A Blog by Jonathan Low

 

Apr 2, 2017

What's Holding Back Artificial Intelligence? Lack of Trust

There is, first, a generalized perception that AI  is no safer than any other technology in that it can be  hacked or that the information it accumulates or generates can be accessed without the owner's (or subject's) permission.
 
 




But there is the larger issue of wondering why you should trust something that may render you economically irrelevant. JL


 
Jurica Dujmovic reports in Marketwatch:

42% of consumers said they don’t trust AI. Only 9.3% of consumers would allow AI to run their finances, and 4% would trust it with HR-related work. 35.6% of respondents rely on AI for various entertainment-related recommendations. The more respondents believe AI is useful, the more they fear being replaced by it.
Not long ago, I wrote about artificial intelligence (AI), its capabilities and its future.
In that article, the chief technology officer of Advanced Micro Devices Inc. AMD, +3.56% one of the largest makers of microprocessors, provided intriguing insights into the topic.
Today, I want to take it a step further: I got my hands on an interesting study by InsideSales.com, an AI-powered predictive sales acceleration platform. The study includes responses from nearly 2,000 Americans from all walks of life about the perceived dangers or opportunities brought on by AI. Is 2017 the year when artificial intelligence will finally reach the mainstream, and if so, is the general population aware of it? Let’s dive into the document and find out.
When asked if they’ve ever used AI, almost 55% of respondents answered affirmatively. The survey uncovers an interesting correlation between income levels and AI adoption: Those that report the most frequent use of AI are from the lowest income bracket (less than $25,000 a year) and the highest (above $200,000).
Americans’ daily use of artificial intelligence.
Navigation apps (60.3%), video (55.2%) and music streaming (47.4%) are the most common ways AI-enhanced content is consumed. That’s understandable, since those technologies had more than a decade to win over the average person.
But what about the more innovative ways in which AI can be used, such as Amazon.com Inc.’s AMZN, +1.16%  Alexa or Alphabet Inc.’s GOOG, -0.23% Google Assistant? Only 12% of respondents find those devices useful, indicating they’re still deemed a novelty. The survey uncovers two more areas that have failed to woo users enough to warrant greater adoption in their daily lives: home automation (5.5% of respondents report regular AI usage) and bots in the workplace (only 1%).
The biggest hurdle seems to be lack of trust. People don’t have enough faith in AI to allow it to work for them. General lack of trust (42% of consumers said they don’t trust AI) varies between the East Coast and the West Coast as well as central regions. The most skeptical were respondents in New York, Pennsylvania and New Jersey (49.2% of them couldn’t name a single AI product they trusted), while users from the Pacific, West, South Central and New England regions were far more ready to rely on AI.
Trust levels also varied among different industries. Only 9.3% of consumers would allow AI to run their finances, and 4% would trust it with HR-related work. On the other hand, 35.6% of respondents rely on AI for various entertainment-related recommendations, 30.1% would let it produce goods made by automated machinery, while almost 19% believe it could enhance automated sales procedures. The percentage of those who would trust medical diagnostics made by learning and decision-making algorithms is smaller.
The most trusted company for AI-related products and services is Google (54.3%), followed by Apple Inc. AAPL, -0.19%  (46.3%), Microsoft Corp. MSFT, +0.23%  (40.05%) and Amazon (39.6%).
What about the future of learning and decision-making algorithms? Here’s what respondents said:
• Almost 49% of consumers believe AI will lead to medical advancements.
• 46.7% of consumers believe AI will take over dangerous jobs.
• 41.7% of consumers believe AI will automate mundane tasks in their personal life.
• Almost 40% of consumers believe AI will lead to advancements in transportation and travel.
• 35.1 % of consumers believe AI will automate mundane tasks in their work life.
Although Americans may be cautious about AI now, many expect it to keep evolving until it’s capable of performing tasks that are currently beyond its capabilities. This conclusion correlates pretty well with the reality of AI and what it can currently do. It also means that the market in general has great, yet realistic, expectations of AI.
Finally, let’s address the cyber elephant in the room: Will AI take your job? When asked that same question (click here to read what I think about the topic), 35.4% of respondents acknowledged being concerned about their job safety. Millennials and members of Generation Z are generally more worried than older generations, with almost 41% of them saying AI could be coming after their jobs in the future.
A further examination of the survey data leads to more interesting correlations: The more respondents believe AI is useful, the more they fear being replaced by it:
• 42% of Generation Z has a positive view of AI in the workplace; 36% believe AI will decrease the number of jobs available.
• 35% of millennials have a positive view of AI in the workplace; 37% believe AI will decrease the number of jobs available.
• 24% of Generation X have a positive view of AI in the workplace; 37% believe AI will decrease the number of jobs available.
• 21% of baby boomers have a positive view of AI in the workplace; 26% believe AI will decrease the number of jobs available.
When it comes to income levels of those interviewed:
• More than 42% of consumers making under $25,000 a year believe AI will decrease the number of job opportunities.
• Fewer than 26% of consumers making over $175,000 a year believe AI will decrease the number of job opportunities.
So, the more someone earns per year, the less he or she feels threatened by artificial intelligence.
AI is not a passing trend. It’s been with us for decades and is here to stay. As technology and science improve, so will the algorithms behind AI and the hardware that’s running it. However, I still believe it must improve before it can become an inseparable and integral part of our lives.

0 comments:

Post a Comment