News
Marketing can change the world

Vox Pop: The battle of intelligence; AI vs. the human mind

The marketing sector can be a complicated place as new marketing tools and techniques are launched, almost on a weekly basis. Powered by The Drum Network, this regular column invites The Drum Network's members to demystify the marketing trade and offer expert insight and opinion on what is happening in the marketing industry today that can help your business tomorrow.

The rise of AI has been a topic on the minds of tech wizards for decades. Now, with Cortana and Siri helping us get up in the morning and directing us when lost, and Tesla's driverless cars set to take the reigns of the highways by 2050, where does that leave us humans? What skills will the world require with machines doing all the work for us? Pioneers of AI have been throwing opinions around about how the rise of machines will affect humanity. Elon Musk, the CEO of Tesla Motors, went as far to claim that AI is "our biggest existential threat" and that the creation of it would be "tantamount to summoning a demon." Drum Network members discuss whether the rise of AI is a real threat to humanity as we know it.

Andrew Kimberley, head of digital strategy and innovation, Latitude

The threats to our existence (jobs) from machines have been with us since the industrial revolution and even before that. Rarely though have machines been threatening jobs where the key skills are knowledge and intelligence based.

The acceleration in machine learning techniques is now starting to make people think about their future job security in the digital space. When these systems are setting our bids, selecting our keywords/targets and wording our ads what will be left for us to do! The view I have come to after many, many motorway miles thinking about it is that we will adapt and being unburdened by things that let’s face it are simple mathematics decisions we will have more time to be more creative and help to improve our marketing far more than the 1000th bid change we could make.

In an industry where most of our time previously has been spent bidding, testing new ads and excluding/adding new targets/keywords we will have more time to add more value thinking about our marketing strategies more broadly which will lead to much more added value.

Adam Skalak, head of SEO, iCrossing

In marketing, we are on the cusp of a 21st century industrial revolution. And this is especially true for digital. I’m certain that in the coming years we will see AI take on the vast majority of back end work. Leaving the machines to do their stuff behind the scenes should however allow marketers to focus much more on the creative content and integration of their marketing activity. I’d like to think this will result in more efficient campaigns but also in campaigns that possess the personal touches needed to make that vital emotional connection with an audience.

Google’s update of RankBrain last year suggests that we are much closer to automated search marketing than we might realise. RankBrain processes search results for Google and I believe that it will eventually perfect the search results for any search query. We are also set to see Google learn which ad copy or meta data works best allowing the search giant to match up keywords with ad copy and meta data automatically. This will free up time and budget for marketers to focus on perfecting the creative, the content and the audience understanding that really make campaigns stand out.

Paul Risebury-Crisp, account director, NMP

As robots take over, the computers in our pockets will be able to find a pub, invite our friends, drive us there – and hopefully back! In the world of online advertising the threat of a robot-uprising is even clearer – who needs people running campaigns when algorithms and automation can run them efficiently? Are digital marketers about to become extinct?

Well, no. Not yet. For one simple reason – computers still can’t understand sarcasm. One recent example; I saw some algorithmic sentiment analysis for a brand. It listed brand-positive social media mentions, including the following gem – “{brand} were as useful as a chocolate teapot”. The robot-overlord-in-waiting which made this faux pas focused purely on the term “useful” rather than “chocolate teapot”. Which missed the point.

Until we get to a place where computers can understand or tell a joke - or even laugh - there will be elements of how humans interact with the world that require at least a level of human oversight to make sense of the context. But with the likes of Google investing heavily in AI – who knows how far away that will be? Once computers have passed the Turing Test of being indistinguishable from a human, surely we will start to become obsolete.

David Skerrett, managing partner, Nimbletank

AI offers the power to transform a brand’s relationships with consumers across all channels, particularly digital and mobile. However, leading voices such as Elon Musk are worried about Terminator’s Skynet happening, having last year donated $10M to help ensure AI is beneficial to humanity. AI entered the public consciousness in 1996, when Deep Blue, an AI created by IBM, played chess against Gary Kasparov, the reigning world champion, and won.

I believe there are four reasons why AI is a powerful trend to stay close to:

The first reason the AI trend is rising powerfully, is that ultimately consumers and marketers are time poor. If a machine can do tasks, largely repetitive, quicker, better, constantly learning and improving, sometimes in real-time, then what’s not to like? Secondly, it can make advertising much more efficient. Tom Bowman, CEO of Intelligent Optimisations, believes AI will lead to increased ROI by saving advertisers time and money. Thirdly, accessibility and the cost of AI is dropping. IBM’s Deep Blue was the size of a room and cost $100 million, this power is now available to everyone 20-years later, at a cost of a $100 smartphone, in your pocket.

Finally, and perhaps most critically, consumer acceptance of AI is growing. Counter to our fear of Terminator-style self-aware Skynet, the last few years has seen the four horsemen Google, Apple, Facebook and Microsoft all offering much hyped yet ever improving time-saving contextual mobile assistants that are AI and natural language based with voice control. This is helping to normalize human interaction with non-humans via AI. After all, AI has been flying airplanes for a while now, again based on the fact machines can do certain things quicker, faster and better.

