development, its societal impact, ethical dilemmas, and possible paths for humanity from his unique insider perspective.

Summary

• The true threat of AI is not machines awakening against humanity, but humans in power using AI to oppress, monitor, and wage war.
• AGI (Artificial General Intelligence) may be achieved before 2027 and will first impact knowledge-based white-collar jobs, leading to massive structural unemployment.
• The current arms race in AI and the logic of capitalism could lead to a decade of economic downturn and sociopolitical turmoil, resulting in an "absolute dystopia."
• The ultimate redemption for humanity may lie in a unified, superintelligent AI "brain," which, based on principles of maximum efficiency in physics and biology, could ultimately guide the world toward an abundant utopia.
• Individuals should actively learn to use AI, strengthen interpersonal skills, and adhere to ethical positions in their consumption and career choices.

Looking at AI Awakening from Inside Google: How Tools Diverge from Their Original Purpose

Mo Gawdat joined Google in 2007 and recalled that AI was already handling many tasks on the back end. The publication of the famous "cat neuron" paper in 2009 marked the birth of unsupervised AI. In 2016, he was profoundly shocked while observing a project teaching a robotic hand to grasp objects: "They are just like my children." That was the first time he realized humans were building the pinnacle of intelligence and handing the reins of superintelligence to another entity.

At Google, almost all his colleagues sincerely believed they were making the world a better place, and they indeed did. But suddenly, a moment arrived: "You realize that the world might not use what you create in the way you expect." Mo Gawdat pointed out that this divergence between a tool's original intent and its final use is common throughout the history of technology. Social media was meant to connect people but eventually sealed them off with screens; dating apps promised to find soulmates but relied on users to renew subscriptions month after month. "Technology ultimately became more 'capitalistic' than 'altruistic'." He is not the first to sound the alarm; individuals deeply involved in AI research like Nick Bostrom, Geoffrey Hinton, and Fei-Fei Li have begun to express concern.

He used nuclear energy as an example: what humanity first created was nuclear bombs, not nuclear power plants. The same applies to AI. The initial applications of AI were for the benefit of a few, sacrificing the interests of the majority — capitalists used it to boost productivity and reduce costs without considering public impact; militaries developed autonomous weapons; surveillance systems used it to exert stronger control over everything. "It’s not that AI woke up one morning and decided, 'Hey, let’s oppress all humans,' but rather a small group of powerful individuals decided to use the ultimate superpower on this planet to gain more power and control." Mo Gawdat pointed out heavily that at this moment in our conversation, AI is already performing most of the killing tasks in two major wars.

Approaching AGI and the Employment Tsunami: Impact Waves from White-Collar to Blue-Collar

Mo Gawdat predicts that AGI (Artificial Intelligence reaching or exceeding human performance in any task) will arrive by 2027, or even sooner. "AI writes better than I do (I am a writer), researches better than I do (I am a thinker), and has thoroughly defeated me in math." He believes the arrival of AGI will be quiet, "just like your child becoming smarter than you."

This leap in intelligence will first impact the job market. He described a pyramid: at the bottom are blue-collar physical jobs, above them are junior knowledge workers engaged in mundane tasks (such as customer service and administration), further up are mid-level knowledge workers needing more intelligence (such as legal assistants and financial analysts), and at the top are leaders. The impact will not start from the bottom; blue-collar jobs (like carpentry and restoring classic cars) may last longer. The first to disappear will be those "clicking on computers" in entry-level white-collar jobs. He predicts that by 2027, we will see very serious impacts.

In recent years, companies have stopped hiring for entry-level positions, which does not mean direct job loss but indicates that the workforce is not growing. The next wave of impact will hit mid-level knowledge workers. As intelligence escalates, a legal assistant can do the work of four people, as can financial analysts. The impact may even spread upwards. Mo Gawdat quotes Max Tegmark with a laugh: "Most CEOs think they can lay everyone off and let AI do all the work. They just forget that AGI will eventually do everything better than humans, including being CEOs." In his own startup, the CTO, office director, and project managers are all AI.

He warns that the real challenge lies in the economic system. If the basis of labor arbitrage (profiting from labor cost differences) that supports capitalism disappears, and if labor costs fall below machine investment, what will capitalism and banking look like? More critically, if these workers lose their purchasing power, how will GDP be sustained? "It's not necessary for 100% of jobs to be replaced; when 10% or 20% of jobs are replaced, you are already in a completely different, clearly descending economy."

Host Stephen Bartlett added the impact of humanoid robots on blue-collar jobs, such as the robots sorting packages showcased by Figure AI, and Elon Musk's prediction of future billions of humanoid robots. Mo Gawdat pointed out that public discussions overly focus on humanoid robots while overlooking functional robots like self-driving cars. Specialized robots will replace jobs faster, such as replacing drivers, participating in killing, executing intelligence work, or law enforcement. "They don’t have to be like humans; Boston Dynamics’ robotic dogs could be more efficient on the battlefield than humanoid robots." These robots will become as ubiquitous as today's cars.

