From hollywood writing rooms to high-level board rooms, AI is on everyone’s mind. As the most talked about trend, it’s either met with enthusiasm, fear, excitement, or just confusion.
On a business level, advancements in AI technology is allowing companies to make data-driven decisions, to engage customers, and to innovate. But do business leaders truly understand the opportunities that lie in their hands? Or are their hesitations clouding their ability to see the role it can play in their organisations?
Enter Aigen, a new type of consultancy that delivers technology-agnostic solutions and thought leadership to businesses worldwide. We meet James Loft, CEO, whose responsibility and personal mission it is to demystify AI, with the aim of helping his clients achieve their strategic goals through this new and fast-evolving technology.
You recently launched Aigen in partnership with Rainbird - can you tell us a bit about this new consultancy and what you’re looking to do?
Aigen is a technology-agnostic AI Consultancy and leadership organisation. We work with forward-looking organisations to help them to achieve new possibilities through AI technology. We always start with the user and base all of our work around service design principles.
The reason we launched the business is that lots of people were talking to us about how they’ve spent months trying to make their business fit AI tools and, actually, what we realised is that we could take a traditional business consulting approach to business problems with the thought leadership and knowledge of AI alongside it.
We’re partnered with Rainbird, which we feel is one of the better AI platforms in the marketplace. The tool is central to a lot of what needs to be done. It’s biggest power is that it tends to be the brain of any system. Rather than the deep mining of any data, it tends to sit best at the centre of what we do. So, working with them has been really helpful. We feed into them about what our partners are looking for and they support us through delivery of the best AI work using their platform.
You’re currently working with a very mixed client base all with different mindsets, cultures, expectations, and strategic objectives. How have you approached companies sitting across different verticals and are there specific methodologies or processes that you use to unearth the potential impact that AI will have on their organisations?
We do. We have the ‘5 Ds’ - which is our delivery structure. It’s taken from the 3 Ds of design structure. Essentially, we try to treat everyone the same, from how we look at their problems, while recognising their problems are different. And this is where our service design and user centred design lens really plays its part, as lots of people come to us with varying problems and it would be very easy to hear someone say something about call centres and sell them a call centre product, which is what most of the AI marketplace is doing.
What we are doing is we’re saying, okay, so you’re operating in an area in which we need to explore what your challenges are so the first two Ds are ‘Define’ and ‘Design’, which is really defining what your problem is, and designing the right solution for it.
Technology doesn’t come into that space in that initial piece, and it’s quite unusual for a tech consultancy to approach it in that way but otherwise we would just be pumping AI out. We would be doing the AI industry an injustice, as we wouldn’t be furthering its cause but more just selling AI and that’s not what we’re here to do.
Going into a business, is it much more about looking at what the potential impact of AI could be in an organisation as an overarching opportunity or is there a specific problem or strategic objective someone is trying to solve when they engage you?
It’s a bit of both. We tend to get invited in for two reasons. One is “I’ve heard about AI and I’ve been told by my senior managers that I need to have AI but I don’t know what it is, help me find where it fits in my organisation”, which is strange for us in a consulting environment to be invited in like that, but it’s very common at the moment, right the way through to solving a specific problem.
I’ve mentioned call centres - that’s a big area where there’s a perceived need to change quite drastically time wise. It’s sort of where the digital transformation piece is focusing itself and call centres is a hot topic. So lots of people are talking to us about specific elements of AI-natural language processing, decisioning and voice detects being the key ones in that space but every example is different. It’s nice to come across a variety of things in the space. Some people come to us with the basics, which is “I need to save money” and some people come to us with a slightly different vision.
We’ve been presented with a couple of interesting scenarios, one of which is “at the moment I’m a cost centre and I want to become a profit centre” - so it’s not just about saving money, it’s about an internal recognition in a large organisation that something that is perceived as being a service problem can actually exist as a sales or marketing channel, and then you can see that AI is not a way of reducing the number of people you’ve got in your team but can take away the long tail of transactional elements so that your team can get back to focusing on relationships.
Who is your customer?
C-suite mainly. CIOs, CTOs, Heads of Call Centre is a really big one in that space. CMOs is an interesting one that we hear a lot from because you wouldn’t traditionally think of it for tech. It’s anybody really that’s got some transformational work already happening or is vastly aware of the fact that they are next on the list of things that need be transformed and they need to get ahead of the curve.
There are two types of people that we meet. People that know they need to do it but don’t know why, and people who have been trying to do something for the last 10 years who finally see this as the possibility to allow them to do it.
And is the business typically at a certain point in its digital transformation journey? If there isn’t a desire or want to actually change the organisation in some way, then one would assume that with a tool as impactful as AI, it can be quite hard to get anything implemented.
