JPMorgan CEO urges slowdown of AI roll-out to ‘save society’

JPMorgan CEO Jamie Dimon warns that the speedy roll-out of synthetic intelligence (AI) all through society will trigger “civil unrest” until governments and firms work collectively to mitigate its impact on job markets.

Talking on the World Financial Discussion board (WEF) in Davos, Dimon mentioned that though AI can ship “a extra productive society… [and] remedy a number of cancers”, the roll-out must be phased to stave off the potential for enormous social unrest.

“My view is, ‘Don’t put your head within the sand.’ It’s what it’s. We’re going to deploy it. Will it remove jobs? Sure. Will it change jobs? Sure. Will it add some jobs? In all probability. It’s what it’s,” mentioned Dimon. “Nevertheless, I do suppose it might go too quick for society. And if it goes too quick for society, that’s the place authorities and enterprise in a collaborative [need to] manner step in collectively and provide you with a technique to retrain folks or transfer it over time.”

Highlighting the instance of the 2 million business truck drivers within the US, Dimon mentioned that if AI is imposed on them in a single fell swoop when efficient driverless autos hit the street, all these folks may probably go from making $150,000 a 12 months to $25,000 of their subsequent jobs: “Do you have to do it abruptly?… No, you should have civil unrest, so section it in.”

Acknowledging that even JPMorgan will possible have fewer staff in 5 years as a consequence of its use of AI, Dimon went on to induce governments to plan for these eventualities now, by growing retraining, wage help and relocation programmes to help staff displaced by the know-how. He added that governments would additionally must develop new incentives for corporations to sluggish their deployments of AI and guarantee they supply earnings help.

“If a city loses a manufacturing facility they usually lose jobs, you could have earnings help, relocation, early retirement, retraining. We might have to try this…to save lots of society,” he mentioned.

When interviewer Zanny Minton Beddoes, editor-in-chief at The Economist, identified that earlier Commerce Adjustment Help programmes within the US had been “extremely poorly completed”, Dimon agreed, saying: “We should be ready to have one thing that works this time.”

Commenting additional on JPMorgan’s personal use of AI, Dimon mentioned the Wall Road lender has up to now developed round 500 use circumstances for the know-how throughout its enterprise models, noting whereas there are clear effectivity advantages to be gained from AI, it may additionally essentially change the enterprise because it develops additional.

“For those who take it to the following step, brokers, that might change what you are promoting, the pace at which issues occur, how folks entry our programs,” he mentioned, including that corporations might want to develop their AI capabilities to stay aggressive, significantly within the monetary sector given the explosion of fintech corporations in recent times.

“For those who put your head within the sand, you’ll lose. I feel that was true 30 years in the past, however it’s in all probability extra true at this time. And the mind energy and cash that’s going to this [AI] factor is extraordinary. And so, if we don’t do our job quicker, faster, we’ll lose quickly.”

Job losses have already begun hitting the tech sector itself, with corporations globally reducing their workforces as they appear to extend spending on and funding in AI instruments. In October 2025, for instance, Amazon laid off 14,000 staff, a call that was particularly prompted and enabled by the agency’s AI investments.

In August 2025, the UK Trades Union Congress (TUC) warned that AI-fuelled financial progress is leaving staff behind and highlighted the significance of collective bargaining because the know-how turns into extra embedded within the office.

Noting that AI could also be utilized by some employers to chop prices and automate current processes, somewhat than make investments, increase and innovate, it mentioned in a report: “Such choices will extra possible displace or deskill staff somewhat than increase, increase or retrain the workforce as a part of technological upgrading.” 

The paper’s authors famous that if machines do extra duties and scale back the demand for expert staff or for labour total, staff may change into much less capable of command a justifiable share, with the excess more and more captured by employers and AI corporations.

In November 2023, the Autonomy suppose tank within the UK argued that whereas automating jobs with giant language fashions (LLMs) may result in important reductions in working time and not using a lack of pay or productiveness, realising the advantages of AI-driven productiveness beneficial properties on this manner would require concerted political motion.

The suppose tank added that this was as a result of it’s clear that productiveness beneficial properties usually are not all the time shared evenly between employers and staff, and rely upon “geographic, demographics, financial cycle and different intrinsic job market elements” equivalent to staff’ entry to collective bargaining.

To ship constructive AI-led adjustments for staff and never simply employers, Autonomy beneficial establishing “automation hubs”, underpinned by commerce union and business agreements, to spice up the adoption of LLMs in methods which can be equitable.