Public sentiment on AI decision-making: A recent Consumer Reports survey reveals widespread discomfort among Americans regarding the use of artificial intelligence in making important decisions across various sectors.
- The survey, conducted in the summer, polled over 2,000 Americans on their attitudes towards AI-driven decision-making in job hiring, banking, renting, medical diagnoses, and surveillance.
- A significant majority of respondents expressed unease with AI’s role in these crucial areas, highlighting a growing concern about the technology’s expanding influence.
Job interview concerns: The use of AI in job interviews emerged as a particularly contentious issue, with a vast majority of respondents expressing discomfort.
- 72% of those surveyed said they would be uncomfortable with AI scanning their faces and answers during job interviews.
- An even more striking 45% reported being “very uncomfortable” with this concept, underscoring the deep-seated reservations about AI’s role in employment decisions.
Financial and housing applications: The survey also revealed substantial concerns about AI’s involvement in financial and housing decisions.
- Approximately two-thirds of respondents expressed discomfort with financial institutions using AI to determine loan eligibility.
- A similar percentage were uncomfortable with landlords employing AI to assess rental eligibility, with nearly 40% stating they were “very uncomfortable” with this potential application.
Surveillance and medical applications: AI’s use in surveillance and healthcare also raised significant concerns among respondents.
- More than half of the surveyed Americans expressed discomfort with AI facial recognition surveillance, with about one-third being “very uncomfortable” with the concept.
- In the medical field, half of the respondents said they would be uncomfortable with AI being used for medical diagnoses and treatment planning.
Transparency and data concerns: The survey highlighted a strong desire for transparency and control over AI decision-making processes.
- An overwhelming 83% of respondents said they would want to know what data the AI algorithms were trained on when making decisions about them.
- Even more significantly, 91% expressed a desire to have a way to correct data when it was wrong, indicating a strong preference for human oversight and intervention in AI systems.
Contrasting industry trends: Despite public reservations, some businesses and government entities are moving forward with AI implementation in decision-making processes.
- California Governor Gavin Newsom announced partnerships with five AI firms to test generative AI in government agencies covering transportation, public health, housing, and taxes.
- In the private sector, companies like McKinsey and Deutsche Bank are embracing AI technologies for decision-making, potentially risking the replication of historical biases in new, algorithmic forms.
Public sector resistance: Some government AI initiatives have faced pushback from concerned citizens.
- A similar AI project undertaken by New York City’s housing department was met with successful tenant protests, demonstrating the potential for public resistance to such technologies.
Broader implications: The survey results highlight a significant gap between public sentiment and the direction of AI implementation in both public and private sectors.
- The overwhelming preference against decision-making AI suggests a need for more careful consideration and public dialogue about the role of AI in society.
- However, if recent trends continue, it seems likely that many organizations will pursue AI implementation despite these concerns, potentially leading to increased tension between technological advancement and public trust.
Americans Absolutely Detest AI That Makes Decisions for Them
