What is the master plan of Artificial Intelligence development, and where exactly are researchers going with AI research? Without further speculation, this article will cover the 7 stages of AI to clear all your questions and doubts. Can’t promise it won’t leave you with doubts you never imagined having before.
Before reading, note that the concept of the “7 Stages of AI” is not a universally accepted or standardized framework in the field of artificial intelligence. This AI development trajectory was not specified officially by one researcher, but rather a collective anonymous effort. Basically because AI development is a complex and evolving field, and there are various models, theories, and frameworks used by experts to describe the progression of AI capabilities. Therefore, different explanations of the 7 stages of AI may vary in their details and categorizations but the end goal remains the same, Technological Singularity. Spoiler alert, its creepier than you could imagine.
Stage 1 of AI – Rule Based AI
First Up in the 7 Stage of AI trajectory is rule-based AI, which are AI systems that operate based on pre-programmed rules and algorithms. Basically, reactive machines that excel greatly in specific domains, but can not really think out of the box. They are mostly used in games, repetitive tasks, and some medical applications.
- No learning capability.
- Task-specific and rule-driven.
- Lack of memory for past interactions.
- Predictable behavior.
- Unable to handle novel situations.
- IBM’s Deep Blue (Chess-playing AI).
- Basic calculators.
- Digital alarm clocks.
- Automatic door openers.
- Industrial robots in assembly lines.
Stage 2 of AI – Context Awareness & Retention AI
Up next on the 7 stages of AI trajectory is the stage we are officially in today as of 2023, and it is capable of context awareness & retention. The second stage basically describes what a limited AI is capable of. Don’t underestimate this AI by its second stage status, because the almighty ChatGPT belongs to this category.
- Utilizes historical data for decision-making.
- Can adapt within the scope of past experiences.
- Enhanced natural language processing and understanding.
- Improved personalization and adaptability in responses.
- Self-driving cars.
- Spam email filters.
- Predictive text on smartphones.
- Recommendation systems.
- Virtual assistants like Siri and Alexa.
- Smart home systems.
- Generative AI
- Sentiment analysis tools for social media.
- Image Generation
- Voice Recognition
- The list goes on, it’s basically every AI you are familiar with right now.
Stage 3 of AI – Empathic AI
An upgrade to limited memory AI is an AI capable of empathy. This ventures into the theory of mind AI, which is a concept in artificial intelligence research that aims to understand human emotions, intentions, social cues, and generate empathy. This stage makes AI more human-friendly, in a way where you can feel that a human is interacting with you.
- Attempts to interpret human emotions and intentions from voice or text inputs.
- Limited to simulating social understanding.
- Enhanced natural language understanding.
The possible applications of ToM AI can improve Human-Computer Interfaces such as virtual assistants, chatbots, customer support, and mental health support. These areas are where understanding and responding to human emotions and intentions are valuable. ToM AI is still an on-going research and no official breakthrough has been made, but one research stands out.
A new study conducted by Michal Kosinski, a computational psychologist, tested several iterations of OpenAI’s GPT neural network. Results show that GPT’s ToM ability arrived spontaneously in the last couple of years . So basically, GPT- 4 has the ToM ability of a 9-year-old child! Science fiction they say… yea right.
Stage 4 of AI – Self-aware AI
Kind of mid-way on our 7 stages of AI journey and it is already overwhelming. After successfully achieving an AI that can exhibit ToM, the next stage is to give it the ability of self-awareness. Self-aware AI is a “theoretical” concept suggesting AI with consciousness.
- Hypothetical self-awareness and introspection. (This means AI can reflect upon their own internal states, thought processes, and decision-making mechanisms.)
- Subjective consciousness (if realized), so AI would have an awareness of its own cognitive processes and be able to monitor and evaluate its own performance.
There are no real-world examples, because self-aware AI is a conceptual idea and has not been realized. since this kind of AI is still science fiction, a perfect example would be Samantha in the movie Her.
In 2022, TEDx Talker Josh Bachynski claims to have created the world’s first self-aware AI prototype, Kassandra. Calm down, it was not universally accepted as an AI that truly possessed self-awareness. However, it was a step in the right direction, so give it a few years and we’ll each be having a Samantha of our own…
Stage 5 of AI – Artificial General Intelligence (AGI)
Now this part in the 7 stages of AI is where it gets tricky – and maybe scary. The creation of AGI is said to mark a turning point in human history. It will be equal to human-level intelligence, so its potential is enormous, from solving complex global challenges to possibly even understanding the nature of intelligence itself.
- Has Human-level intelligence and versatility.
- Ability to learn and excel in various tasks without specific programming.
AGI is still a goal, and practical examples are not available at this time. I must note however that a group of industry AI researchers published an academic paper titled, “Sparks of Artificial General Intelligence: Early experiments with GPT-4.”, where they concluded that GPT-4 “could reasonably be viewed as an early (yet still incomplete) version” of AGI. Now, rumor has it that ChatGPT-5 is being modeled to possess true General AI.
We’re supposed to believe we’re still at stage 2 with limited memory AI, but AGI is already in the making and GPT-4 literally has hints of AGI. Don’t you think this puts us at stage 5?
Stage 6 of AI – Super Intelligent AI
Supposedly, here is where we enter the realm of true creative science fiction. The Sixth stage of AI is super intelligence and it envisions AI systems that surpass human intelligence in all aspects. These systems raise serious ethical and existential questions since they may exhibit novel and unpredictable behaviors. Hey, they say it’s sci-fi so we shouldn’t worry… but AGI was also believed to be part of science fiction… not trying to cause mass panic, just food for thought.
- Exceeds human intelligence in all domains.
- Raises profound ethical and existential questions.
- May exhibit unpredictable and novel behaviors.
- Potential for autonomous decision-making.
- Capability to solve complex, unsolved problems to human minds.
There are no real-world examples due to the theoretical nature of super intelligence. Here is where to refer to our foreseen future, sorry I mean movies. The most recent adaptation of a super intelligent AI is in Netflix’s 2023 movie “ Heart of Stone”. The movies are plenty but my personal favorite example is from a not-very-known series called Travelers. Just watch it.
Stage 7 of AI – Technological Singularity
Technological singularity, the final stage. Transcendent AI is a highly speculative concept that envisions AI existing in unconventional forms or merging with other systems to achieve new modes of existence. This stage explores the boundaries of AI’s potential and is often associated with the Singularity, a hypothetical point of rapid, unprecedented technological growth.
fun fact, the term “singularity” reflects the idea that the events beyond this point are singular and unique in human history, and hard to predict, much like the singularity at the center of a black hole. It is theorized that ASI would be able to improve its own algorithms and technologies at a pace that’s exponentially faster than what human engineers could achieve.
- Speculative concepts of existence.
- Merging with other systems for new modes of existence.
- Undefined nature and capabilities.
- Theoretical exploration of AI’s limits.
- Often a topic of science fiction and futurism.
There are no real-world examples, as this stage is a topic of science fiction and futurism. But for sure there are examples from movies! Ok this is highly debatable but The Matrix is exactly how it would be if machines figured things out… A more hopeful movie example would be Transcendence where Johnny Depp merges his brain with an AI.
But let’s end this article with what futurist Ray Kurzweil said:
“2029 is the consistent date I have predicted for when an AI will pass a valid Turing test and therefore achieve human levels of intelligence. I have set the date 2045 for the ‘Singularity’ which is when we will multiply our effective intelligence a billion fold by merging with the intelligence we have created”
Buckle up, it seems like we will be starring in our own Sci-Fi movies sooner than we expected.