Do you need to understand AI?

Whilst speaking to a prospect yesterday, the realisation dawned on me that not everyone understands AI… Not at a technical level but from an abstract commercial application perspective. This is perhaps not surprising given the media hype, the mountain of new acronyms and the inaccurate simplifications of both the threats posed and opportunities presented.  Perhaps…

Whilst speaking to a prospect yesterday, the realisation dawned on me that not everyone understands AI… Not at a technical level but from an abstract commercial application perspective. This is perhaps not surprising given the media hype, the mountain of new acronyms and the inaccurate simplifications of both the threats posed and opportunities presented. 

Perhaps there is a need for a foundational, hype free webinar, along the lines of “Data Science for Non-Data Scientists and others that need to know their AI from their ML”?

Some of the misunderstandings that I increasingly see (not just yesterday):

– Believing that OpenAI and ChatGPT are AI in its entirety. 

– To use AI you need to write something called a prompt… and you probably need to hire a “Prompt Engineer”

– AI can be used interchangeably with ML depending on which acronym looks more visually pleasing in your document. 

– AI changes everything, for everyone, forever… and other well qualified and quantified assertions regarding the role of AI. 

If I was to create a webinar to try and push back the tide of misunderstanding, as well as provide some sound foundational concepts for the less excitable, where would I start?

– Consider Generative Artificial Intelligence, Analytical Artificial Intelligence and Artificial Intelligent Automation; what are the traits, boundaries and use cases? In other words what can you do with it and which “it” do you mean?

– When we say “the machine learns”, what do we really mean? How and what does it learn? What are the realities, risks and practicalities of the learning process? Actually what is a Model?

– If you had a viable use case… Actually, how would you know you had a viable use case? Wait!! If you can’t spot this then is there not a danger that you might be missing some commercial opportunities?

– OK… so now you know what you need to know, you now what “it” is and you can spot a use case for “it”… But in broad terms how do you deploy “it” and what are some of the headline costs? Do you need more than an OpenAI account and paid access to ChatGPT? (At this point there is not an emoji with its eyes rolled back far enough!) 

Could be interesting? Would you attend (for free)?

Be the first to hear

Subscribe to my email newsletter and be the first to see useful content, news and the occasional offer delivered straight to your inbox.

View more articles
  • I forgot

    It turns out that I am in fact not very good at this diary writing stuff! Who knew. It’s not the actual writing that causes me a problem. I can in fact write for England. But remembering to connect to this site and create a post seems tricky. There are probably ways I could remember…

  • Time for a tidy up

    For some time now I have thought my digital presence has been a bit of a mess and in need of some TLC. This is particularly ironic given I am supposed to know about such stuff! The problem I think is one of abundance! I have social profiles on every platform except Twitter. Then for…

  • Will AI replace Developers?

    Curiously I just “asked” ChaptGPT to write me a Python script to generate random numbers and to include an infinite loop in the code. You would think it could surely generate the correct response to that? After all some people think that AI will replace Developers. But strangely it didn’t reply with “what upper and…