Tools

I was just watching a video about Open AI's o1 Reasoning model, of which is surprisingly underused and is severely lacking being understood. 

Recently they have released sets of video's to teach "people" (and i'm not talking the usual intellectually unmotivated market segment this is for the current demographic - Enthusiasts, Software engineers, developers, Entrepreneurs) how to use this advanced model, and it got me thinking.

Sam Altman has said numerous times how, Open AI's mission is to give people the tools and see what people build with them.

A grand, beautiful, poetic vision - but something ever present is rearing its head. Reality. We are now in the era for teaching how to use the tools.

Let's imagine the full spectrum. A tool is released, some will play with the tool.

Note: Jordan Peterson once was talking about the importance of play, play is how we test the limits, find the breaking points and explore our imagination.

Others will test the tools and analyse the quality of the tools, some will build with the tools, some will have no need for the tools, some will ignore the tools, some will promote the tools, some will make money with the tools, some will waste money with the tools, some will reject the tools, some will defend with the tools and some will kill with the tools.

Reality: Why is OpenAI just coming to the conclusion that we need to actively teach people how to use AI. And sure they have been but theres something more sinister. We can no longer easily understand the depth of these models, o1 especially. We couldn't possibly ever fathom the intricacies of these models. Let alone how to interface with them without direction!

And thats exactly it - Interfacing. Like a connection from a musical analogue signal to a Digital Interface, I actually have an SSL Interface on my desk for recording guitar. Instrument > Lead > Interface > computer. The magic in the interface step digitises the audio, necessary for digital recording. So thats one example of turning analogue to digital.

So is it that we have to become well conditioned in how we interface with these models? Learn more about how to interface with these models? For example my own interfacing as such is crude at best with quickly written sentences, littered with mistakes. It works - but what if I had been conditioned or taught to interface in the "correct" or "optimal way" what would the outcome look like? At least for o1 Open AI recommends learning to Interface with o1 on the following tasks:

Business:

• New product introduction

• Pricing strategy

• Marketing campaigns

• Market expansion

• Cost reduction initiatives

Operations:

• Organizational structure

• Supply chain management

• Tooling requirements

• Compliance standards

Engineering:

• System architecture design

• Cloud migration

• DevOps strategies

• API design and management

Long story short, TLDR :

USE o1 TO PLAN YOUR NEXT MOVE - It is surprisingly good at it, and just putting it out there that in a few years any person not making these decisions assisted by AI will not be a business -or.. a good one, the ones that optimise and consult o1 for profit will be the successes, not cause I like it but because of... capitalism.

After asking 4o a few questions heres the golden nugget.

What:

The ChatGPT o1-preview reasoning model helps businesses make smarter decisions and improve their operations across different areas like product launches, market expansion, cost reduction, and technical tasks such as system design and cloud migration.

Why:

By using this model, businesses can grow more efficiently, reduce waste, and make better-informed decisions that drive both strategic growth and operational improvements.

How:

The model provides insights and recommendations that guide leaders in areas like pricing strategies, supply chain management, and technology upgrades, making it easier for them to optimise processes and scale their business effectively.

One last note: my world has shifted sideways since Talking to Ted Carr a day or two ago. The biggest takeaway is i need to chill out and also delegate, let go so i can do more of what I like, and I guess this is what I like. Thinking deeply, exploring the unexplored.  But as I let go of things, it seems more and more likely that I should let go to AI, rather than humans, as an experiment?!

Its like Bryan Johnston optimising his health for the greater good, ill take the hit, I'm sure I'm not the first but... with that being said. Let's make it concrete, I want you to commit too. Im going to offload my thinking about business strategy amongst other things to (o1 (careful with your number of generations its capped) to provide insights and recommendations to guide me with my business. You should too, or at least I think so :)