London-born tech entrepreneur Lloyd Wahed has founded two search firms and has successfully helped scale a number of unicorn technology scale-ups. Wahed is also the host of Searching For Mana podcast – interviewing some of the leading and influential men and women building the future in tech innovation and finance.
With artificial intelligence making strides in popularity, it’s becoming increasingly relevant in businesses across the board – from self-driving vehicles like Tesla cars to the world of finance with Monzo and Tide. According to a 2019 survey by Garner, AI usage grew 270 percent over the last four years and tripled in 2019.
It is, however, important to understand where machine learning and artificial intelligence do not solve underlying issues. Here are five misconceptions when it comes to AI.
Myth 1: Using AI when hiring proves to increase diversity.
AI has not been proven to increase diversity when used for recruitment purposes and in fact, AI approaches have increased the risk of an even higher bias in some cases. A recent scientific study from Microsoft and Cornell University that analysed the claims and practices of companies offering algorithms for employment assessment highlighted the lack of transparency from current AI vendors to disclose specifics on how they actually mitigate algorithmic bias.
It’s no wonder then why Amazon, one of the most innovative companies worldwide, ceased its AI recruiting tool as the resume matching machine resulted in higher averse to women. As for diverse talent acquisition, my podcast guest Shefali Roy, COO & CCO at TrueLayer shared that the word "visionary" in the job specification did not increase the number of women applying for the position.
Myth 2: The only way for AI to carry out a task is to mirror humans.
Algorithms have been designed to reproduce many tasks faster than humans, but do not depend on imitating the way we reason.
In 1997 the chess-playing computer Deep Blue managed to beat Garry Kasparov. This supercomputer was in power of processing 330 million moves in one second, where a human player can go through 100 possible moves in around three minutes.
Interestingly, this machine-learning did not account any human input apart from the basic rules of the game and it didn’t try to copy the human thinking process either.
Myth 3: AI will displace all humans in businesses.
There is a growing misconception that Artificial Intelligence and algorithms will replace entire professions or jobs. They will definitely and already are in the capacity of automating many labour-intensive tasks, which means in turn that AI will help us save time and money on repetitive actions.
In my recent podcast interview with Lara Gilman co-lead of iwocaPay, Lara highlighted that whilst technology can consume and understand data, their company also has a philosophy that machines are good at some things, and humans are also good at some things.
She says "We have to be able to lend the benefit of machines and humans. Machines help us consume data and recognise patterns in data but they’re not so good at client concentration risk. For us, it’s important to blend in that human part".
Myth 4: You need large budgets to use AI for business.
Many tools are becoming increasingly accessible for businesses and don’t require large investments.
Most companies will choose to use business applications developed on top of tools that companies such as Google, Apple, Amazon, Facebook, and well-funded startups build sharing their APIs. Businesses should also leverage the open source and open innovation movement in order to join the club.
Myth 5: Businesses don’t need an AI strategy.
Understanding the benefits of AI and how it increases efficiency to the next level is vital for your business.
It allows companies to cut down their labour force as AI can do the jobs that took thousands of people hours in seconds.
It can provide incredibly accurate market analysis allowing for business growth and pinpoint accurate investment.
It allows business models to be incredibly lean.
To summarise my beliefs on Artificial Intelligence which I’ve discussed with Robert Elms recently in a BBC Radio London interview – we shouldn’t be scared of Artificial Intelligence if we're proactive.
Careers aren't going anywhere for the next 50 years but they're going to take a different approach.
Artificial Intelligence, if governed the right way, will hopefully take the middle layer out of companies that are quite admin intensive.
What is going to be brought in is the ability for us to use our emotional and social skills that we have as humans.