AI artificial intelligence tech

Artificial intelligence: myths vs. reality

Artificial intelligence (Ai) is having a moment right now. 

Google launched their Ai platform to help teams bring Ai-powered projects to market faster. Amazon is using machine learning and Ai to improve personalisation. Facebook launched an internal Ai team within their research department to advance the science and technology of artificial intelligence. That’s not to mention a certain content publishing platform you may have heard of.

Despite all this, the artificial intelligence myths and misconceptions generally persist. We’ve compiled three of our (least) favourite and will hopefully go some way towards reassuring you that the end isn’t nigh.

It’ll cost human jobs

In the same way that combine harvesters replaced hand scythes, technological advancements will forever change the nature of work.

However, the difference between the industrial revolution and the Ai revolution is simply that the jobs that will be replaced with problem-solving tech are still being defined. 

It’s true that many low-skilled jobs may become obsolete in the coming decade. However, as industries evolve, so do the disciplines required to keep them running efficiently. 

Rather than a human replacement, think of Ai as an efficiency enhancer. The inflo.Ai tool, for example, allows you to create an original, optimised blog in a quarter of the time it would take you to do so normally.

Artificial intelligence takes care of all the sifting through information, trawling through SERPs and identifying trends. The user still adds their own dash of personality and presses ‘publish’, but they do so with all the ingredients they need within arm’s reach from the outset.

While you could argue that it’s potentially putting freelancers out of work, I’d argue that the flip-side is that allowing SMEs to publish at a higher level while reducing their overheads. Freelancers and contractors still have undeniable value, they just needn’t be such a crutch.

Ai development won’t take long

Recent breakthroughs include self-driving cars, healthcare technologies and speech recognition.

However, the idea that every home, office and school will be teeming with Smart Staplers and NLP-powered Robo-Nannies by the next World Cup remains science fiction (for now at least). In truth, artificial intelligence is very much in its infancy.

It has taken more than half a century to get to this point. This may sound pretty impressive, but it’s a snail’s pace compared to the progress so much other digital tech has made in the same timeframe. If you don’t believe me, ask your dad what his first mobile phone could do.

It’s also important here to note the difference between artificial general intelligence (AGI) and machine learning (ML).

AGI is the ‘science fiction’ Ai which most people imagine. One system that can solve general purpose tasks like running a bath or enslaving humanity.

ML techniques are more concerned with tasks like speech recognition, image classification etc. These ML algorithms can match and exceed human intelligence reasoning, but only on the tasks for which they have been built/trained.

We are just now unlocking the potential of disciplines like ML and the next few years will dictate just how far we can take technologies like ours.

It’ll bring down society

The majority of artificial intelligence myths come from science-fiction fantasy. 

When someone says they fear artificial intelligence, what are they imagining? Robots taking over the world (Terminator), turning on their masters (Bladerunner) or corrupting humans via more ‘friendly’ means (Ex Machina). 

The reality is that machines are not resourceful and have no interest in us. Yes, Ai plays an increasingly important role in our lives. But we have a ways to go before deep learning and machines are solving all of our problems. The real worry should be how humans use this technology, as we are already seeing with mass surveillance, deep fakes and bot usage.

The main reason behind this is straightforward and subscribers to this fear ignore it too often.

Humans develop Ai. Ai is not self-sufficient, just self-improving. Artificial intelligence is capable of doing what it has been programmed to do and nothing more. Trust us, inflo.Ai’s NLP programmes were painstakingly created over a period of years by a developer called Shashank. He’s currently sitting just a few feet away from me.

Once again, we are beholden to those who create, build and train the models that power Ai. 

The truth of things

Poorly designed models or poorly designed datasets can introduce biases in Ai systems that could potentially make unethical decisions. For example making decisions about credit score and approving loans on the basis of gender or ethnicity.

Our tech can do a myriad of incredibly exciting, impressive stuff, but right now, only what Shashank allows it to do. Human build Ai for a singular purpose, so even if they were capable of making moral decisions, these decisions could only be made at Shashank’s discretion (God help us if that ever happens).

For now, as long as Ai powers people, people will continue to power Ai.


To learn more about artificial intelligence myths or ask a question about the future of Ai, email our team on