Guys, have you ever fantasized about one day being able to chat with Jarvis in "Iron Man"? Or like the protagonist in "Her", falling in love with a super-intelligent system? The scenes in these sci-fi movies seem to be getting closer and closer to us!
Recently, OpenAI's CEO Sam ·Altman threw a bombshell: super artificial intelligence, which is often referred to as AGI (general artificial intelligence), may be realized in the next 10 years! As soon as this news came out, it simply made the entire technology circle boil. As a "trendsetter" in the AI world, what does Ultraman's prophecy mean? How far are we from a true "super AI"? Let's take a look at the ins and outs of this hot topic!
First of all, we have to figure out, what is AGI? Why is it so important?
AGI, the full name is Artificial General Intelligence, Chinese called General Artificial Intelligence. To put it simply, it is an AI system that can think, learn, and adapt to various situations just like humans. Unlike our current common specialized AI (e.g., playing chess, recognizing images), AGI can handle any type of intellectual task and may be more efficient than humans.
Imagine what it would be like if AGI was realized. Maybe you wake up in the morning and AGI has already made the best work plan for you based on your schedule, weather forecast, and traffic conditions. When you encounter a problem, AGI can provide you with a variety of innovative solutions like the best partner. Even when you are in a bad mood, it can also act as a psychological counselor to enlighten and solve your worries. Doesn't that sound cool?
However, the impact of AGI goes far beyond that. It has the potential to revolutionize the way we work, the way we live, and even the way society as a whole operates. Some people believe that the emergence of AGI will bring about a new technological revolution and solve many problems faced by mankind; There are also concerns that super AI could get out of control and threaten the existence of humans.
So, why does Ultraman think AGI will be realized in 10 years? Let's take a look at his arguments:
Technological progress is unbelievable
Altman pointed out that AI technology is developing much faster than expected. Take OpenAI's own products as an example, from GPT-3 to GPT-4, the performance improvement is staggering. And this progress seems to be accelerating. Do you remember the sensation when ChatGPT first came out last year? In retrospect, it may have been just a prototype of AGI.
Exponential growth in computing power
Moore's Law tells us that the performance of a computer doubles every 18 months. Although this growth rate has slowed down recently, the development of AI-specific chips is in the ascendant. Companies such as Nvidia, Google, and Tesla are all developing increasingly powerful AI chips. With these "super brains", will the implementation of AGI be far away?
The explosion of data volumes
As you know, data is the "food" of AI. With the popularization of the Internet of Things and 5G technology, the amount of data generated every day is increasing exponentially. This massive amount of data provides rich material for AI learning. Think about what would happen when AI was able to "digest" all the knowledge accumulated throughout human history?
Breakthrough advances in algorithms
Deep learning, reinforcement learning, transfer learning... These lofty-sounding words are actually important algorithms in the field of AI. In recent years, these algorithms have made continuous breakthroughs, making a qualitative leap in AI's learning ability and generalization ability. Altman believes that we are getting closer and closer to simulating the way the human brain works.
Strengthening interdisciplinary collaboration
The development of AI relies not only on computer science, but also on the support of multiple disciplines such as neuroscience, cognitive science, and psychology. With the deep integration of disciplines, our understanding of the nature of intelligence is also deepening. This provides a solid theoretical foundation for the realization of AGI.
After listening to this, do you also start to believe that the arrival of AGI is really just around the corner? But don't get too excited, it's not that simple.
First of all, many AI experts have reservations about Ultraman's prophecy. They believe that while AI technology is indeed advancing at a rapid pace, it is still a long way from true AGI. Why?
The nature of intelligence is not yet clear
While we have been able to create very "smart" AI systems, humans still know very little about the nature of intelligence. We don't even fully understand how our brains work, so how can we easily replicate human-level intelligence?
There are still limits to AI's "understanding".
Existing AI systems, no matter how advanced, are essentially doing pattern recognition and statistical reasoning. They lack true understanding and creativity. For example, ChatGPT can write articles that seem to make sense, but does it really understand what it is saying?
Lack of common sense reasoning skills
Humans do a lot of common-sense reasoning in their daily lives, which is a breeze for us, but a huge challenge for AI. How to make AI systems have "common sense" is a problem that has not been well solved until now.
Ethical and safety issues need to be addressed urgently
As AI becomes more powerful, how can we ensure that they behave in line with human values and ethics? How to prevent super AI from getting out of control? If these issues are not addressed, the implementation of AGI could have disastrous consequences.
Limitations on computing resources
While computing power is constantly improving, the computing resources needed to simulate intelligence at the human brain level may be far greater than we can imagine. Studies have shown that the human brain can have up to 1 trillion calculations per second, and the energy consumption is extremely low. Can our technology really reach this level in the short term?
Seeing this, you may feel a little disappointed. AGI seems so far away from us, isn't Ultraman a little too optimistic?
In fact, no matter when AGI will eventually become a reality, one thing is certain: AI technology is changing our world at an unprecedented rate. While we're discussing this topic, there may be another new AI breakthrough on the way.
So, in the face of such an era of rapid development of AI, how should we ordinary people respond?
Keep your enthusiasm for learning
The development of AI may automate a lot of work, but it will also create new jobs. Only by continuing to learn and improve one's abilities can we be invincible in this era of change.
Develop creativity and emotional intelligence
By far the most difficult human traits for AI to replicate, creativity and emotional intelligence are among the most difficult human traits for AI to replicate. Cultivating these capabilities will allow us to play a greater role in the future of human-robot collaboration.
Focus on AI ethics
With the popularity of AI, each of us may face ethical issues related to AI. Understanding and participating in discussions about AI ethics can help us make more informed choices in this new era.
Embrace change, but stay vigilant
The changes brought about by AI are irreversible, and we should embrace them with an open mind. But at the same time, it is also necessary to be vigilant about the risks that AI may bring, and participate in the discussion of AI governance.
Valuing relationships
In an increasingly digital world, authentic relationships may become even more precious. Don't forget to nurture and maintain your network outside of technology.
Finally, ask yourself: If one day, you really met a super AI, what question would you ask it? Perhaps, starting to think about this now will help us better understand ourselves and understand what makes us unique.
Whether AGI is achieved in 10 or 50 years, one thing is certain: we are living in an era of unprecedented technological change. In this day and age, being curious and passionate about learning is perhaps our best way to survive. After all, the future is here, it's just not evenly distributed yet. Are you ready?