AI at work: why employee monitoring may be creating new pressures

As artificial intelligence (AI) continues to make leaps, it has also kickstarted a race among India’s information technology-focused businesses to incorporate AI use into their business operations as rapidly and extensively as possible. Earlier this year, a number of Indian technology companies instructed employees to use and implement AI in their work as much as possible. Known as “tokenmaxxing”, this is the practice of pushing employees to maximise their AI ‘token’ consumption, based on the belief that high usage of AI will eventually lead to better business outcomes.

A token refers to a unit of AI use and can be whole words, parts of words (subwords), or even single characters. The “-maxxing” suffix comes from Gen Z slang like “looksmaxxing” or “sleepmaxxing,” meaning to optimise or maximise something. 

Tokenmaxxing took much of the tech industry by storm, with teams treating AI usage like a game and chasing the highest possible usage numbers. 

In Big Tech companies in India, employees found their AI token usage being tracked and ranked on internal leaderboards, with power users being awarded titles like ‘Token Legend’. At many companies, power users were awarded generous perks and incentives to encourage them to push their use of AI tools to the limit. 

Feelings of anxiety
Employers promoted the idea that the greater an employee’s use of AI, the more productive they must be.

However, for workers themselves, the battle to get to the top of the leaderboard – and avoid being ranked bottom – led to feelings of unease, stress and anxiety.

Commenting on the pressure they felt from their employer to use AI tools more extensively and the fears about job security that this created, a tech worker stated on Reddit: “Management at my company has set up leaderboards to track who is using the most tokens, and there is speculation that bad things may happen to you if you aren’t in the top 25%.” 

Meanwhile, a tech executive at a Hyderabad-based startup said: “I know of a coder who probably spends more than his salary on Claude [Anthropic’ agentic coding tool for IT, tech and software developers]. Of course, his employer pays for his tokens.” 

Many tech workers argue that tokenmaxxing simply involves using AI in pointless tasks in a bid to comply with employer mandates to increase AI usage. They also argue that token counts do not measure the actual business value, or effectiveness, of using AI at work. “Inside large tech companies, it’s becoming a career risk to not use the technology at an accelerated pace, regardless of output quality,” observed one tech worker. 

At Google offices in India, an internal dashboard is now reportedly used to track how developers use AI tools, but the data is not treated as performance benchmarks, according to a report in Analytics India Magazine.

“There is no productivity benchmark. It’s more like just to see how much engineers are using AI,” a Bengaluru-based staff engineer at Google told Analytics India Magazine. “Just to visualise how many people are using it and what the token usage is.”

Rapid pace of adoption of AI
The rapid rise of AI is also reflected in the pace at which organisations are adopting the technology.

A 2025 report by Amazon Web Services (AWS) revealed that as many as 83% of organisations in India have appointed a dedicated AI executive, namely a chief AI officer (CAIO), to speed up the adoption and manage the implementation of generative AI innovations. 

Overall, to date, the scale of adoption of Generative AI (Gen AI) in Indian businesses has been on par with their global counterparts, with 98% of Indian organisations using Gen AI tools and 95% actively experimenting to unlock new applications, according to research. 

Satinder Pal Singh, head of solution architecture at AWS India and South Asia, said: “For India, this is not merely about technology adoption; it’s about cultivating a mindset of continuous innovation and harnessing AI as a strategic lever to reimagine customer experiences and rearchitect how work gets done.”

But are any of these tokenmaxxers producing anything worthwhile for the business, or are they simply keeping up with the times, as (they predict) the era of human coding appears to be coming to an end?  

“In the absence of true metrics, organisations created [AI token] usage leaderboards, which people quickly learned to game,” Neil Dhar, SVP for IBM Consulting, wrote on LinkedIn. “Usage soon became a proxy for value.”

Increasingly, companies are sharing their experiences about what happens when AI token usage is employed as a productivity metric. Pushing tech developers to consume as many tokens as possible can encourage AI adoption, but it also opens the door to enormous expenditure on AI.

So, when technology companies that have encouraged high levels of AI token usage received the resulting invoices from AI tool providers like Anthropic and OpenAI, the amounts owed were huge. As a result, Microsoft has reportedly decided to cut most of its Claude Code subscriptions amid financial concerns. Uber, meanwhile, reportedly used up its entire 2026 AI budget in four months due to high AI usage, and its chief operating officer is now apparently questioning the value of high usage of AI tokens. An AI consultant told Axios news channel that one of their clients was recently left with a bill of half a billion dollars in a single month after failing to put usage limits on Claude licences for employees.

However, this doesn’t necessarily mean that all companies will stop spending large amounts on AI. Meta reportedly told employees that the company was on track to spend billions on AI use this year, but wanted to “find places we can spend less while getting similar or better business results”. 

Uber’s chief technology officer Praveen Neppali Naga’s recently stated that the company has shifted its strategy from simply consuming AI services to systematically embedding ‘Agentic AI’ – autonomous AI systems powered by large language models that can set goals, make multi-step plans, use external tools, and execute actions with limited human supervision – into the core of its business, achieving a level of adoption where 99% of its engineers now use AI tools.

