We hear a lot about it wherever we look. This is leading many to believe it is nothing but hype.
Sadly, this is likely to hurt a lot of people as they will ignore what is taking place.
There is a rule in life: you either adapt to technology or get run over by it. This is a lesson from history. In fact, there is a name given to this, Luddites.
AI is the fastest moving, most powerful technology we ever saw. Some are claiming things are slowing down as a "wall" is being hit. This is not the case.
Sam Altman, the CEO of OpenAi, actually stated the exact opposite.
Image generated by Ideogram
AI Will Be The Only Game In Town
Altman came out stating that OpenAI knows how to build AGI. This will happen over the next 12 months.
/@taskmaster4450le/re-taskmaster4450le-2nehqmdym?referral=taskmaster4450le
Here we have an alignment between him and Elon Musk. The latter previously targeted 2025 for the time when AGI is achieved.
This means we are in for a massive shift for humanity. To start, it is likely to set off a wave of job loss, likely starting in 2026. We are already seeing the capabilities of AI agents expanding rapidly.
More importantly, we need to consider the concepts we are presently dealing with.
What we are basically looking at is generating units of cognition. Think about that (pun intended). That is the unit of output.
Which brings us to the next concept: AI factories.
This is something we will explore in greater detail but basically data centers are being transformed. No longer are they predominantly storage facilities. The old image could be one of a glorified filing warehouse, but digital.
People basically "entered" the data center to either add more "files" (data) or to retrieve. The software was designed to aid in this process, helping to provide people with the information they were seeking.
Generative AI has changed this.
Instead of the data center simply spitting out what was input, it has a product. This output is no different than what humans are presently doing.
For example, the units of cognition could be realized as legal contracts or software code for engineers.
Ultimately, we are going to reach a point where these factories will be able to do all human services (at least in the non-physical realm).
AGI
What is Artificial General Intelligence?
While the answers can vary, this is what Venice.ai came up with:
Artificial General Intelligence (AGI) is a hypothetical stage of artificial intelligence (AI) that possesses the ability to understand, learn, and apply knowledge across a wide range of tasks, similar to human intelligence. In other words, AGI would be capable of performing any intellectual task that a human can, and it would be able to learn and improve on its own without being explicitly programmed for specific tasks.
AGI is often considered the "holy grail" of AI research, as it would enable machines to learn, reason, and apply knowledge in a general and flexible way, rather than being limited to specific domains or tasks. If achieved, AGI would likely revolutionize many areas of society, including healthcare, education, finance, and more.
Some of the key characteristics of AGI include:
- Reasoning and problem-solving: AGI would be able to reason and solve complex problems using logic, inference, and other cognitive processes.
- Learning and adaptation: AGI would be able to learn from experience, adapt to new situations, and improve its performance over time.
- Knowledge representation: AGI would be able to represent and manipulate knowledge in a flexible and abstract way.
- Natural language understanding: AGI would be able to understand and generate human language, including nuances, idioms, and context.
- Common sense: AGI would possess a deep understanding of the world, including common sense, and be able to apply this understanding to real-world situations.
This is the basic parameters that most tend to agree with. Essentially, we are looking at the ability to do cognitive tasks better than almost every human in all fields. There will be outliers, such as in the field of physics or mathematics where the best people will still be better than AI. However, the number, as a percentage of the whole, will be minimal.
We are to the point where, in my view, all else should be secondary. There are articles daily about how companies need to get on board with this. Those that are late to the game will not be in it long. Here we see tremendous opportunity for startups and smaller entities, due to agility, to step in and gain market share.
It is also going to affect everything. All aspects of life are going to see a major shift. Once again, it is best to remind people that ChatGPT3 was released at the end of November, 2022. Think about how far we came in that time. We are now just approaching the two year mark.
One of my forecasts is the next generation of LLMs will far exceed the capabilities we see today. That means, before the end of 2025, we will see at least one new iteration (some might have 2). The impact is going to be enhanced by the development of AI agents.
And much of this is going to start rolling out of the AI factories that Big Tech is building.
AI is going to be the only game in town.