20+ AI Bias & Ethics Statistics (2025)

of people globally are willing to trust AI systems — leaving a majority sceptical

Across 47 countries surveyed, less than half the world trusts AI. Trust is the central challenge for ethical AI deployment.

of US adults say AI in daily life makes them more concerned than excited — up from 37% in 2021

Rising concern (a 13-point jump in 4 years) reflects growing public unease about AI's societal impact and ethical risks.

of people globally believe AI will result in benefits — but trust in how it's deployed remains low

The paradox of AI ethics: most people see potential benefits but don't trust the systems or the companies building them.

of people say they 'always trust' information from AI — the vast majority remain cautious

Just 1 in 12 people fully trust AI output, underscoring the ethical obligation for transparency and accuracy.

of people would be less likely to engage with content if they knew it was AI-generated

The 'AI label penalty' — simply disclosing AI involvement reduces engagement by half, creating a transparency paradox.

of people have witnessed AI making factual mistakes — eroding trust when transparency is lacking

More than 7 in 10 users have seen AI hallucinate or produce errors, making transparent AI labelling an ethical imperative.

of people admit relying on AI output without evaluating its accuracy — a transparency and literacy problem

Two-thirds of users don't verify AI outputs, raising ethical concerns about deploying AI in high-stakes domains without safeguards.

of people are worried about AI's environmental impact — an under-discussed ethical dimension

AI ethics extends beyond bias: nearly three-quarters of people worry about AI's environmental footprint.

of Gen Z trust AI to be objective — vs. just 18% of Baby Boomers

A 31-point generational gap in AI trust shapes how different age groups experience and perceive AI fairness.

of Boomers say 'I don't trust it' as their main reason for not using AI

Trust — not access or skill — is the primary barrier to AI adoption among older generations.

of women would be less likely to engage with AI-generated content — vs. 42.54% of men

A 13-point gender gap in AI content engagement suggests women apply more scrutiny to AI-generated information.

of people aged 60+ want less AI-generated content — the strongest age-group resistance

Older demographics overwhelmingly resist AI-generated content, creating an ethical imperative for transparency in content origin.

gap between AI expert optimism (56%) and public optimism (17%) about AI's societal impact

Experts building AI are far more positive than the people affected by it — an ethical concern about who defines AI's future.

median across 35 countries say they are excited about AI — global enthusiasm is the exception, not the norm

Globally, only 1 in 6 people express excitement about AI — the ethical burden of deployment falls on a largely unexcited public.

of Americans believe AI will positively impact education — yet AI is rapidly being deployed in schools

A 24% approval rate for AI in education raises ethical questions about deploying AI where public consent is weak.

believe AI will positively impact jobs — vs. 32% who say negatively — a net-negative public assessment

The public sees AI as a net negative for employment — ethical AI deployment must address these workforce concerns.

organisations have mature AI governance — leaving 80% without formal ethical safeguards

Ethical AI requires governance infrastructure — the vast majority of organisations don't have it.

of people globally say national and international AI regulation is needed — a mandate for ethical governance

A global supermajority demands regulation — the ethical case for AI governance is overwhelming.

of organisations have effectively enforced a ban on generative AI — policies without governance fail

AI bans are virtually unenforceable — ethical AI requires active governance, not prohibition.

of enterprises rate their AI strategy as highly prepared — but 58% are ethically and operationally exposed

Less than half of enterprises are prepared for the ethical and compliance demands of responsible AI deployment.