According to IBM, the goal for effective deployment of AI in business should not be token maximisation or token minimisation. “The goal should be ‘valuemaxxing’, a term coined recently by Marc Boroditsky, the chief revenue officer of Nebius,” IBM recently stated. “Valuemaxxing moves the conversation toward direct outcomes. Instead of asking, ‘How many tokens did we use?’, organisations should ask: How many tasks were completed? How much developer time was saved? How much modernisation work was accelerated? How many vulnerabilities were resolved? How much rework was avoided?” 

In a post on Google’s DevOps Research and Assessment (DORA) website, Evan Conaway, lead UX researcher at Google, stated: “Gamifying token consumption might give you a temporary spike in visible activity, but it comes at the cost of technical debt and developer burnout. True, sustainable performance comes from empowering teams to collaborate, measuring holistic outcomes, and finding balance in how we build. It’s time to leave ‘maxxing’ behind.”

Impact of AI use on employment
However, the debate about the value of using AI to improve business processes and outcomes stretches beyond productivity and costs. The impact of AI use at work on employment more generally is becoming an equally important question, particularly in India, which has a large and rapidly evolving workforce.

In a post on LinkedIn, former union minister Suresh Prabhu wrote: “Every new #technology, right from the First Industrial Revolution, has brought disruption and #job losses. But history also shows that new technologies create substantial gains in other sectors, and over time the benefits have often outweighed the initial pain. #AI, however, may not necessarily follow the same pattern in terms of creating more jobs than those it displaces.”

He added: “#India is particularly vulnerable to emerging technologies, especially AI. The reported job cuts at Oracle are a clear example. The company has already undertaken significant #workforce reductions and is now reportedly planning another round of job cuts, while making massive #investments in AI infrastructure.”

He further stated: “We must welcome new technology and the #progress it brings, but we cannot ignore its impact on employment. A thorough assessment of how emerging technologies will affect jobs needs to be undertaken at the earliest, followed by necessary #policy measures and action. 

“At the same time, we need to identify the sectors where new jobs will be created and prepare our people for them through #training and #reskilling. Technology will continue to change the way we work. Our responsibility is to make sure our #people are ready for that change. 

“Technology is also changing how employers monitor and measure work, adding another dimension to its impact on employees.”

Monitoring work
Commentators say that the rapid adoption of AI by business does not merely raise questions about whether the technology will create or displace jobs; it also means that the use of AI in the workplace is changing how employers monitor and measure work.

For instance, Tata Consultancy Services (TCS) has allegedly deployed a Digital User Experience Monitoring tool on company-issued laptops used by its workforce, according to a report by Moneycontrol news website.

The reported move has triggered discussions about whether India’s largest IT services company can maintain an appropriate balance between meeting cybersecurity requirements and ensuring the monitoring of employees using AI remains reasonable and fair, and does not have a negative impact on worker wellbeing.

Sources cited by Moneycontrol said the tool enables TCS to view information such as applications accessed on employee devices, such as computers, and the time spent on those applications. The deployment is reportedly being viewed internally as both a measure to strengthen data security and a mechanism to gain greater visibility into employee activity on company-owned devices.

Following the news report by Moneycontrol, TCS shared a statement with People Matters, a news website for talent and HR professionals in Asia, rejecting allegations of employee surveillance.

“The reports regarding surveillance of employees are baseless and inaccurate,” said TCS. “TCS does not track individual employee activity on laptops and respects their privacy. We use tools to monitor macro-level network performance to ensure security, availability, and enhanced user experience for our employees.”

However, although TCS rejected allegations that it is using an AI-powered tool to conduct surveillance on employee activities, Indian multinational technology company Wipro is reportedly rolling out a productivity-tracking tool on company-issued laptops and desktops in India, according to the Indian business and financial news website Mint. According to the Mint report, the tool provides employees and managers with a dashboard showing digital activity and time spent on work applications.

The software, called TimeScope, allegedly identifies periods of inactivity and may trigger a deduction in annual leave if there has no device login or other attendance record for a scheduled working day, according to the Mint report. 

Health and safety risks
Psychologists warn that monitoring tools of this nature can place pressure on employees to work faster and harder, which may pose “serious health and safety risks” for workers. Other commentators say the deployment of AI-powered monitoring tools raises thorny questions about where and how employers draw the line between maintaining employee productivity and intrusive surveillance of their teams.

In a post on LinkedIn, Neeraj Gour, lead analyst at Wipro, said: Accountability has to be a two-way street. If a company automatically penalises an employee with forced leave for being inactive during regular hours, they must be equally diligent in compensating that employee when they are active during ‘inactive’ hours, like weekends or late nights. Transparency and tracking shouldn't just work in the employer’s favour – if you track the gaps, you must also honour and pay for the extra miles.”

Another individual commented on LinkedIn: “Employees are professionals, not prisoners. Accountability and performance expectations are perfectly reasonable – but micromanagement, excessive control, and treating employees as though they need to ‘prove their innocence’ every day is not a healthy workplace culture. 

“Trust goes both ways. Respect goes both ways. You cannot build a high-performing organisation by hiring talented professionals and then treating them as if they are untrustworthy by default. Leadership is about creating accountability without compromising dignity and trust.”