02.14.24
If you look at the news, you know that Artificial Intelligence is everywhere. There are articles about everything from the danger of AI deepfakes… to concern about whether AI will make jobs obsolete… to optimism about how scientists are using AI to classify heart disease… to excitement over how AI can help you reach your personal fitness goals. AI is creeping into all aspects of our lives. There’s a lot to be excited about, but also a lot of legitimate concern, so it’s no surprise that public perceptions of AI are mixed.
Concerns over AI seem to fall into two main categories: the impacts on our economy and the impacts on our society.
Recent polling shows that Americans are optimistic about the potential for AI to benefit businesses. 66% of Americans believe that Big Companies will be the biggest beneficiaries of this new technology. Afterall, AI can empower greater productivity rates and boost bottom lines. Significant pluralities are optimistic that AI will have a positive impact on the way we learn, work, and access information, most notably, a majority of the general public is favorable towards the use of AI to increase the accuracy of medical diagnoses.
But Americans are also concerned about AI’s potential disruptive effects on the job market for the middle class. Greater productivity allows companies to do more with fewer resources – which could make some jobs obsolete.
Americans are also concerned with the potential for AI to cause significant disruptions to the security of online information and systems. In fact – in recent IPSOS polling – this is the only category (information security) in which Americans believe that AI will have a net negative impact.
When it comes to the impact on society, there’s the potential for groups to become marginalized. There have already been numerous studies about how facial recognition software (a form of AI) doesn’t accurately “see” people of color.
And as we look at AI’s potential impact on societal interactions, we see concerns grow. There is a general lack of comfort with allowing AI to model people in ways that impact real life (e.g. who is approved for a mortgage loan and who is not.)
There’s also a generational gap when we look at perceptions of AI. Americans are MUCH more likely to say that AI will benefit younger generations. As AI becomes more prevalent in our daily lives, Americans are concerned that older adults may get left behind.
What can we do with this information?
Given simultaneous excitement for and fear of AI, it’s important to remember that many opinions on AI are still actively forming. Roughly a quarter of adults are unsure of the impact AI will have on areas like public transit, climate change, school curriculum, allocation of government funds or modeling for things like loan eligibility. This presents an opportunity for interested parties to shape AI’s narrative about its uses and impacts – both positively and negatively.
More than three-quarters of Americans are ready for bipartisan common sense policy guidelines to help regulate how AI is used. Two thirds want regulation that helps people whose jobs are threatened by AI. Sixty-three percent want to establish a liability regime for consumer fraud using AI. And about six-in-ten support protections for people who may face discrimination from AI models.
That said, there is little clarity about who should enact these regulations. About a quarter want an entirely new AI department to be created within the federal government, while only about 12% believe this is within the jurisdiction of Congress. Moreover, a plurality of voters don’t think that AI can be regulated at all.
What is clear is that American sentiment on AI is still forming. Most have heard of AI, but many don’t know exactly what to think, or how they might be advantaged or disadvantaged as adoption of the new technology spreads. Smart policy solutions can help mitigate negative effects.