Responses to the survey question: “What’s the first word that comes to mind when you think of AI?” (Image via WordClouds.com)

AI is the hottest topic in tech. Corporate giants and startups are racing to embed generative AI capabilities into their products and services. Regulators are trying to set rules governing its use. And experts are debating whether AI will have a positive or negative impact on society at large.

As part of GeekWire’s recently relaunched “Bot or Not” series, which examines the line between human ingenuity and artificial intelligence, we surveyed readers to get a pulse on how people are thinking about AI, and to help inform our coverage of the important topic.

Naturally, we asked ChatGPT to analyze the results.

I gave the chatbot a spreadsheet of the survey responses. It sent back a basic recap of the answers. (Note: we used GPT-3.5, as our full dataset exceeded GPT-4’s caps.)

Here’s what it produced after analyzing answers to the question “Do you feel AI will benefit or harm society in the future?”

“Responses were mixed, with some expressing that AI will bring benefits, while others believed it will bring harm. Some responses mentioned both benefit and harm, and others emphasized the importance of responsible implementation and regulation.”

I asked ChatGPT to do further analysis, and that’s where the AI displayed its capacity for nuanced (though not entirely accurate) takeaways.

Here’s how ChatGPT responded to the follow-up request. I bolded the last sentence of each blurb for emphasis.

  1. Sentiment Analysis: Analyzing the responses to various questions, we can determine the overall sentiment towards AI. While there were mixed sentiments, a majority of the responses leaned towards a positive outlook on AI, with words like “wonderful,” “opportunity,” and “amazing” being mentioned. However, there were also mentions of concerns and negative associations, such as “anxiety,” “job displacement,” and “deceit.” It indicates a complex perception of AI, with both optimistic and cautious viewpoints.
  2. Adoption of AI: In terms of personal use, a significant portion of respondents mentioned not using AI in their personal lives. However, those who did use AI mentioned leveraging it for assistance, content creation, research, problem-solving, and learning. In the context of work, while some respondents reported not using AI, others mentioned using it for automation, writing, research, marketing, customer support, and data analysis. This suggests that AI adoption is not yet widespread across all individuals and industries.
  3. Concerns about Job Replacement: The responses to the question about worries regarding AI replacing jobs were distributed across the scale. While some expressed low worries (rating 1-3), others indicated higher levels of concern (rating 7-10). This disparity suggests varying levels of awareness and apprehension about the potential impact of AI on employment.
  4. Industries Impacted by AI: The responses indicated that a wide range of industries are expected to be impacted by AI. Some frequently mentioned industries include technology, entertainment, healthcare, customer support, media/advertising, and education. The diversity of mentioned industries highlights the broad scope of AI’s potential influence.
  5. Future Impact of AI on Society: The views on the future impact of AI on society were diverse. While some respondents believed that AI will bring significant benefits to society, others expressed concerns about potential harm. Some respondents emphasized the need for responsible implementation and regulation to ensure positive outcomes. The mixed responses highlight the ongoing debate and uncertainty surrounding the societal implications of AI.
  6. Coverage of AI in News: Respondents expressed interest in various topics related to AI, including AI safety, industry impacts, job displacement, ethics, regulation, advancements, and non-traditional AI applications. The desire for accurate and responsible news coverage indicates a need for comprehensive reporting to address the wide-ranging concerns and interests in the AI domain.
  7. Leading Cloud Brands in Generative AI: When asked about the leader in generative AI, respondents mentioned Microsoft, Google, and OpenAI as notable brands. However, there were also respondents who couldn’t identify a clear leader or were unsure. This suggests that the perception of leadership in generative AI is subjective and dependent on individual awareness and experiences.

These answers from ChatGPT weren’t 100% accurate.

For example, we asked respondents: “What’s the first word that comes to mind when you think of AI?” ChatGPT said “a majority of the responses leaned towards a positive outlook on AI.”

But after looking at answers myself, only about 10% of the answers would qualify as a “positive.”

Maybe ChatGPT has a bullish view of its inner workings.

For the question about job displacement, we asked readers to provide a number from 1-10, with 1 being least worried about robots taking their job, and 10 being most worried. ChatGPT told us the answers varied. That was accurate, but 34% answered with “1,” and another 18% answered 2-5 — indicating that most people aren’t that worried.

In other words, ChatGPT didn’t quite nail this one.

We asked readers to provide a number from 1-10, with 1 being least worried about robots taking their job, and 10 being most worried — 34% answered with “1.”

But for the most part, ChatGPT’s deeper analysis was pretty spot on, and summed up the survey results faster than any human could have done.

Here are some actual answers from the survey:

Readers were split on whether AI will benefit (55%) or harm (45%) society.

  • “I see a lot of good but do worry that people will start learning less and rely on the bot.”
  • “AI will have a negative effect on the 2024 election and will be a major source of disinformation.”
  • “It will destroy creativity.”
  • “While there is significant potential for harm, I feel the benefits will outweigh it.”
  • “AI should only be used to do things humans can’t do.”

We asked readers for the first word that comes to mind when they think of AI.

  • Uncontrollable
  • Wonderful
  • Frightening
  • Layoffs
  • Terminator
  • Transformative
  • Dangerous

Here’s how people are using AI at work and in their personal life:

  • Automating tasks
  • Image creation
  • Writing blog posts, emails, and product descriptions
  • Talking to fictional characters
  • Assistance with creativity
  • A sounding board
  • Amusement
  • As a new search engine
  • Restaurant search

And here are some of the industries people think will be most impacted by AI:

  • Entertainment
  • Writing
  • Coding
  • Legal
  • Customer service
  • Accounting
  • Food
  • Taxes
  • Journalism

Stay tuned for more from the “Bot or Not” series, and email us at tips@geekwire.com with your story ideas and tips. The series will feature insights from AI experts and other leaders, and explore the ethical and practical implications of AI. We’ll get a sense for how it’s impacting work and life everywhere, as viewed from our perspective on business and technology in Seattle and the Pacific Northwest.

This editorial series is reported and written independently by GeekWire’s news team, with underwriting support from Google Cloud. (Read more about underwriting and sponsorship on GeekWire.)

Like what you're reading? Subscribe to GeekWire's free newsletters to catch every headline

Job Listings on GeekWork

Find more jobs on GeekWork. Employers, post a job here.