AI & Automation development – Common Myths

Having worked in this industry for a number of years, attended many AI & Automation conferences and spoken with numerous businesses there are a few common misconceptions that crop up again and again. This post takes a quick look at a few of the most common myths and hopefully dispels some of them.

Myth 1: AI & Automation is expensive

Now don’t get me wrong it can be. Operating massive and complex deep learning models, the kind that drive technologies such as chat GPT, is resource intensive and expensive to develop. However I have yet to meet a business team that couldn’t benefit from AI & Automation adoption to improve their decision making or business processes. In 90% of these use cases delivery can be very economical and even self fund in cost saving terms very quickly. For example by utilising pay as you play “AutoML” offerings by the likes of Google or Amazon or scripting Python solutions for simple day to day office processes. For example automating data extraction, processing and population to reporting systems.

If your teams spend endless hours per week in excel completing the same tasks, that is the perfect example of an “easy win” automation solution that can generate significant human time saving for relatively low cost.

Myth 2: AI is only for big enterpise with big teams and big budgets

Connected to the cost angle is a commonly held belief that you require big data science teams to operate and develop AI and automation solutions. The truth is that with ever easier access to solutions and the right delivery partners, AI and Automation are now available to everyone. Whether you are looking to manage your admin as a sole trader, an SME with excel heavy day to day processes, or an enterprise function with a limited budget but an unlimited to do list finding the right partner can help with all of these things.

Creating and productionalising automated solutions that lift your teams out of day to day drudgery and free them to focus on the real human value they can generate is one of the core reasons why we exist.

Myth 3: IT should own AI

AI and automation are tech so should belong to IT, right?? Well certainly this is a common approach and certainly IT teams are essential partners to maintain and manage the systems required for data and productionalised solutions. However, sole ownership, especially of strategy by IT is often a huge blocker to progress. IT teams are typically incentivised and targeted directly on the wrong things for driving innovation. It’s their job to make sure systems run reliably, technical resources are allocated appropriately and risk is minimised. Innovation often requires exploration and failure whilst challenging existing systems.

We recommend cross functional ownership delivered in an agile way, led by the owner of the business problem and including IT as the systems enabler. It’s the ongoing iterative delivery partnership between the “Why” (business owner) with the “What” (data/dev specialists) and the “How”(data/dev. specialists and IT) that typically delivers a successful solution. This way you avoid the all to common experience of specifying a solution to IT you think you need, enduring a lengthy procurement or development process, before finding out that the expensive solution delivered six months later is not quite what you need.

Focus your IT resources on what they are good at, keeping the ship’s engine running effectively not deciding where it goes.

Similar Posts