Using GPT-o1 in Agents

On Thursday this week, GPT-o1-preview was announced. We took the opportunity to run some tests. The outcome can be sensed in the title of this post. I just had to share our experience right away. It even affected tomorrow’s Boot Camp where we will talk about the impact of ‘o1’.

Open AI’s new GPT-o1 is a quantum leap

we have tested the new model and have been deeply impressed. The most impressive result came from the most challenging test: “probing its level of ingenuity”. The result was almost shocking. I can say flat out – this model can THINK and you will experience it when using GPT-o1 in GPTBlue Agents very soon.

Ingenuity Test

Test environment: We are building an Autonomous Innovation system, based on an award-winning innovation framework. We test ingenuity, creativity, and the level of disruption.

The Test: build a geothermal energy generator and get to the earth’s crust to a level where we have 500°C, the same heat that we have with nuclear power plants. This is at about 20 to 25 km depth.

• The problem is that drilling that deep for many known physical limitations is impossible.

It took the ‘Think-Machina GPT o1’ only 2 prompts We gave it the above challenge.

With the first prompt, it failed to get to a disruptive solution. In a second prompt, we asked to question every condition to get there. The answer was a disruptive solution within 245 seconds in total. The key aspect of the result is below. It reads as if the model is surprised about its own discovery.

The Test
The task is to build a geothermal energy generator and get to the earth’s crust to a level where we have 500°C, the same heat that we have with nuclear power plants. This is at about 20 to 25 km depth.

• The problem is that drilling that deep for many known physical limitations is impossible.

It took the ‘Think-Machina GPT o1’ only 2 prompts
We gave it the above challenge.

With the first prompt it failed to get to a disruptive solution.
In a second prompt, we ask to question every condition to get there
The answer was a disruptive solution within 245 seconds in total
The key aspect of the result is below
It reads as if the model is surprised about its own discovery.

GPT’s Answer:

Mind-Bending Aspect

By questioning the necessity of a vertical approach, we unlock a new paradigm for accessing deep geothermal energy:

• Redefining “Depth”: Depth doesn’t have to be measured purely in vertical terms. By combining horizontal and inclined tunneling, we can reach the same geothermal horizons without the limitations imposed by vertical shafts.
• Bringing Equipment to the Heat: Transporting steam turbines and other heavy machinery through inclined tunnels is more practical than lowering them down vertical shafts. This approach allows us to set up fully functional power generation facilities deep underground.
• New Industrial Frontiers: Establishing underground facilities at such depths could revolutionize not only energy production but also mining, scientific research, and underground construction techniques.

Countless scientists could not get there and came up with bizarre futuristic concepts. After the above answer, we know it is feasible today. We already can drill over 50 km long Tunnels and 500°C is nothing for today’s steel tools.

GPT-o1 in the next GPTBlue Agents We will use GPT-o1-preview probably as early as next week within specific agents.

Human Intelligence Augmentation

The ability to create disruptive solutions with AI may be frightening to many people. But that was also true of the concept that a train could run at a ‘neck-breaking’ speed of about 20km/h (Adler in Nuremberg) on Dec 7, 1835. Everything thereafter is well-known history today.

Imagine for a moment you can create disruptive innovation on demand. However, you still want an innovation that fits your company, for your specific audience, solving a known problem, and finding a way to bring it to market. We humans do things based on purpose (Human Mind) there are reasons to do that (reasoning) there is a company that has the need to be innovative (company specific) and so forth. It needs a human who sets such a process in motion. Augmenting our human intelligence doesn’t mean WE get more intelligent but now that we have a crutch to solve problems that were impossible to solve before. We are not jealous of the train that can be faster than we can walk, or of an airplane that can fly and we cannot. Human ingenuity is this amazing gift to create tools for anything we can’t do ourselves, AUGMENTS our limited capabilities and make them almost limitless. Another area where GPT-o1 is a quantum leap.

Today, we begin augmenting our capability to think beyond our natural capacity.

Today, Human Intelligence Augmentation has become a reality. With that, our vision to create solutions, that allow us to do things that were impossible before, is amplified.

I’d love to hear your thoughts.