Skip to Content

The Hopes and Fears for AI: A Halloween Special

Capgemini
2018-10-29

With Halloween around the corner, I wanted to explore a technology that is both exciting and spooky: AI. From HAL in 2001: A Space Odyssey to Ava in Ex Machina and everything in between – AI gone wrong has been a common topic in films over the years. Even tech leaders like Elon Musk have expressed their worries, saying that AI is “a fundamental risk to human civilization”[14]. But how accurate are these fears? Do people know what AI is? How are they feeling about it? We’ll explore such questions in this post.


Ava in Ex Machina[21]

What do people know of AI?

Let’s start with the basic – what is AI? In the most general terms, AI can be described as a “field of technology designed to create computers, machines and software that are as smart as humans, and which are able to perform the same tasks that normally require human intelligence”[2]. There are many different technologies which fall under this umbrella definition; natural language processing, deep neural networks, cognitive computing – just to name a few. Some of these technologies can be grouped together, as can be seen in the visual below, put together by the Capgemini Digital Transformation Institute[1]:

What’s important to realize from this is that AI is many different parts, each with specific functional capabilities. Sometimes they’re referred to as “Narrow” or “Weak” AI. Their scope may be limited, but nonetheless useful, such as using computer vision for medical diagnosis.

The type of AI popularized in media is Artificial General Intelligence (AGI), sometimes referred to as “Strong AI”.  Gartner gives one definition here; “an AI with the ability to apply intelligence to any problem it confronts, and can exhibit consciousness, sense of self, situational awareness, and can reason its way through novel situations.”[5]. Horror movie examples of this include “The Terminator” or “I, Robot”. But, as Gartner continues, AGI is currently not in sight and probably won’t be within the next 10 years[5].

This brings us to our first point – people, in general, aren’t knowledgeable about AI. Weber Shandwick and KRC Research conducted a survey of 2100 adult consumers across 5 countries and found that the majority of consumers don’t know much about AI[2]:

Weber Shandwick and KRC Research[2]

Interestingly, the highest percentage of consumers who reported knowing a lot about AI (31%) were from China, almost twice the percentage compared to countries surveyed. We’ll see further examples of China’s interest in AI as we go on.

In the same survey, 150 chief marketing officers were contacted by phone. Results were similar; 32% reported knowing a little or nothing about AI. Even for CMO’s ramping up for AI, about half know a little or nothing[2]. This blind investment, in a way, is scary itself.

Consumers were asked what the first thing that came to mind about AI, Robots are unsurprisingly at the forefront of their mind.[2]

Weber Shandwick and KRC Research[2]

When asked where these impressions came from the overwhelming majority was from media exposure, the two most sources cited being internet social media (52%) as well as TV/Movies (52%)[2]. The way AI is depicted in media combined with a general lack of understanding perpetuates certain myths. Gartner sums it up well: “AI is overhyped as a socioeconomic phenomenon. The media, governments, corporations and individuals each have an opinion about AI, mostly based on vague ideas of what it really is”[3].

Looking Forward to the Future

A lack of knowledge doesn’t equate to a lack of opinion, and when it comes to AI the general outlook is surprisingly optimistic. In the same Weber Shandwick/KRC Research survey, it was found that more people think the societal and personal impact of AI will be positive.  And, as seen below, China seems to be the most enthusiastic country.

Weber Shandwick and KRC Research[2]

The most cited benefit that consumers think AI will bring is the completion of tasks that are too hard/dangerous for people (72% of respondents)[2]. Many other benefits people think AI will bring about are centered around ease and convenience. Easier access to relevant news/info, easier decision making for purchases of products/services, etc. We can already see this in personalized news feeds and search results, each link clicked or post liked is data used to show you relevant information (creating a filter bubble in the process). But there are more meaningful benefits expected, like a better use of energy/natural resources, positive impact on the environment, improvement to health/longevity, and, interestingly companionship[2]. An example of this is “Samantha” in the movie Her.

