Artificial General Intelligence
(AGI)
Description
AI Reimagined: Dreaming of a Truly Thinking Machine
We’ve come a long way in AI. It’s what lets us open our phones with our faces, makes online shopping smarter, talks to virtual assistants, and even drives cars. But all these systems share one thing: they are meant to do some particular thing. They really don’t comprehend or think as humans do. This is where Artificial General Intelligence or AGI comes in.
Artificial General Intelligence
AGI is a term that is buzzing in technology at present; though not a technical term, it is all technology for it is the end purpose of creating a machine which can think, learn, change and use intelligence almost all sort of tasks as we do. But what does AGI mean? And how far are we from such an achievement? And what will this signify for the human future? AGI would exhibit ‘strength’ and ‘flexibility’ compared to the ‘weak’ or ‘narrow’ intelligence in use today and for a specified purpose. Not needing to be retrained for a change of task and improving over time.
ChatGPT and other AI help in writing articles, but they are unaware of what they have written. AGI will. It could engage in meaningful conversation, then spend time really learning a new language or making an engineering plan from scratch all by itself.
What’s the difference between AI and AGI?
Let’s unpack this:
- AI (narrow) — Performs single-purpose task
- AGI (general) — Performs multiple tasks simultaneously
- AI — Being hardwired to learn
- AGI — Can be applied to applications in other areas
- AI — Lack of consciousness
- AGI — Possibly conscious
- AI — Used in applications such as email spam filters, personal assistants, and map services
- AGI — Future AGI applications
AGI should be as calculators are smart in one thing but in mathematics and writing a poem and cooking dinner and discussing politics.
What Desire AGI?
There are many reasons that researchers, technologists, and futurists get excited when they hear the words Artificial General Intelligence, like:
Learning With No Limits
It could be that AGI will have a solution to problems that people haven’t even conceived of yet; to give an example: cures for diseases, or new scientific theories. Machines never sleep, never get tired, and never lose focus. An AGI can work all the time in healthcare, education, research, and new inventions.
Control of Complexity
AGI could afford to control worldwide systems such as meteorological predictions, allocation of resources, and interplanetary travel with exceptional speed and accuracy.
Barriers to AGI
AGI is not just difficult; it is one of the hardest problems in computer science to have ever been conceived. Thus:
- Understand Human Intelligence: We still do not know everything about how our brains work. How can we copy something we cannot explain?
- Common Sense and Emotion: A present-day AI exhibits no gut reaction or emotional depth. Hence, it’s very difficult to teach a machine to be kind, moral, or even funny.
- Safety and Control: How can we guarantee that AGI is doing what’s right when it is above us in intelligence? This is where AGI ethics will be crucial.
Save for a handful, they range from 2040-something to hundreds of years, if ever. ‘Ways that AGI could be far off’, he stated in Report 2018,” as Goodfellow Professor noted. Some are sanguine enough to suggest that the former date is realistic; others contend that it may take centuries or not at all.
AI is a lot smarter these days, sure…but not by much. An ‘understanding of’ rather than an ‘information processing about’ intelligence would have to make massive breakthroughs in neuroscience, cognitive science, ethics, and engineering to construct the real AGI.
Advanced narrow AI like ChatGPT as well as Google DeepMind’s Gemini, not AGI.
The Power of AGI raises a number of anxiety about the morality and the danger it presents:
- Unemployment: What happens to jobs and people’s lives if everything is better done by AGI than people?
- Should AGIs take decisions for people? Who governs them?
- Existential Risk: Some scientists, for instance, Stephen Hawking and Elon Musk, have maintained that if not cautiously approached, AGI could pose a threat to humankind.
Dream or Fate?
Artificial General Intelligence is the dream final destination of AI: a machine that can think, learn, and feel like a person, and maybe even better than one.
It’s got tremendous potential and is highly dangerous. If AGI is developed in an ethical and prudent manner, the net effect could be very positive. However, if it is rushed or mishandled, this could be a repudiation of the human being.
AGI for the moment is just a concept. Some are hoping for it, some fearing, and everybody else confused about it. But one thing is for sure that as technology progresses, so will be the discussion on AGI.