Chandrayaan, humanoid robots, driverless cars—been there, done that.
But are we ready for AGI? Will AGI solve problems or become a problem itself?
Machines with common sense, consciousness, and the ability to think define AGI.
With such systems, will the age-old debate: “Can machines be at same or better state than humans?” finally have an answer?
AGI: What’s all the hype about?
Artificial General Intelligence (AGI) is a theoretical stage in the development of machine learning.
AI which levels up to do literally everything a human can do, maybe even better. 💡
AGI systems will learn new skills in no time. With common sense, general intelligence, adaptability, and problem-solving, AGI will be the G.O.A.T. of ML 🤖🔥
Imagine an AGI system finds a GitHub project, learns how to spin it up, detects errors, fixes them, and even cleans up the code like it’s no big deal.
It’s just like having a coder buddy who is a genius! 👨💻
The AI Face-Off
Consider an AI trained to play chess like a pro. It’ll ace the game, but it can’t do anything else.
That’s your classic ANI (Artificial Narrow Intelligence) — super good at one thing, but that’s it.
Ever heard of the fictional character “The Terminator”?
He can ride a bike, drive a car, or even handle guns. He’s not just smart at one thing, he’s a jack-of-all-trades but still fictional! 🦾
Imagine a system making best decisions in tricky situations, all on its own. That’s our dream with AGI.
Then there’s “Skynet” from “The Terminator”, who gained consciousness and could also outperform humans.
Cool, right? That’s ASI (Artificial Super Intelligence).
BTW, did you know self-driving cars aren’t thinking like humans yet?
They may seem to be AGI but are actually a team of multiple ANI systems.
One detects pedestrians, another controls acceleration.
Right now, ANI is the only AI that’s out there.
Are LLMs AGI in disguise?
LLMs (Large Language Models) are making headlines.
Researchers such as Blase Agüera y Arcas and Peter Norvig say that LLMs, like GPT, Llama, and Claude, have already hit AGI, with “generality” being the key.
However, current LLMs have limitations.
To reach AGI, models need real-time learning, thinking power, and higher compute.
They may also require a modified cognitive architecture.
In a nutshell, LLMs are not AGI. AGI goes far beyond.
Courtesy: Alex Albert @AnthropicAI on Twitter
Fun claim: Anthropic already found AGI.
2 stones made of 3 minerals: Azurite, Goethite, and Iodargyrite (initials A, G, and I).
With a note beside saying, “Do not plug in,” haha—because the real AGI might just blow a fuse!
Is AGI closer than you think?
There has been a heated debate about when we can expect AGI to arrive.
While companies like OpenAI, Google DeepMind, and Anthropic are pushing the boundaries of AI rapidly, AGI is still a dream.
OpenAI recently showcased the o3 frontier model that was able to score highest in ARC-AGI, which is significantly better than the o1 model at similar benchmarks.
Even these reasoning models are good at specific benchmarks but fail to solve problems that young brains could do with relative ease 🧠
That’s why ARC PRIZE remains undefeated.
Courtesy: ARCPRIZE.org
Here is a sample task in ARC PRIZE.
Some say AGI could show up in the next few decades, but it’s far from a sure thing.
Sam Altman, CEO of OpenAI, thinks humans might crack AGI in a “few thousand days.”
But the exact timeline remains uncertain.
AGI may arrive as expected or as a complete surprise.
Let’s hold on to our hats until then 👋
