
Why AI tools aren't creating more free time at work
In 1930, economist John Maynard Keynes optimistically predicted that humans of the future would need to work only 15 hours a week because, in the next 100 years, technology would have automated many tedious and time-consuming tasks. Nearly a century later, the fantasy of 3-hour work days hasn't come true, though many companies around the world are opting for 4-day work weeks thanks to improved efficiency in their operations due to automation. The flip side of this trend is, of course, a surge in layoffs as well as proportional reduction in the salaries of employees working shorter shifts to maximise the company's topline.
In 1930, economist John Maynard Keynes optimistically predicted that humans of the future would need to work only 15 hours a week because, in the next 100 years, technology would have automated many tedious and time-consuming tasks. Nearly a century later, the fantasy of 3-hour work days hasn't come true, though many companies around the world are opting for 4-day work weeks thanks to improved efficiency in their operations due to automation. The flip side of this trend is, of course, a surge in layoffs as well as proportional reduction in the salaries of employees working shorter shifts to maximise the company's topline.
Even where an artificial intelligence (AI) mandate is yet to be formally adopted at the organisation level, AI tools are being widely used by individuals to reduce the drudgery of their daily grind. For perspective, a recent report shows, India is the largest user of ChatGPT, surpassing the United States and Indonesia. Time that was once spent on executing repetitive tasks, like workflow management, has been freed up, allowing employees to focus more on strategic goals. The logical conclusion, based on these shifts, would be that AI's transformative hand is improving our work-life balance. If you have chunks of time freed up from your schedule, you would be most likely spending it with family or friends, or whatever activity gives you joy, right? Quite wrong, as it turns out. Also Read | Overcoming accent bias in the global workplace
The 2025 Work Trend Index Annual Report recently published by Microsoft shows, based on the company's telemetry data, that the notion of the 'infinite workday" continues to thrive all over the world. If you thought early risers were fiercely protective of their morning routine, that's not entirely true: 40% of people are online at 6am, checking their emails. On an average, employees receive 117 emails, which most skim in under a minute, and 157 messages on Teams, each day. That's not all.
Data showed 57% meetings were called without any prior intimation, and average time between interruptions by a meeting, mail or message during core work hours was 2 minutes. About half of these meetings were called between 9am-11am and 1 pm-3pm, which are peak productivity hours for most people, according to a wide range of research into circadian rhythm. Due to this fragmented focus through the day, along with attendant distractions on social media and the internet, the evenings organically fuse into the 'workday". And so, 29% people log back into their inbox by 10pm on weekdays, and around 20% actively work over the weekend. Work Without End
According to data tracked by scholars Wei Jiang, Junyoung Park, Rachel Xiao and Shen Zhang between 2004 and 2023, workers in AI-intensive occupations had more hours added to their work week relative to those in less AI-exposed jobs. In an article published on the Centre for Economic Policy Research (CEPR) website in March, they showed that an increase from the 25th to the 75th percentile in AI exposure added over 2 hours to the work week of employees.
So what explains this conundrum, especially in heavily AI-exposed jobs, which are supposed to increase efficiency and create more leisure time?
As the CEPR article argues, by increasing worker productivity, AI-exposed jobs ironically end up incentivising longer work hours. It's a case of market economics colluding with the innate human desire for more. As automation takes away the cognitive load of mundane labour, employers begin to set higher benchmarks of productivity for their employees. Individual goals are reset, the metrics of success are amped up, pushing workers to work ever longer hours to achieve their targets. Especially for those incentivised by productivity-linked pay, working longer becomes a way of making more money. And who wouldn't want to make some extra cash in the midst of this cost of living crisis? Also Read | Firing underperformers can boost your team's retention
The second fallout of the AI revolution in the workplace is the increased surveillance of workers. Bill Gates famously memorised the license plates of cars belonging to his employees to keep track of the time they came into, and left, work. Now all you need is an AI tool to monitor how your employees are spending their time, especially when they are working remotely or in a hybrid setting.
In many AI-intensive roles, employers are rolling out automated performance scores to provide real-time feedback to workers—not only on how well they are doing their tasks but also where they stand vis-à-vis their peers. It's back to school again except, instead of a human, you now have an intelligent machine hovering over you, rapping you on the knuckles, and making you feel miserable for being not as good as your classmates.
The outcry against the so-called AI apocalypse has largely focused on layoffs so far. But even for those who survive this tech revolution, there are ever new worries that lurk on the horizon. While technological strides may give them a spike in wealth in the short term, it won't do any good to their well-being in the future.
Work Vibes is a fortnightly column on ideas to help you thrive at what you do. Also Read | Instead of setting SMART goals, why not make a PACT? Topics You May Be Interested In

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Hans India
14 minutes ago
- Hans India
ChatGPT Go Plan Launched in India at ₹399 With GPT-5 Access
OpenAI has launched a new plan in India, and it is called ChatGPT Go. The price is ₹399 per month. This plan gives access to GPT-5. It also allows for more image creation. Users can also check more documents and files. What subscribers will get with ChatGPT Go: Subscribers get higher message limits They get more image generations. They get more file uploads. They get a longer memory than the free plan. OpenAI said limits may change sometimes. This is to keep the app smooth for all users. The Go Plan is now available for subscribers To subscribe: Log in to ChatGPT. Click on the profile icon. Select 'Upgrade Plan'. Pay using UPI or a credit card.


