-
@ AI Buzz
2023-08-02 13:42:06Photo by Jason Leung on Unsplash
AI is a broad term that encompasses many different types of systems that can perform tasks that normally require human intelligence. The examples are reasoning, learning, planning, decision-making, and problem-solving.
However, not all AI systems are created equal. In fact, there are different levels of AI, depending on how general or specific their abilities are.
The AI that we see today in our smartphones, search engines, social media platforms, and smart devices is mostly narrow AI, or artificial narrow intelligence (ANI). ANI systems can excel at one or a few specific tasks, such as recognizing faces, playing chess, or translating languages.
They can do these tasks faster and better than humans, but they cannot do anything else. They are not flexible or adaptable to new situations or domains. They are not aware of themselves or their surroundings. They are not intelligent in the same way that humans are.
The AI that we often imagine in science fiction movies and novels is general AI, or artificial general intelligence (AGI). AGI systems can perform any intellectual task that a human can do, such as understanding natural language, reasoning logically, learning from experience, and creating new knowledge.
They can also transfer their skills and knowledge across different domains and contexts. They can think and act like humans, or even surpass them. They may have consciousness, emotions, and goals.
AGI is the ultimate goal of AI research, but it is also the most elusive and controversial one. There is no consensus among researchers on when or how AGI will be achieved, or even if it is possible at all.
Some are optimistic and predict that AGI could be developed by the mid-21st century, while others are skeptical and think that it is a distant or unrealistic goal.
Some are excited and hopeful about the benefits that AGI could bring to humanity, while others are worried and fearful about the risks and challenges that it could pose.
One of the biggest challenges of developing AGI is unsupervised learning, which is a type of machine learning that does not require labeled data but learns from the inherent structure or patterns in the data.
Unsupervised learning could enable a system to learn from any kind of data without human supervision or guidance. It could also enable a system to learn more generalizable and autonomous skills that could be applied to different tasks and situations.
However, unsupervised learning is also a very difficult and active area of research, with many open problems and limitations. Some of the recent advances in unsupervised learning include generative adversarial networks (GANs), which can generate realistic images or texts from random inputs. The other one is self-supervised learning, which can learn useful representations from unlabeled data by solving auxiliary tasks.
Photo by Possessed Photography on Unsplash
Another challenge of developing AGI is ethical and social responsibility.
If we ever create a system that can think and learn like humans, how should we treat it? How should we ensure that it aligns with our values and goals? How should we regulate its use and development?
These are not easy questions to answer, and they require careful consideration and collaboration among researchers, policymakers, industry leaders, civil society groups, and the general public.
We need to establish clear guidelines and principles for the design, development, and governance of AI systems. We need to ensure that AI systems are accountable, ethical, safe, and beneficial for all.
We also need to be aware of the potential dangers and threats that AI systems could pose to humanity if they are not aligned with our ethical values.
One of the worst scenarios that we want to avoid is autonomous weapons: weapons that can select and engage targets without human intervention.
Autonomous weapons could increase the speed, scale, and scope of warfare beyond human control. They could also lower the threshold for armed conflict and violate international humanitarian law.
AI is a remarkable technology that has already changed our world in many ways. But it is also a technology that has not yet reached its full potential or its full implications.
We need to be mindful of the opportunities and challenges that AI presents us with as we move forward in this exciting and uncertain journey.
We need to be curious and creative in exploring the possibilities of AI for enhancing our lives.
We have to be cautious and responsible in ensuring that AI serves our values and respects our dignity.
By：Rudi Widiyanto Link：https://medium.com/predict/how-far-are-we-from-achieving-agi-ad10bd641bb8