AI speeds up work Breezes into offices
AI speeds up work
Breezes into offices
AI speeds up work
but apparently seems to be slowing down mental wellbeing
As AI breezes into offices, it not only ensured speed and smarter decisions, but it also left many workers anxious, tired and unsure of who they will be tomorrow. Is this the human cost of ‘endless learning’: can kinder leadership and wiser policies stop a generation from burning out?
Is this the human cost of ‘endless learning’: can kinder leadership and wiser policies stop a generation from burning out?
For many employees, though, the promise has come with a bruise: the constant pressure to learn new systems, the fear of being replaced and the loss of the confidence that comes from mastery.
This is not just a training problem, it is a human problem.
Across India and beyond, anxiety is spreading quietly through workplaces. From coders and designers to teachers, marketers and analysts, millions of professionals now spend part of their day not just doing their jobs, but learning how to keep them.
Conversations once centred on promotions or pay hikes now revolve around upskilling courses, AI certifications and fears of redundancy. For many, the excitement of innovation has been replaced by an undercurrent of unease, the sense that progress is outpacing people.
A 2025 Acta Psychologica paper (Chen et al., 2025) found that anxiety about AI, fear of obsolescence and relentless learning pressure, reduces workers’ passion for their jobs and increases emotional exhaustion.
The study described how people who believe “no amount of effort will keep you permanently relevant” feel drained and disengaged.
As per Ankur Agrawal, who has worked with technology teams, sums it up: “Mental health: people are facing a triple whammy here: pressure learning new tools while their roles evolve, constant anxiety about obsolescence, and the erosion of mastery achieved with years of learning and practise. This is disrupting people’s professional self-identity and confidence in their own competence.”
He called the feeling ‘treadmill anxiety’, running hard but not getting further.
Numbers back up the unease.
The OECD report ‘Using AI in the Workplace’ (2024) found about three in five workers worry they could lose their job to AI in the next ten years, and two in five expect AI could push down wages in their sector.
McKinsey’s ‘State of AI’ work (2024–25) shows many organisations are gaining value from AI, yet only a few have paired adoption with strong worker support or change management.
A large global survey led by KPMG and the University of Melbourne (2025) asked nearly 48,000 people about AI use at work. It found 57% of workers hide their AI use from managers and only 47% report receiving AI training.
Alarmingly, 66% do not verify AI outputs for accuracy. The Institute for the Future of Work also found that heavy technology roll-outs can harm quality of life when they increase surveillance, workload and job insecurity.
Leaders need to re-think re-skilling;
Ankur Agrawal insists leaders must stop treating AI anxiety as an individual failing.
“Leaders must move beyond treating AI anxiety as an individual problem and recognise it as an organisational design challenge,” he says.
He urges managers to create environments where people can admit confusion, ask ‘stupid’ questions, and opt out of tools without penalty. Empathy is vital, but it must be combined with structural change, such as making upskilling time part of the working day.
Companies that rush technology in without clear role definition, training and mentorship risk hollowing out professional pride.
Instead, organisations should choose which tools to adopt quickly and which to pilot more slowly, keeping humans in charge of judgement and ethics while using AI for speed and volume.
There are questions about scale. Can firms realistically slow adoption? Not entirely, competitive pressure will continue to push rapid implementation.
advertisement
But organisations can choose which tools to adopt quickly and which to pilot, and they can prioritise measures that reduce cognitive load: clearer role definitions, fewer simultaneous new platforms, and scaffolding learning plans.
Further according to Agrawal, “Ultimately, they need to resolve this basic conflict of treating a design problem (how work is organised) as an individual adaptation problem (why can’t workers just cope?).”
Allowing teams time to grieve a loss of mastery, to relearn at a human pace, and to recliam professional dignity may be the single most effective antidote.
Organisations can take these steps now to protect employees while benefiting from AI:
Pace adoption. Pilot tools with small teams, learn, then scale.
Paid learning time. Make upskilling part of work hours, not extra.
One change at a time. Avoid multiple platform rollouts simultaneously.
News Edit KV Raman
