84% Of US Voters Say AI Threatens American Workers — Far More Than Illegal Immigration
AI Job Fears Surge Across America As Voters Demand Tougher Government Rules
The Political Fight Over AI Jobs
A new nationwide poll finds concern about artificial intelligence now cuts across party lines, with voters also backing tougher regulation, limits on data centres and restrictions on military targeting.
Artificial intelligence has overtaken illegal immigration as a perceived threat to American workers in a striking new snapshot of US public opinion.
Eighty-four per cent of registered voters surveyed said AI poses a threat to American workers. Sixty per cent said the same about illegal immigration. Competition from foreign workers was seen as a threat by 74 per cent, while 73 per cent said large corporations threatened workers.
The numbers do not prove that AI will destroy more jobs than immigration, foreign competition or corporate restructuring. They show something different: how voters currently perceive the risks.
That distinction matters. Public fear around artificial intelligence has moved far beyond specialist debates about algorithms and chatbots. It is now tied directly to wages, job security, economic power and the role of government.
AI Job Anxiety Has Become Mainstream
The result is unusually broad because concern is not confined to one political camp.
The poll found that 84 per cent of registered voters regarded AI as a threat to American workers. That makes employment one of the clearest pressure points in the public debate over the technology.
The fear also fits a wider pattern already visible in 2026. Earlier polling of American workers found that two-thirds expected AI to make the worker experience worse by eliminating jobs and increasing pressure on employees rather than improving work by removing repetitive tasks.
The concern is understandable because AI is already reshaping how work is organised.
The most immediate change is not necessarily a machine replacing an entire occupation. It is software absorbing individual tasks once performed by people.
Drafting, coding, administration, customer support, document analysis, data processing and research can all be accelerated or partly automated. Once enough tasks inside a role become cheaper to perform with software, employers can redesign teams, reduce hiring or expect fewer people to produce more.
That can affect workers even before mass redundancies appear.
A company that once hired ten junior employees may decide it needs six. A department may leave vacancies unfilled. Entry-level work can shrink while experienced employees use AI to handle a larger workload.
The result may be a labour-market shock that arrives gradually rather than through one dramatic wave of replacement.
Why The Immigration Comparison Is So Striking
Immigration has occupied an enormous place in American political debate for years, particularly around wages, border security and competition for work.
Yet in this poll, significantly more voters saw AI as a threat to workers than illegal immigration.
That does not mean immigration has suddenly become unimportant to voters. The comparison instead shows how quickly artificial intelligence has entered the same economic territory.
AI presents a different type of employment pressure.
Immigration changes the supply of human labour. Artificial intelligence can change the amount of human labour required to complete a task in the first place.
If a business can automate part of its workload, the economic question is no longer only who competes for a job. It becomes whether the same number of jobs needs to exist at all.
That is why arguments about what happens economically when AI replaces or reduces human work are moving from theoretical discussions into mainstream politics.
The crucial uncertainty is scale.
AI could increase productivity enough to create new industries, lower costs and generate new forms of employment. It could also remove large amounts of routine work faster than economies can create replacements.
Both outcomes can happen at the same time.
Voters Also Want Tougher AI Regulation
Employment is only one part of the anxiety.
The poll found that 80 per cent of Democrats and 61 per cent of Republicans supported stricter government regulation of artificial intelligence.
That degree of cross-party support is significant because US technology policy is often pulled between two competing goals.
One is speed. American policymakers want domestic companies to remain ahead of Chinese competitors and maintain leadership in advanced computing.
The other is control. Voters increasingly want safeguards around systems powerful enough to affect employment, infrastructure, security and personal decision-making.
The Trump administration has repeatedly emphasised American technological leadership and the expansion of AI infrastructure. It has also supported voluntary safety commitments from major AI companies.
The political difficulty is that voluntary principles are not the same as enforceable regulation.
That gap sits at the centre of the wider debate over what meaningful AI safety rules could actually look like.
Rules can range from transparency requirements and independent testing to restrictions on particular uses, liability standards and mandatory human oversight.
The challenge is writing rules that address real harms without freezing useful technology in place.