Even today it seems consumers don’t mind using AI. A survey of 10,000 consumers performed by Capgemini has found that “(73%) say they are aware of having interactions enabled by artificial intelligence” and that “69% of these AI-aware consumers were satisfied with their AI-enabled interactions”[13]. In fact, the majority of consumers prefer a mix of human and AI interactions for low-consideration products and services (a virtual assistant directing you to the correct human representative), but that’s not always the case for high-consideration products (such as shopping for a wedding ring)[13]

Overall I was surprised by how favorably people are viewing AI. But that’s not to say they don’t have their fears, and one fear in particular.

Job Loss: the Big Bad

Job loss is easily the biggest fear of AI’s impact. Overall 64% of consumers concerned about AI adoption in some way (Chinese consumers being the least concerned at 52%)[2]. The first concern to come to their minds is job loss[2]. This coincides with Capgemini’s survey of senior executives from around 1000 organizations around the world, 61% say the majority of their employees worry about losing jobs to AI[1].

Although some executives are optimistic about AI creating new jobs, or automating repetitive tasks and freeing up time for workers to focus on more creative aspects of their job[1], consumers have clear views. 82% think AI will lead to job loss rather than job creation on the whole[2]. 65% were concerned about losing their own job personally, a worry echoed by those who reported knowing a lot about AI or a little to nothing of AI[2].


Weber Shandwick and KRC Research[2]

And indeed they may have reason to be concerned. In a 2014 pew research survey of 1896 experts – (48%) expected robots and digital agents to displace a significant amount of both blue- and white-collar workers by 2025[15]

Regardless of job loss or creation, AI is expected to change your organization in some way. According to a 2016 Forrester survey of 612 technology specialists and business leaders, 16% reported their organization had already done some restructuring due to AI implementations, with another 13% planning to do so. 51% anticipated that restructuring will occur, and only 20% felt that AI would have no organizational impact.[16]

Generational Differences

There’s an interesting disparity in the level of concern across generational differences in the workforce. Millennials are most concerned about AI replacing our jobs, whereas boomers are the least concerned[2]:


Weber Shandwick and KRC Research[2]

As millennials are young in their careers, they will comparatively have the most time to experience the disruption caused by AI. That’s of the current workforce, the next generation will be in even more trouble if their parents’ concern is anything to go on. Those who have children younger than 18 will witness their children grow up and enter a workplace which has probably already incorporated AI, and they are just as worried as millennials. On the other hand baby boomers, who some say are retiring at a rate of 10,000 a day[17], will most likely be out of the workforce, or leaving shortly, when the robots come to take our jobs.

Of course, there are other fears besides job loss, the most prominent of which are cyber-security risks, companies or governments gaining too much information (think 1984 or Minority Report), and machines making bad choices[2]. If an AI makes the wrong decision it can lead to accidents[2], as seen in auto-pilot accidents[18].

Beyond physical harm, AI can repeat the scarier aspects of humanity. It may further the divide between the rich and poor; 2/3 of consumers agree AI will be accessible to wealthy people rather than everyone[2]. Or it could continue existing discrimination, which can already be seen in recent news with Amazon’s recruiting engine being biased against women[19].  AI learns from us, so we need to take a critical look in the mirror to understand what it is we’re creating.

Full Speed Ahead

Despite these fears and cautions, AI development is not likely to slow anytime soon. We see increasing amounts of research and news articles being written on AI, more buzz being generated (including this post). The money is there – in 2015, $8.5 billion was spent on AI companies, nearly four times as much as in 2010[20]. The enthusiasm is there – 77% of consumers want AI development to either continue it’s current pace or accelerate, that percentage increases for those who reported knowing more about it. As is this case with most technological advances in the past century, what was once science fiction will soon be reality; whether that will lead to an AI-assisted society or a Terminator-filled horror battleground is yet to be seen.

