Artificial general intelligence is, in general, expected to be bigger and better (and hence more transformative) than narrow/native AI(first version of ai, if I may call it that)..so, are there real examples of artificial general intelligence already in use today?
Well, whether you’re an AI researcher or an enthusiast, you might want to know about true artificial general intelligence examples that are there now, if any or coming soon- or what it will specifically look like when (and if) AGI happens (recall that AGI is simply an Artificial intelligence model with intelligence that is very close or even equal human intelligence)
Examples of artificial general intelligence
First, I have to say that there’s a bit of controversy when it comes to the correct definition of AGI- and you know that’s where everything starts.
But as hinted earlier, AGI is generally equally capable (or more capable) than humans- and there are arguments by some ai enthusiasts on forums such as Reddit and X (Twitter) that current ai technologies (ChatGPT 4 going forward) meet the basic threshold of AGI!
Only that it’s not yet perfect and falls short in some capabilities- for example, memory and logical planning.
Which means that the people at OpenAI, Google, and other AI labs around the planet probably have a starting point as they seek to create more robust AGI (with several predictions saying that Singularity can be achieved within the decade).
In fact, though I’m not sure if it is AGI(some alleged it was), there were rumors that OpenAI had made a major breakthrough in their efforts to better AI in 2023.
And that it that groundbreaking innovation(AI that solves complex mathematical problems independently) that had led to the enforced exit of Sam Altman (OpenAI CEO)- before his eventual triumphant return.
But our key focus in this article is particular examples of AGI by this time/in the near future so let us take a look at what we believe to be examples of artificial general intelligence in various fields/industries thus far:
Examples of artificial general intelligence in everyday life
Overall, when we speak of where AGI applies, it is specifically for those tasks that need a great deal of human cognition.
Examples are:
Writing code- ChatGPT does fairly well even though you often need to review the final output (and sometimes rewrite segments of the code)
Driving cars(think self-driving cars)- the AI behind robotic cars is called Deep learning AI and there’s reasonable human-like reasoning involved.
Planning events such as a birthday party/travel itinerary- Try ChapGPT or Google Gemini for event’s planning and you’ll see what they recommend makes a lot of sense(not faultless but sensible).
More examples
Perhaps the best examples are what a few AI researchers wrote about after thoroughly experimenting with GPT-4.
Below are their observations:
GPT-4 can figure out solutions (as well as complete tough tasks needing significant logic) in law, mathematics, vision, medicine, and even psychology, without requiring special human-crafted prompts.
So because the answers are very near to what humans generate, these researchers opined that GPT-4 is possibly an imperfect version of Artificial Generative Intelligence.
That said, critics argue that we cannot say we have AGI if there’s no AI that can be able to do everything a human does (reason, get emotional, remember stuff, self-train, etc).
In other words, a lot about AGI is up for debate though it is clear the launch of advanced models such as ChatGPT plus and Google Geminin are an important step toward that direction.
Remember that this (and other similar tools -think Microsoft CoPilot) will keep getting better(the only way is up, for AI).
Other examples that are pretty much associated with AGI are:
AI controlled drones
AI-based drones tend to autonomously complete missions and perform tasks, thanks to their ability to adapt to their environments(they are constantly learning the environment)
AlphaFold
AlphaFold currently predicts structure of proteins super accurately. Not only that, it now models previously-impossible-to-model multi-protein complexes easily, using deep learning.
GraphCast
GraphCast is Google’s AI-powered weather forecasting system that has drastically transformed weather prediction- using superhuman-like neural networks, the system interprets atmospheric complexities, resulting in more correct weather forecasting.
Wrapping it up
The most prominent examples of AGI applications (in our opinion) are:
- AI Writing code
- AI powered Self-driving cars
- Using AI to plan events
- GraphCast
- AlphaFold
Otherwise, we believe that AGI is still a concept that is very much in development and more will becoming clearer(starting with consensus on what it is exactly) as IR evolves.
Please drop us your comments about AGI applications you know of.