Why AI presents benefits and risks for jobs, wellbeing and society

Many employees and businesses are already using AI technologies to complete tasks and monitor staff. It’s vital that businesses manage AI use in the workplace in the interests of their employees. To maintain a happy, productive workforce and ensure that the benefits of AI are shared equally, employers should give staff a say over the way AI is used at work.

The government also has a responsibility to actively manage AI’s impact on the labour market. The tax system should reward businesses that raise workers’ productivity rather than automating roles and workers should be supported by the skills and welfare systems through any disruption caused by AI.

Joseph Evans: "Anyone displaced by AI should be equipped with skills that allow them to secure new employment."

Are we about to witness a ‘jobs apocalypse’?

The spectre of job losses looms large over the AI debate. Tech leaders like Microsoft AI’s Mustafa Suleyman and Anthropic’s Dario Amodei, alongside prominent politicians like Bernie Sanders and Sadiq Khan, have warned that AI could automate a huge number of jobs. The public are concerned, too: seven in 10 say they are worried about AI’s economic impacts, and six in 10 believe it will destroy more jobs than it creates.

The public’s concerns are understandable at a time when businesses are cautious about hiring and unemployment is rising. However, macroeconomic evidence does not support the idea that AI is currently replacing humans en masse. Instead, the recent rise in unemployment is being driven by a complex, interrelated mix of global economic uncertainty and rising costs for businesses.

AI certainly demonstrates the potential to automate many tasks currently done by humans, and its capabilities are rapidly improving. IPPR estimates that up to eight million jobs in the UK could be exposed to AI if its capabilities develop significantly. However, the impact of AI on these jobs is far from predetermined.

Rather than ushering in a ‘jobs apocalypse’, AI is currently more likely to change people’s experience of work – for better or worse – than replace them.

On the positive side, AI could improve people’s experience of work. If AI technologies become capable of performing monotonous tasks, this could free us up to focus on the more meaningful parts of our jobs. In the legal sector, for example, anecdotal evidence suggests that agentic AI is automating routine tasks and freeing up junior staff to focus on more valuable parts of their role.

There’s a risk, however, that AI worsens the quality of jobs by intensifying workers’ workload, surveillance and management. One study found that when employees at a US tech firm started using generative AI, their workload became more intense. AI is also enabling greater surveillance and algorithmic management of workers. Both can worsen people’s experience of work by diminishing their autonomy, intensifying their workload, increasing their stress levels and potentially undermining fundamental workplace rights. 

What do workers think about AI?

When we asked workers how AI is impacting them at work, roughly equal numbers said it was improving or worsening their experience of their job. A much smaller number said it had replaced them altogether.

In the legal sector agentic AI is automating routine tasks and freeing up junior staff to focus on more valuable parts of their role. Photograph: iStock

This suggests that AI is currently having a more mixed impact on the labour market than doomsday warnings of a ‘jobs apocalypse’ might suggest.

It also highlights that AI is having an unequal impact on workers – helping some, while making work worse for others. This reinforces the need for business and government to act swiftly to ensure that the benefits of AI are shared more equally.

When we asked workers what they needed to feel supported at work in the era of AI, their responses indicated a clear role for employers and government in managing AI’s impacts:

• Fair pay and job security – 29%
• Protection from unfair use of AI at work – 24%
• Legal advice or support for individual problems – 18%
• Training and skills for new kinds of work – 17%
• A say on how AI is used at work – 16%
• A community or network of other workers in my situation – 12%.

Employees should be given a say over AI use

Employers should feel motivated by the potential for AI to strengthen organisational performance. 

If AI use is managed properly, it could support workers to increase their output, benefiting individual organisations and the wider economy. The OECD anticipates that AI will boost productivity growth, especially if business adoption is strong and the capabilities of AI technologies 
are significantly expanded.

The benefits of productivity growth should be shared fairly. At a macro level, AI has the potential to generate a significant productivity windfall which could accrue disproportionately to businesses and capital owners. The distribution of this windfall will be determined by what happens at the micro level of the firm, and specifically workers’ ability to determine how AI is adopted.

Giving workers a say over AI use will also determine whether AI technologies improves workers’ experience of work, makes their job worse, or renders their skills obsolete. It is therefore a vital mechanism for ensuring that workers enjoy the benefits of AI, including its potential to boost job satisfaction and drive economic growth.

Employers can meet this challenge by engaging their workforce directly over AI use. This could involve disclosing how they plan to use AI to change staff workload, management or monitoring. This might be facilitated by direct engagement with a trade union or the creation of a staff consultative body.

Ultimately, meaningful consultation is likely to benefit both employers and workers. Perceptions of unfairness can have a damaging impact on staff trust and morale, potentially harming staff retention and workforce productivity. On the other hand, direct and open consultation could lead to AI use which benefits both the business and the workforce.

The government should actively manage AI’s impact on the labour market

Government also has a vital role to play in steering AI adoption to support stronger, fairer economic growth. There could be huge upsides for the public if AI strengthens UK productivity, translating into higher tax receipts for the state and greater potential to invest in essential public services. On the other hand, a mismanaged transition could lead to backlash from a public which is becoming increasingly worried about AI’s impacts.

As a first step, governments should begin to rebalance tax and subsidy schemes so firms are rewarded for raising worker productivity rather than automating roles. In the longer term, they should also consider strengthening taxes on capital income as value in the economy shifts away from labour and towards capital.

Anyone displaced by AI should be equipped with skills that allow them to secure new employment, and workers should be signposted towards services which help them to navigate the disruption. IPPR has recommended that government should fund a new package of ‘portable benefits’ to support workers through any disruption caused by AI.

AI could be leveraged to strengthen economic growth, improve people’s experience of their jobs, and create shared prosperity for an entire generation. It could also worsen inequality, displace a significant number of workers and drive public discontent. The path we take will depend on the decisions taken now by governments and employers. 

For more information see: ippr.org

Joseph Evans is Research fellow, economy and environment team at IPPR