That tends to happen a lot more with the “we think we want this but we don’t know what it is or why we want it” camp. They tend to talk themselves out of it when they hear the possibilities of AI and they almost get a bit scared of the culture impact and recoil slightly. But, on the other side, I think we tend to get people who are either heavily involved in the transformation piece and they’re not getting the results they need, so they need to supercharge it, or are on the journey to building a new transformational piece and at the exploratory stages before it.
But also, we get a lot of organisations who have, for many years, been perceived to be at the back of the pack and they are looking at doing digital transformation but they are looking at doing it in a way of leapfrogging to the front of the pile. If you think about it in the consulting world we refer to them as the ‘big 4’, and we do the same in the banking world. Where there are fives, sixes, sevens, eights, and nines in those kind of industries, it’s about looking to take the top spot and they see this new technology as a way of leapfrogging themselves up there. And it’s not just consulting and banking - it’s every industry.
A recent report by marketing intelligence firm Tactica claimed that cumulative worldwide spending on AI will reach $40.6 billion by 2024. Do businesses risk extinction if they don’t take AI seriously? Especially when their competitors are already well underway in exploring AI as a tool for growth.
I don’t think it’s ever articulated in that way, but I think it’s at the back of their minds and they are probably vastly aware of it but they don’t necessarily say that to us. I think that they know on a lot of occasions that they’re not going to achieve what they set out to do if they don’t use this, and I think there is a wariness to go on a digital transformation path without it because there’s a fear that they won’t get anywhere. It’ll be another false start almost.
We see that the journey the innovation market took - and in many ways is still taking - saw them finding a way to define its place through having conversations with clients around the fact that, if they don’t look at innovation as a tool for growth and change, they face serious disruption from new competitors coming into the market. What place is AI in in its journey - is AI viewed more as an innovation tool or is it viewed more as a brand tool?
It’s a little bit of both and I think that in the same way as innovation, it struggled initially because of the type of organisation you’re talking to. So if you’re talking to an innovative organisation who has gone through that process then it’s about innovation. For others, it’s about using AI as a brand tool
For instance, some organisations want to be seen as forward-looking in technology and those organisations tend to be quite service heavy. Some have either grown with rapid success or want to continue that brand engagement by being the people first at doing something. There are also some who have a brand problem due to the historical performance of their technology, and they’re really looking at AI to be a way to leapfrog their success within a marketplace but actually they’re also using it to be a brand flag.
With that, there’s a caveat with where no one wants to sell driverless cars yet because they don’t want to be the first ones through the gate so everyone’s shuffling forward as quickly as they can without being first, but there is this thing that if we can land something well and brand it ‘AI’, then we can turn around to our customers and say, “look how forward-thinking we are because you’re already using AI when you communicate with us”.
What challenges are you facing in educating the C-Suite about the power and opportunity of AI? Are there things that the C-Suite should consider before speaking to an AI consultancy or attempting to roll out a new platform?
The thing to say is that AI is easily marketed as magic and it’s not. It’s very, very clever but it’s not magic and, it needs to be able to succeed in the environment it’s put in. There are two clear questions. Firstly, if you’re using it to solve a particular problem then be prepared to take a step back to figure out what that problem is before you try to fix it with a new technology.
And secondly, there is a message that is we’re probably one of the few places that gives you the comfort of talking to us without being embarrassed about not knowing anything. In the way Virgin Wines has succeeded so that you didn’t have to go into a wine shop and be embarrassed that you didn’t know anything about wine but still buy a nice bottle, we’re kind of that. We are the Virgin Wines to the 30-year-old male for CEOs and CTOs. We’re a place where there is no stigma about not knowing about AI when you come to talk to us as an AI organisation because part of our role is to get it right for you and to educate you. And we learn ourselves.
We don’t know everything about AI; we wish we did but it moves fast and changes so often. I talk about something on a Friday that came into the news on a Monday and pretty quickly it’s considered as ‘old news’. In a market place like that who can keep up? We’re in the market place and it can be hard.
The most obvious concern with AI and organisation change is jobs being replaced by robots. What do CEOs need to consider around this? Are many businesses facing criticism if they seek to use AI as a tool to replace humans?
It comes up in every conversation at some point in a different guise. From the perspective of talking about it, there’s this underlying thought that if you’re doing something around cost reduction then it’s not the first conversation you’ve had around cost reduction.
This whole ‘AI’s going to take your job’ discussion is a very emotive one and I think the media doesn’t do us any favours. The media has a very distinct view of AI and I think driverless cars being the thing every one is focusing on is ridiculous - the thing about AI is that as soon as it has a success everyone broadcasts it as not being AI so we’ve got this cultural battle going on generally around AI. There’s this belief that we’re striving for ‘strong AI’ or ‘wide AI’ where you have autonomous robots and that’s a very unrealistic thing to put in people’s view because that’s a very long way off and, actually, AI is going to play a major role in our lives in and out of work before we get there.