Most Voters Think Washington Is Not Taking The Risk Seriously Enough
The poll also found dissatisfaction with the federal response.
Fifty-seven per cent of registered voters said the Trump administration had not taken AI risks seriously enough or at all. Fifty-four per cent said the same about Congress.
Those figures matter because AI policy is becoming an electoral issue rather than a specialist regulatory question.
The United States votes in midterm elections on 3 November 2026, with control of Congress at stake.
Artificial intelligence is now increasingly connected to questions voters already understand clearly: Will my job still exist? Will electricity bills rise? Should algorithms make decisions about people? Who is responsible when automated systems fail?
Those are more politically tangible questions than abstract arguments about model capability.
Data Centres Are Becoming Part Of The Backlash
The infrastructure behind AI is also drawing scrutiny.
Fifty-six per cent of registered voters supported limits on building data centres. Support stood at 69 per cent among Democrats, 55 per cent among independents and 43 per cent among Republicans.
That debate is likely to intensify because powerful AI systems require enormous computing capacity.
Data centres need electricity, cooling, land and grid connections. Large projects can create investment and construction jobs, but they can also trigger local concerns over power demand, water use, noise and electricity prices.
The technology therefore has a physical footprint that voters can see.
The question is no longer simply whether people want better AI systems. It is whether communities are willing to absorb the infrastructure needed to run them.
That pressure is already making power and cooling central constraints on the AI boom.
Military AI Is Another Emerging Red Line
The survey also found growing resistance to the use of artificial intelligence in warfare.
Fifty-seven per cent of voters said the US government should not use AI to determine the target of a military strike.
That was up from 49 per cent in a comparable poll conducted in August 2025.
The wording is important. Artificial intelligence is already used in military systems for functions such as analysis, logistics, surveillance and decision support. The more controversial question is how much authority a machine should have when a decision could lead directly to death.
That concern mirrors a wider public instinct: people may tolerate AI as an assistant more readily than as a final decision-maker.
The dividing line becomes especially important in hiring, medicine, policing, military action and other areas where mistakes can carry irreversible consequences.
Fear Does Not Mean AI Adoption Will Stop
There is a paradox inside the polling.
Americans are worried about artificial intelligence, but AI use is still expanding.
Businesses continue to invest in automation. Workers are increasingly using AI tools. Technology companies are spending heavily on models, chips and computing infrastructure.
That means public resistance is unlikely to stop AI development by itself.
Instead, it may determine the conditions under which development continues.
The next political phase may focus less on whether artificial intelligence should exist and more on who bears the cost when it changes the economy.
Workers could demand stronger protections. Companies could face tighter rules around automated decision-making. Governments could invest more heavily in retraining, wage support or transition programmes.
The deeper question is distribution.
If AI allows an economy to produce more with fewer hours of human labour, that can create enormous wealth. The political conflict begins when societies decide who receives that wealth and who absorbs the disruption.
The latest polling suggests American voters are already thinking about that problem.
For years, the public debate around jobs focused heavily on immigration, trade and offshoring.
Artificial intelligence has now entered that argument with extraordinary speed.
And for a large majority of US voters, it already looks like the bigger threat.
Sources
Reuters — Most US voters say Trump, Congress don't take AI risks seriously, Reuters/Ipsos poll finds — Supports the headline polling figures on jobs, regulation, data centres, military targeting and confidence in federal action.
Ipsos — Reuters/Ipsos October Survey — Provides the survey dates, sample design, weighting and methodology for the nationwide poll.
Ipsos — American Workers Assess AI Impact — Provides additional 2026 evidence on worker attitudes towards AI, job pressure and uneven workplace adoption.
Next Reads
The Data Shows AI Is Already Reshaping Work — Examines how AI is changing tasks, roles and labour demand before full job replacement occurs.
When AI Replaces Jobs, Who Pays You? The Next Economy Is Already Taking Shape — Explores the economic-policy choices that could follow large-scale AI displacement.
What AI Safety Rules Could Actually Look Like — Looks at testing, enforcement, oversight and the practical design of AI regulation.
