Artificial Intelligence – the branding implications (Q&A with Annie O'Rourke)
February 24, 2017
Digital assistants powered by artificial intelligence have created a frantic corporate arms race.
At the moment it’s all about computer programmers and technologists grappling with the infinite possibilities of artificial intelligence. But there’s more to success in this field than technology. If digital assistants are to be part of our lives we have to like them as well as find them useful.
This article captures the views of 89 Degrees East’s Director, Annie O’Rourke, who has been deep-diving into the artificial intelligence space with the opening of her new venture ‘Digital Workforce Australia’.
Q1. What do you see as the future for artificially intelligent digital assistants?
Digital assistants are already part of our lives (think Siri and ‘Hey Google’).
But they are propagating at a rate that will soon see them infiltrate every aspect of our lives and transform the way companies interact with their customers. Each generation improves exponentially on the next. Digital assistants should enhance and extend rather than replace existing capacity.
Q2. What do you see as the risks for organisations implementing an artificially intelligent digital assistant?
The first risk is doing nothing and being left behind as this technology revolutionises the way everyone does business. The next biggest risk is rushing into a ‘cookie cutter’ approach informed by IT salespeople or consultants with a vested interest in selling their product. Before an organisation starts its AI journey, it is critical to understand exactly how digital assistants can better assist, engage, inform and support your unique relationship with customers and other stakeholders. Off-the-shelf style digital assistants run the risk of devaluing your brand and alienating your customers.
Q3. What role does branding and communications play in the development of digital assistants?
Up until now, most of the focus has been on the technological capacity, but there is another factor that defines which digital assistants will succeed and become an essential part of our lives. Technologists call it ‘level of humanity’ - but most people will be more familiar with the term ‘branding’.
All the things we think about with brand - including the look, voice, tone and personality - impact on how we will receive and accept digital assistants and the access we will allow them into our lives.
Q4. To successfully activate a digital assistant, a communications strategy is of equal importance as the IT behind the AI. You simply can’t have one without the other. So should digital assistants be made as human-like as possible?
Digital assistant interactions are increasingly human-like. Of course they are not people, but they can and in many cases should be designed to have personalities, where appropriate.
The digital assistants that are succeeding at the moment, like Meeting Manager’s Amy Ingram (x.ai), demonstrate high ‘level of humanity’.
Clients describe Amy as warm, collegial and sometimes even flirtatious. Her interactions are so human-like she has even been sent flowers and chocolates and asked to meet in the foyer. One client tweeted about how she tried to connect with Amy, saying “Ha! You work for 2 of my friends!” before the dawning realisation that Amy was a computer.
Q5. How do you create voice and tone that is appropriate to your brand?
Most interaction with your digital assistant will be via voice platforms. But for many people, a robotic-style voice has sinister connotations – starting with HAL in 2001 A Space Odyssey and more recently in the disembodied voice of Ava in Ex Machina. Voice must be considered a powerful tool for increasing the ‘level of humanity’ quotient.
The developers of Amy Ingram took at lot of time creating a friendly, authentic voice for the emails she sends. One of the many communication activities they undertook as part of their strategy was to hire a professional scriptwriter to develop the persona and consistency of voice.
Cubic Robotics claim their AI ‘butler’, tailored to control your smart home, is more like a friend than a machine. If you make references to Game of Thrones it will follow up with GOT jokes in its banter.
Mark Zuckerberg’s personal assistant Jarvis is voiced by the wonderful Morgan Freeman, and our own Cate Blanchett has lent her voice to the National Disability Insurance Scheme’s virtual assistant Nadia.
Q6. How do you develop ‘personality’ for AI?
Just like all brands have (or at least should have!) a unique presentation, engagement and voice, so too should your digital assistant.
As we all know, personalities are complex and evolving. Successful digital assistants are like a good book or film with a series of layers and reveals needed to enable a relationship to grow, change, challenge and surprise.
Q7. What relevance does using a ‘name’ have for a digital assistant?
If you want and/or need your clients or stakeholders to engage and trust your digital assistant, it will need to be a name they can connect with.
At the moment there are far too many female digital assistants being developed – and white ones at that. Like in all areas of life, we need to break that mould and help companies and organisations feel more comfortable with diverse ‘assistants’. It is great that Mark Zuckerberg made Jarvis a male.
Q8. How far can a digital assistant go before people feel uncomfortable?
If a person can’t tell if they are dealing with a person or a machine they often feel uncomfortable and reluctant to engage. This feeling even has a name – 'uncanny valley'. So it’s important to be upfront with people and let them know about your digital assistant and what you are hoping to achieve from developing and activating it.
Your digital assistant has to have a likeable and relatable persona, but don’t over play it – no one likes a fake. Except maybe Donald Trump.