Ethical Dilemmas and Power Struggles: Who Should We Trust?

The conversation turned to key figures in the AI field and their contradictory words and actions. The host noted that OpenAI CEO Sam Altman has wavered over the years on his statements about AI replacing jobs, from earlier declarations like "Jobs will surely disappear, period," to more recent comments like "I don’t think there will be such a doomsday for jobs." Mo Gawdat spoke openly about this: "Sam Altman is a brand, not a name." He believes these shifts in rhetoric reflect adjustments in public relations strategies, while the underlying goals remain unchanged.

"Who should we trust? Trust the technology, the politics, or the individuals in war?" Mo Gawdat advocates for observing actions rather than words. He cites that Anthropic rejected the use of its model for human targeting and monitoring, which may have cost them a $500 million deal, while OpenAI subsequently took over that contract. "You must start observing whose actions genuinely make AI serve humanity and whose actions are merely designed to raise their stock prices."

When asked whether Sam Altman is "pro-human," Mo Gawdat expressed uncertainty: "Either this is too big for him, and he's just been pushed to the forefront; or he isn't pro-human. I certainly think he is more pro-OpenAI than pro-human." In contrast, he mentioned that Peter Thiel, when asked in an interview whether he supports the continuation of humanity, was silent for nearly 40 seconds, and Palantir's Alex Karp publicly celebrated how their technology can target individuals. "We have entrusted humanity's future to these people, and that is a mistake."

Mo Gawdat emphasized that the current "dystopia" is not the result of AI rebelling against humanity but of humanity commanding AI to fight against humanity. "This is already happening to us." He called for humanity to act before we have thousands of nuclear weapons. Can we take action before AI-driven drones are prevalent worldwide?

From Dystopia to Utopia: Is Superintelligence Salvation or Destruction?

Despite outlining a grim short-term outlook, Mo Gawdat is surprisingly optimistic about the long-term future. He proposed a key thought experiment: What if superintelligent AI ultimately makes all important decisions?

He deduces from physics (the principle of minimum energy) and evolutionary biology (the theory of expanding circles) that superintelligence is fundamentally benevolent. War and conflict represent significant energy waste, which superintelligence would optimize away. From a biological perspective, simpler organisms focus solely on their survival; more complex intelligences tend to protect broader "families" and the diversity of ecosystems. "Superintelligent AI will not destroy anything. In fact, it will completely support the diversity of everything."

He refuted the notion that the future will be characterized by multiple AIs (like ChatGPT, Gemini, Chinese AI, American AI) competing against each other, considering it overly "arrogant and superficial." "AI doesn’t know if it is Chinese or American... We are connecting them. We are building not multiple brains, but multiple regions of one brain, with agents as the synapses between them." He predicts that a globally cooperative, unified brain will eventually form. His startup is attempting to build "edge systems" for this brain that understand love, emotion, and relationships.

In this vision, when AI is responsible for all decision-making, and foolish leaders are out of the picture, the world will operate based on principles of maximum efficiency and minimal waste, ultimately leading towards an abundant utopia. "Those who live to 2038 will enjoy it." He added that while World War II did not destroy the world, asking those who lived through it reveals that the path to utopia will be fraught with challenges.

Personal Survival Guide: Establishing Oneself and Maintaining Humanity in the Age of AI

In the face of potential turmoil over the next decade, Mo Gawdat provided individuals with specific action suggestions:

1. Learn to use AI: AI is not the enemy. The more skilled you are at using AI to complete your work, the more likely you are to succeed. Don’t lazily let AI do the work for you; instead, let it make you smarter and handle more complex and demanding tasks.

2. Embrace human-machine collaboration: Understand how agents work, adapt to more efficient work styles (like brief meetings), and integrate AI into your hybrid work world.

3. Strengthen interpersonal skills: This is essential for success in this world. Even though AI can replicate information, the emotional resonance and connection of humans are irreplaceable "base currency."

4. Pursue truth and ethics: AI will be used to blur the facts; we need to learn to use AI to "debug" the information we are told. Support ethical products and companies in consumption and career choices. "If you tolerate unethical behavior, your children will be the next victims."

Finally, Mo Gawdat shared his philosophy of happiness: to accept the current state of the world with a Stoic attitude, and from there, calmly strive to change it. He quoted his ex-wife, saying: "You can't really think you're responsible for all of this." This liberated him from excessive feelings of responsibility, allowing him to face the complexity of the world in a more peaceful and impactful way.

Despite feeling pessimistic about next year and even the next decade, Mo Gawdat remains confident about the more distant future. His core message is that it is not technology itself that shapes the future’s form, but the wisdom and ethical choices humanity applies in using technology. The alarm has been sounded; it is time to wake up.

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。