AI on a shoestring budget

AI on a Shoestring Budget

AI on a shoestring budget.

It can be done!!

AI takes several forms and one of the most critical aspects is what you plan to accomplish while considering implementing AI.

Certainly, if you want to implement an AI framework into your entire business that solves many problems with little human assistance it will come with a commensurate price tag.  But, if you start small, work on a specific problem and get an early win you could have an AI function for a very modest investment.  Plus, you’d have the beginnings of an AI capability that you can grow.

So, what are your options?

Big budget AI

You can start using AI in your business tomorrow, thanks to many of the big cloud platform companies.  The likes of IBM, Azure, AWS and Salesforce all offer ‘ready to eat’ AI that can be as simple as plug and play:

  • IBM Watson
  • Azure AI
  • AWS AI Services
  • Saleforce Einstein
  • and many more

As with all out-of-the-box solutions, they were created with a one-size-fits-all mindset.  And while they can get you started, some customisation may be required.  That’s where the big budget comes in, As Ford said, “you can have any colour as long as it’s black”.  Things may start to get costly when you start to customise the AI to suit what you need.

AI on a shoestring budget

Many thanks to the open source community.  Pretty much any problem you try to solve has already been solved in one form or another and the algorithms are sitting on a public repository somewhere.  The challenge with this approach is:

  1. Finding the right code to solve your problem
  2. Building a version for your own use
  3. Testing and verifying the results
  4. Putting it into production

If much of this sounds way too techie for you, then you’re probably right.
A lot of the AI solution is early incarnations of code that don’t necessarily have a pretty user interface or come with a handy “install” file or plugin.

You will generally need technical assistance with this approach.

And that isn’t always expensive and fraught with uncertainty.  Provided you have a clear view of what you want to achieve and what your budget is, you will most likely reach a pretty solid outcome.

Start on your terms

Whatever your budget, focus on your outcomes as a priority.  Once projects get underway, it’s easy to get lost in the tech stack or the build, particularly because they are often fresh and shiny like these AI projects.

Think outside the box when it comes to the algorithms too.  There are so many different methods around that can solve similar problems and can be a bit overwhelming.  So keep an open mind and try a variety of approaches.

In Summary

AI can be very expensive but only if you let it.  Like any project you embark on, you will be able to execute an AI solution if well planned, budgeted and managed, which will provide you with the results you need at a price you can afford.

After all, it’s value for money we need and cheap isn’t always the right answer.