There’s a cultural journey that AI has to go on. Internally in organisations it’s a new technology and new technologies tend to repurpose jobs. We’re very focused on the principle of augmented humans rather than automated humans and the biggest reason for that is our partnership with Rainbird. Most AI tools start with the data and go up, Rainbird starts with the human and goes down.
Rainbird is a way of codifying human knowledge. It’s about augmenting a human, about capturing the thought process of a human and allowing that to be expanded out. It’s also about getting best practice and about uniformity. It’s not about automation and replacement and that’s one of the reasons we’re keen to be partnered with Rainbird as that’s our philosophy. We’re very interested in how we add tools to humans.
We’re not naive - it’s always in the back of our minds that someone might use what we do for bad and, in the same way that any software that anyone builds has to take that potential risk, so do we have a moral calculus that we apply.
There is a lot of talk about this being the second machine age focused on impacting how organisations are run. How is AI now impacting management especially as organisations start to become more data-driven? Do business leaders need to rethink their roles and how they fit within the organisation alongside AI?
When I was at university, I read a book published in 1992 that said, “In 10 years we will have abolished the concept of middle management”. It’s all to do with knowledge teams, which are far smaller and far more autonomous. It likened the organisation to an orchestra where everyone in the orchestra knows what they’re doing and so they only need one conductor as everyone knows their job. What this augmentation of roles is doing is taking simpler, more day-to-day tasks out of the organisation and forcing an upskill in the workforce. In terms of that, some of it is well accepted, and some of it is forced, but it’s happening either way.
The interesting thing is that we’re moving towards this world a lot faster than anyone was expecting. There’s a great quote from a banking organisation we work with which said that their focus was to move away from training staff to training the system. They want to take away the need to have these really extended periods of training and train the system. So, essentially, train it once, and you’ve got a uniform approach to it. You then need a community of people to specialise in dealing with exceptional.
You need people to know more, and you need them to actually be more remote. You don’t need them to be all together because you don’t need to inseminate information to them in the same way. We’re going to start seeing people working remotely more and smaller collections of people with larger knowledge spaces starting to be the thing of the future. In these structures you don’t need management - they are self managed. We’re going to see a lot less people coming into one place to sit with their managers rather more knowledge managers managing systems to allow other people to do the work.
We’ve seen huge leaps and bounds in as little as the last couple of years and change seems to be happening faster and faster:
- What are your predictions for AI in today’s organisations this year?
- How do you see 2017 playing out?
We’re going to see a couple of things. So I have a personal mission around this, which is to demystify some of the perception around AI.
I think we’re going to see a lot more voice-to-text. It’s going to play a major part this year. We’re going to see the rise of systems such as Alexa from Amazon. We’re already seeing others come to the marketplace. We’re going to see the functionality within those products coming to life a bit more.
I’m always amazed by Alexa. Even if the question is answered incorrectly, what actually happens in the millisecond before the answer comes back is astounding: the voice file is sent to a cloud-based service, which strips the sentence apart, identifies the elements of the sentence, and confirms an intent and a sentiment, and then forms an answer to give back all in .2 of a second. We’re going to see a lot more in natural language this year.
We’re going to see a lot more around chatbots and I know that’s the 2017 thing, but we’re going to see a split between chat and bot. So less chat, more bot. That’s where we’re going to go because if you think about an AI system that builds the knowledge to be able to answer a question from a human, plus the AI that needs to communicate the intent and the sentence, they’re two very difficult pieces of work and you’re going to find that companies are not as willing to create chat interfaces for their bots because of the effort involved. So we’re going to see a lot more fixed-answer bots that have AI supporting their decisioning and because of AI it’s going to be easier to do and quicker to do.
We’re going to see a lot more in the utilities space coming through those kind of channels and I think that the likes of Alexa are really going to accelerate the number of AIs we see. My personal mission is to slightly refocus how people think about that. So there’s this perception at the moment that every AI in the home needs to be a general AI that can answer everything for them. I want to put the responsibility of that back on organisations.
There’s no reason why you should have a super complex AI system in your house and they shouldn’t have a super complex system that can talk to a slave-based system in your house that knows your likes and dislikes. So, for example, if someone like O2 sends you a phone bill in the middle of the night, and it’s higher than usual, and the system in your house knows that when you receive your bill and it’s higher than normal you like to interrogate that information.
While you’re asleep your AI can do this communication piece which says ‘I’m going to need more information on that’ and so by time you get the information, it’s already in a structured format for you to look at. I think that’s huge, and we can see that in the way Amazon have gone about opening up their services. It’s a clear statement that we’re going to go very far with this and that the uses of it need to come more from organisations. Aviva have already started doing an insurance interaction with Amazon, and we’ll see a lot more of that.
So we’ll see a lot more of your personal AI system, be it Siri, be it Alexa, learning more about you and acting on your behalf.