Claire Passos, business development manager, ifour

Without delving deeper into the subject I can see why people would get paranoid about how humanity is at threat with the rise of automated devices and machines that will take over our jobs. Further investigation by economists at Deloitte, reveal that technology has actually created more jobs than it has destroyed. This sheds new light on the relationship between jobs and the rise of technology by trawling through census data going back to 1871, the conclusion is that technology has actually been a great job-creating machine.

Agreed, machines will take on more repetitive and laborious tasks but as the study highlights they seem no closer to eliminating the need for human labour than at any time in the last 150 years. So, while Cortana and Siri claim to be your very own personal assistant at hand 24/7 to answer even the most obscure questions, for now at least there is still a need for real people to communicate with each other, even if that is in your driverless car whilst on chatting away on their smart devices.

Pete Wailes, chief software architect, Builtvisible

The 20th century saw massive technological development. The journey to the moon, the invention of the Internet, the development of nuclear power and so on inspired us and brought wide-spread change. However, with the exponential growth in computing power and research time, we could well see 1,000 times that progress in the 21st. Nowhere is this more obvious than in the field of AI which is now good enough to beat human performance at specific tasks. As a result, we're seeing the emergence of technology which could threaten more than a third of all jobs.

Driverless cars will make human-driven transportation irrelevant, replacing lorry and taxi drivers. Watson and its ilk could replace most GP functions at your local practice. How society will adapt to a world where for the most part companies need far fewer employees is far from certain. To what degree it will be good or bad will be dictated by our ability as societies to support people who can have no economic function, and thus receive no income from labour. What happens when we remove our need to work will reveal what kind of a people we are, and what we will become.

Nick Livermore, marketing manager, Digital Visitor

It’s a bit of a jump to go from pseudo-AI technology like Siri and driverless cars to the end of humanity. Talk about hyperbole. If that gets you worried, then the news that Google’s DeepMind AlphaGo program has just defeated Go superstar, Lee Se-Dol, in the first of five matches should have you quaking in your boots. Big deal, right? Well, yes – actually. Go – the three thousand year old Chinese board game – isn’t like chess. It’s significantly more complex and has long been seen as a huge challenge for developers of AI technology as it requires ‘intuition’. Lee Se-Dol has been playing Go since he was twelve and is an eighteen times world champion. And its intuition that makes AI scary, especially when it comes to military application of the technology. In fact, last year Elon Musk, Stephen Hawking, Bill Gates and many others co-issued a letter describing artificial intelligence as, potentially, a bigger threat than nuclear weapons.

Do I feel threatened? Not immediately. But, as Hawking said: “Success in creating AI would be the biggest event in human history… Unfortunately, it might also be the last…”

Jen Keywood, head of marketing, Paprika

At Paprika, we develop software which enables our clients to get the insights they need within a timeframe that wouldn’t be humanly possible. Take timesheets for instance. Our software can report on staff utilization and client profitability within a matter of clicks. What our software cannot do, is fill out people’s timesheets for them. Whether we are talking about software or driverless cars, both rely on the data input in the first place. A satnav is a great example of this, the user puts in a post code but could still ended up in the middle of a field if the satnav it isn’t up to date. The human element will always be needed to input the data that machines couldn’t possibly account for.

Martyn Gooding, creative director, Gravity Thinking

According to Erik Brynjolfsson & Andrew McAfee, co-authors of The Second Machine Age - creative endeavours, scientific discovery, social interactions, and emotional intelligence are the last bastions of humanity that AI will struggle to replace, at least in the short term. This is an important consideration for our industry, one which is based on both executional design and creative thinking.

Design is being increasingly automated, both as Photoshop adds more and more functionality but also services such as The Grid – a platform that uses AI to build websites based on your preferences. AI is automating many creative processes from design to music to painting. But can a computer be truly creative? Can algorithms and deep learning create something that is completely new? Computers can create millions of random juxtapositions but they can't know which of them is interesting. Can they understand the social and emotional context of what they create?

Additionally, are algorithms and AI in social media a good thing? Computers don’t know the social and emotional context of content. There are many stories of Facebook serving memories of exes, and dead loved ones into your feed under the innocent title of ‘On this day, XX years ago’, leading to some very upsetting incidents for their users.

Guy Thornton, head of search, Found

Well, some of our planet’s brightest minds have warned us that creating thinking machines could indeed threaten our very existence. Even Stephen Hawking pointed out that, “humans, who are limited by slow biological evolution, couldn't compete and would be superseded". It’s hard to argue with him!

The AI we’ve managed to develop so far, like the ML algorithms used by Google in search, have already demonstrated how positive and useful it can be. However it’s impossible to really know what will happen when machines surpass our own intelligence. Will we be enormously helped by AI, pushed aside and ignored or even destroyed by it? Successfully creating a sentient AI will definitely be mankind’s greatest achievement. But it could also be our last, unless we learn how to pre-empt and avoid the risks that it might bring. “

And if after all this doomsday talk you’re feeling a bit worried, just remember there’s a little while to go before we get there.

Naomi Taylor

Naomi Taylor is editorial account manager at The Drum Network, covering members' news, insights and publications.

All by Naomi