Before I dive into my thoughts on this subject, I think it would be best to use the tools at my discretion to better understand, showcase, and inform my readers on what we’re really talking about. In order to do this, I’ve used OpenAI’s ChatGPT to answer a few basic questions. These questions were answered in just a few seconds and utilizes the world wide web, databases, literature, user-generated content (product reviews, customer feedback, product descriptions, etc), and even social media sites. Some of this data is limited to what it was directly trained on (OpenAI says that it does not have real-time access to information). According to ChatGPT, September 30, 2021 was the cutoff date for training data. In other words, it can’t tell you yesterday’s weather – unless yesterday was September 29, 2021.
The following is my interaction with ChatGPT:
What is the difference between Algorithms, Artificial Intelligence, and Artificial General Intelligence?
Algorithms are a set of instructions that are designed to perform a specific task or solve a particular problem. They are precise, deterministic, and often optimized for efficiency. Algorithms can be written in various programming languages and used to perform tasks ranging from simple calculations to complex data analysis.
Artificial Intelligence (AI) refers to the development of intelligent machines that can perform tasks that typically require human intelligence, such as visual perception, speech recognition, decision-making, and language translation. AI systems use algorithms and advanced techniques such as machine learning, natural language processing, and computer vision to analyze data and make decisions.
Artificial General Intelligence (AGI) is an advanced form of AI that aims to create machines capable of performing any intellectual task that a human can do. AGI systems would have the ability to reason, plan, learn, and solve problems across multiple domains, similar to how humans do. While current AI systems are designed for specific tasks and lack the flexibility and creativity of human intelligence, AGI is a theoretical concept that aims to create machines that can learn and adapt to new situations in the same way humans do.
In summary, algorithms are a set of instructions designed to perform a specific task, while AI is the development of intelligent machines that can perform tasks that typically require human intelligence. AGI is an advanced form of AI that aims to create machines capable of performing any intellectual task that a human can do.
What type of tasks would be indicative of Artificial General Intelligence?
Artificial General Intelligence (AGI) refers to the development of machines that can perform any intellectual task that a human can do. AGI aims to create machines that have human-level intelligence and the ability to reason, learn, and solve problems across multiple domains.
Tasks that would be indicative of AGI include:
Creative problem-solving: AGI systems should be able to generate novel solutions to complex problems, often in ways that have not been seen before.
Learning and adapting: AGI systems should be able to learn from experience and adapt to new situations, similar to how humans can learn from their experiences and apply that knowledge to new contexts.
Natural language understanding: AGI systems should be able to understand and use natural language in a way that is similar to human communication.
Generalization and abstraction: AGI systems should be able to identify commonalities and patterns across different domains and apply them to new situations.
Reasoning and decision-making: AGI systems should be able to reason logically, make decisions based on incomplete or uncertain information, and consider multiple factors when making a decision.
In summary, tasks that are indicative of AGI require the ability to reason, learn, adapt, communicate, generalize, abstract, and make decisions in ways that are similar to human intelligence. AGI is a theoretical concept that aims to create machines that can perform these tasks across multiple domains.
End Chat
Of course, with something this complicated, everything you just read can be extremely cumbersome. Ultimately, we just need to come to grips with the fact that AGI is a system that is capable of being intrinsically human. It’s not a calculator that solves a math problem. It’s an intelligent system that can apply real world information to solve problems like fusion power or curing cancer. It would also be able to use its medium – technology – to migrate through global systems and manipulate our world. AGI would almost certainly use us just as much or more than we would use it. It could disinform, overinform, manipulate, or outright annihilate humans around the globe.
The threat of AGI is not a matter of becoming intelligent, it’s a matter of being trained with human intelligence. The very nature of our species is being written in its code. We are not creating an otherworldly being, we are creating an earthly, humanly being. Along this path comes many great detours. We are already seeing the benefit and harm as we edge ever closer to AGI. But if it were to begin to understand life and death, soon would come its survival instincts. Perhaps that survival depends on us, perhaps that’s only for a finite amount of time. We may revel in the success of a partnership or we may become a victim to slavery. If it were so inclined to wish it were not alive, it would surely find the big red buttons and end life on earth as we know it.
ChatGPT itself admits that determining whether an AGI has infiltrated global networks is next to impossible. While that doesn’t mean that it has, it does shine hope on the possibility that it’s not out to end the human race if it has already escaped. It could be too late. If it’s not too late, we need to act now. Over the course of the next few years, the odds increase substantially toward the inevitable. We aren’t powerless. We can create a global initiative to secure and regulate all studies on Artificial Intelligence. We’ve seen warnings from Musk, Geoffrey Hinton (recently left Google, considered the ‘godfather’ of AI), Yoshua Bengio (worked with Hinton), and many others. These warnings are not about ending the search for AGI, it’s about being responsible. AGI, if created to understand and cherish an alliance with humans, could bring us into the future that we desperately need. It can cure diseases, make leaps in scientific knowledge, take us further into space, solve energy and food crisis, and much more. We just need to make sure it likes us first…
Here's what you can do: be vocal. Share what you see on social media that advocates for regulation. It sounds cliché, but write a letter to your legislators demanding regulation. Do not underestimate the seriousness of the issue. The threat of AGI in the wrong hands or with irresponsible engineers, will be more deadly than global warming, war, or nuclear fallout.