Millennial Innovation Council

The following article was submitted on behalf of Capgemini NA’s Millennial Innovation Council (MIC) Employee Resource Group (ERG).  The mission of MIC is to engage all employees to strategically position Capgemini in raising its future generation of talent in a changing business environment and an evolving world. By understanding the living situations of millennials Capgemini is putting themselves in a position to attract and retain talent.

This article was written by Scott Smith. Scott has worked at Capgemini for a little over a year as a business analyst in the NA FSSBU, specifically with insurance. After graduating, Scott, like other millennials, moved back into his parent’s house.  He looks forward to sharing more interesting articles.

Sources

1. https://www.capgemini.com/wp-content/uploads/2018/01/turning-ai-into-concrete-value-the-successful-implementers-toolkit.pdf

2. http://www.webershandwick.com/wp-content/uploads/2018/04/AI-Ready-or-Not-report-Oct12-FINAL.pdf

3. Sicular, S., & Brant, K. (2018). Hype Cycle for Artificial Intelligence, 2018. Gartner. https://www.gartner.com/doc/3883863/hype-cycle-artificial-intelligence-

4. http://www.shaperssurvey2017.org/static/data/WEF_GSC_Annual_Survey_2017.pdf

5. Austin, T., Linden, A., & Rollings, M. (2017). Hype Hurts: Steering Clear of Dangerous AI Myths. Gartner. https://www.gartner.com/doc/3754263/hype-hurts-steering-clear-dangerous

6. Purcell, B. (2018). The Forrester Tech Tide™: Artificial Intelligence For Business Insights, Q3 2018. Forrester. https://www.forrester.com/report/The+Forrester+Tech+Tide+Artificial+Intelligence+For+Business+Insights+Q3+2018/-/E-RES143252

7. Le Clair, C. (2018). AI Is Ready For Employees, Not Just Customers. Forrester. https://www.forrester.com/report/AI+Is+Ready+For+Employees+Not+Just+Customers/-/E-RES141543

8. Evelson, B., Goetz, M., & Hopkins, B. (2017). Predictions 2018: The Honeymoon For AI Is Over. Forrester. https://www.forrester.com/report/Predictions+2018+The+Honeymoon+For+AI+Is+Over/-/E-RES139744

9. https://www.gsam.com/content/dam/gsam/pdfs/common/en/public/articles/global-equity-outlook/investing-in-the-millennial-effect.pdf?sa=n&rd=n

10. Brethenoux, E. (2018). AI Use Cases, Tales From the Trenches: A Gartner Trend Insight Report. Gartner. https://www.gartner.com/doc/3891498/ai-use-cases-tales-trenches

13. https://www.capgemini.com/service/the-secret-to-winning-customers-hearts-with-ai/

14. https://www.washingtonexaminer.com/elon-musk-artificial-intelligence-a-fundamental-risk-to-the-existence-of-human-civilization

15. http://www.pewresearch.org/wp-content/uploads/sites/9/2014/08/Future-of-AI-Robotics-and-Jobs.pdf

16. Forrester’s Q2 2016 Global State Of Artificial Intelligence Online Survey

17. Friedberg, B. (2017). Are We in a Baby Boomer Retirement Crisis? Investopedia. https://www.investopedia.com/articles/personal-finance/032216/are-we-baby-boomer-retirement-crisis.asp#ixzz55g2PEuf8

18. Matousek, M. (2018). Tesla’s Autopilot used to give the company a huge advantage — but now it’s becoming a problem. Business Insider. https://www.businessinsider.com/teslas-autopilot-faces-scrutiny-after-accidents-2018-5

19. Dastin, F. (2018). Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G

20. Standage, T. (2016). The return of the machinery question. The Economist. https://www.economist.com/special-report/2016/06/25/the-return-of-the-machinery-question

21. Ava in Ex Machina. (2014). Film4. DNA Films. https://www.cinemablend.com/new/Meet-Ava-5-Things-You-Need-Know-About-Ex-Machina-Star-70795.html