Study: The AI Recommendations We Trust Most
Study: The AI Recommendations We Trust Most
View an infographic about some AI recommendations that people trust and whether of not people trust AI over actual people.
Join the DZone community and get the full member experience.
Join For FreeBias comes in a variety of forms, all of them potentially damaging to the efficacy of your ML algorithm. Read how Alegion's Chief Data Scientist discusses the source of most headlines about AI failures here.
Millions of Americans make decisions based on the recommendations of Artificial Intelligence, and many don't even realize what's happening.
From Netflix's "top picks for you," and Spotify's recommended playlists and artists, to even the profiles shown on dating applications, Artificial Intelligence (AI) is using algorithms to offer users suggestions designed to match our interests. But how much do people really trust those recommendations?
Innovative Technology Solutions surveyed 2,000 Americans to identify the types of AI recommendations they trust the most and least. Their findings revealed that while many people use and trust AI, to a certain extent, they remain more likely to trust the recommendations of an actual friend over those of a device.
The study evaluated consumers satisfaction with seven types of Artificial Intelligence: GPS navigation, television and movie streaming, music streaming, restaurant and bar applications, job search applications, dating applications, and news applications (Facebook). By and large, the type of AI recommendations people are most satisfied with are those from GPS navigation followed by television and movie streaming services and music streaming.
But when comparing the trustworthiness of the recommendations of a flesh-and-blood friend to those of an application's Artificial Intelligence algorithm, people said they trust their friend more for six of the seven AI services. The only case in which people say they trust AI more is when it comes to giving directions via GPS.
When asked how well AI recommendations and suggestions expand a person's horizons and introduce them to new things, the vast majority of respondents said that all seven types of AI either expand their horizons "a lot" or "a little."
The future of Artificial Intelligence is less clear. Many industries are incorporating Artificial Intelligence algorithms in some capacity or another, but Americans' overall trust of a device-created-recommendation is hesitant. When asked about the potential use of AI for aiding with a person's financial investments, respondents rated it a 6.2 on a trustworthiness scale of 10. When asked about AI aiding in the presidential election or other elections, people ranked that much lower, a 2.8 out of a trustworthiness scale of 10.
Share your thoughts in the comments!
Your machine learning project needs enormous amounts of training data to get to a production-ready confidence level. Get a checklist approach to assembling the combination of technology, workforce and project management skills you’ll need to prepare your own training data.
Like This Article? Read More From DZone
Published at DZone with permission of Tricia Harte . See the original article here.
Opinions expressed by DZone contributors are their own.
{{ parent.title || parent.header.title}}
{{ parent.tldr }}
{{ parent.linkDescription }}
{{ parent.urlSource.name }}