Mint
14 minutes ago
- Mint
Sam Altman says there's an AI bubble. What Wall Street thinks.
Big technology stocks are pricey, leading Wall Street to a debate whether artificial intelligence stocks are overvalued. Tech earnings have given investors a lot to celebrate over the last couple of weeks. Hyperscalers such as Microsoft, Alphabet, and posted better-than-expected earnings growth. The AI giants, including Meta Platforms, also committed to spending billions more on AI infrastructure in the months ahead. 'These huge investments support earnings because they are revenue for someone, and they help drive productivity gains and boost profit margins—not just for the technology companies but for all of corporate America," wrote Jeff Buchbinder, chief equity strategist at LPL Financial, on Monday. As of Friday's close, the S&P 500's market cap gain for the year was $4.865 trillion, according to Dow Jones Market Data. The Magnificent 7 has contributed roughly 43% of that. The market cap value of the S&P 500 as of Friday's close was $54.67 trillion, while the market cap for the Mag 7 was $19.68 trillion, or 36% of the S&P 500's market cap. Valuations are also high. The S&P 500's forward price to earnings valuation is 22.5 times. Six of the Mag 7 stocks are trading at higher multiples than that. Nvidia is trading at 34.9 times earnings expected over the next 12 months; Microsoft is at 32.7 times; Apple at 29.6 times; Amazon at 31.7 times; Meta at 26.1 times; and Tesla at 151.6 times forward earnings. Apple, Meta, and Tesla are also trading above their five-year historic averages. Other tech stocks are pricey, too. Netflix is trading at 41.2 times forward earnings, Oracle at 34.5 times, and Broadcom at 38.2 times. According to an article from The Verge on Friday, OpenAI Chief Executive Sam Altman told reporters that he thinks the AI market is in a bubble. 'Are we in a phase where investors as a whole are overexcited about AI? My opinion is yes. Is AI the most important thing to happen in a very long time? My opinion is also yes," Altman said, according to the article. Barron's has reached out to OpenAI for comment. On top of concerns that near-term expectations for AI stock returns are too high, the broader economic environment is uncertain. The July jobs report was a disappointment, and May and June revisions to the downside were a shock. Wholesale inflation came in hotter than expected for July, and the most hefty of President Donald Trump's tariff policies have recently started to go into effect. Wall Street is now looking to the Federal Reserve's monetary policy symposium this week, where Chair Jerome Powell could possibly offer a more hawkish view on rate cuts than investors might want to hear. This puts highly valued tech stocks at risk. 'If everybody is fully invested, or nearly fully invested in a small cadre of very highly valued stocks, who's there to buy them if they hit an air pocket or sentiment changes?" Steve Sosnick, chief strategist at Interactive Brokers, told Barron's on Monday. He added that it's like if there's an 'overcrowded building and everybody needs to rush to the exits, people get trampled, and that is the risk here." Not everyone on Wall Street is concerned. Richard Saperstein, chief investment officer at Treasury Partners, wrote on Monday that 'big technology stocks have led the market higher and will continue to dominate market performance. We expect continued earnings growth, reinvestment of cash flows and expansion of their globally dominant footprints." Saperstein noted that investors should remain fully invested in U.S. stocks with a concentration in large-cap technology stocks. 'Stocks will likely benefit from deregulation, onshoring and capital expenditure expensing which will eventually become a tailwind for economic growth," Saperstein added. Write to Angela Palumbo at


Time of India
17 minutes ago
- Time of India
'Saying AI will replace developers is like saying calculator will replace CAs': Techie's post goes viral, others say learn prompt engineering
Amid mounting reports of layoffs in the IT sector, a data engineer's Reddit post has gone viral after challenging the narrative that artificial intelligence will soon replace developers. 'Saying AI will replace developers is like saying calculator will replace CA,' the engineer wrote, dismissing the fear-mongering around AI gobbling up tech jobs. The engineer argued that AI would first replace unskilled, non-IT roles before making a dent in skilled developer jobs. 'I see a lot of propaganda by delusional people saying AI will replace IT jobs in the future. If that were true, AI would first replace some unskilled non-IT jobs. As a data engineer, I just use ChatGPT to clear doubts—like we once used Google,' the post read. by Taboola by Taboola Sponsored Links Sponsored Links Promoted Links Promoted Links You May Like Join new Free to Play WWII MMO War Thunder War Thunder Play Now Citing personal experience, the engineer said current AI tools are limited. 'Whenever I put a client requirement in ChatGPT, it often goes blank. Even for coding, it gives only basic code. It can't understand client-specific environments, which differ every time,' they noted. The post quickly drew reactions from fellow techies. Live Events One quipped: 'Being a data engineer and not being able to extract intended data from an LLM is astonishing.' Another suggested he improve his prompt engineering: 'Buddy, ChatGPT isn't the best AI tool for your use case. Try Claude Code, Roo with Claude 4 API, or even Kiro. The tools are evolving, but so should we.' A third added: 'It's not AI, it's you. Remember—AI won't replace you, but people who know how to use it will.' Some, however, agreed with the original post, stressing that while AI can support developers, it still lacks the depth and adaptability of human expertise. Meanwhile, the broader IT industry is reeling from large-scale layoffs. Tata Consultancy Services (TCS) cut 12,000 jobs in July, Microsoft has axed 15,000 so far this year, and Intel is preparing to slash 15–20% of its workforce—impacting nearly 10,000 employees. Similar cuts are unfolding in automotive and manufacturing. According to Indeed Hiring Lab , tech job postings in July 2025 were down 36% compared to early 2020. While AI and machine learning roles remain in demand, traditional and entry-level tech jobs are rapidly shrinking. Brendon Bernard, senior economist at Indeed, warned that many positions are 'losing steam,' urging job seekers to adapt and rethink career strategies.