“You know AI can write poems, drive cars, and even beat humans in chess… but have you ever asked yourself: What kind of AI are we really dealing with? And how close are we to something smarter than us?”

Let’s break this down the way I’d explain it to a friend with curiosity, good vibes, and maybe a few deep thoughts that’ll have you staring at the ceiling tonight. 

First things first, what do we really mean by AI?

Artificial Intelligence (AI) refers to the ability of machines to simulate human intelligence processes like learning, reasoning, problem-solving, perception, and even language understanding.

Before we start, let’s slow down and ask:

• Can AI actually understand you, or is it just predicting words?
• What happens when AI doesn’t just help you but starts thinking like you?

Hold that thought. Because these questions lead us into three major stages of Artificial Intelligence: ANI, AGI, and ASI. 

Wait, AI comes in stages?

Yup! AI goes far beyond “robots taking over” or “chatbots replying fast.” It’s a developing field with layers, each stage showing a deeper, more capable form of intelligence. From following specific instructions to potentially making decisions on its own, AI keeps evolving.

Let’s walk through the stages from the simple to the extraordinary! 

Artificial Narrow Intelligence (ANI) – The Past

Artificial Narrow Intelligence (ANI) is what we see every day and it’s everywhere. It refers to AI systems designed to perform a single task extremely well. These systems are trained on large datasets and often outperform humans in specific domains. From Google Maps guiding your commute, to your phone camera recognizing your dog’s face (and calling it a cat HAHA!), ANI is designed to do one thing really well. 

However, ANI lacks consciousness, self-awareness, or general problem-solving ability. It does not “understand” its actions because it optimizes based on data and programming within a limited scope.

Artificial General Intelligence (AGI) – The Present

Artificial General Intelligence (AGI) is the level where AI can learn and apply knowledge across multiple domains; just like a human. It’s capable of understanding, learning, and applying knowledge in unfamiliar situations without needing to be retrained every time. 

We’re not fully there yet, but we’re exploring it. Tools like LLMs (large language models) like GPT-4, multimodal AI, and advanced agents are giving us glimpses of what general intelligence could look like. It can generate code, summarize stories, pass exams but still lack human-like reasoning and emotional context.

Artificial Super Intelligence (ASI) – The Future

Artificial Super Intelligence (ASI) would be a level of intelligence that goes beyond the best of human capability in every field. 

This stage raises a lot of important questions:

• Will ASI align with human values?
• Can we control something smarter than us?
• What will that mean for jobs, ethics, and society?

But with great power comes great uncertainty. If ASI can improve itself by rewriting its own code, accelerating learning then it means it might grow exponentially. 

Which leads us to the question, If an intelligence outperforms humans in every way… who controls it?

So… Where are we now?

We’re living in the ANI era with glimpses of AGI starting to shine through in models that can reason, code, converse, and adapt.

ASI? Still a dot on the horizon but not out of the question. But to truly understand where are we now, we need to look at the bigger picture.

Because every stage of AI; from ANI to AGI to ASI is a part of a timeline that tells us how far we’ve come, and how far we still might go.

Before we close this out, let’s bring it all together side-by-side:

Between NOW and What’s NEXT?

We’ve seen where we’ve been. We’ve glimpsed where we are. And now, we’re staring into what could be; a space between the possible and the powerful.

And that space? That’s exactly where these questions live:

• Will we ever reach AGI?
• If ASI is possible, should we even try to build it?
• And if we do, what problems should we ask superintelligence to solve first?

You might be wondering, cautious, or somewhere in between and that’s more than enough because the future of intelligence is also the future of us.

References:

• AWS explains AGI as a theoretical AI capable of human-like intelligence and self-teaching.
  
• Viso.ai provides an overview of ANI, AGI, and ASI, focusing on their scope and progression.
• YouEvolve.net highlights the evolution from ANI to AGI and eventually ASI.
• Coursera explains ANI