Office workers begin to "perform" AI efficiency gains: workplace oddities under Skill, weekly reports, and usage-rate assessments

Wen | Dingjiao One, Authors | Jin Yufan, Wang Hanxing, Lei Jing, Wang Lu, Chen Dan, Editor | Wang Hanxing

With just over two months left in 2026, big-tech workers are about to make it through another year dominated by AI.

At the start of the year, Jensen Huang's remark that "if an engineer earning $500,000 a year consumes less than $250,000 worth of tokens in a year, I would be deeply uneasy" set off the Tokenmaxxing wave. Everyone was using AI, everyone was burning tokens, and it seemed that if you used too few tokens, you weren't doing your job properly.

By June, big companies found they could no longer quite afford to burn tokens, and began focusing instead on the return on AI usage. This did not ease employees' AI FOMO (fear of missing out); it made it even harder for them to "please" their bosses through AI use. They had to prove that AI was helping them work more efficiently, that what they delivered had to contain AI, and that every token they consumed was producing value.

To that end, some people wrote up "having AI transcribe half an hour of the boss's meeting recording into text" as "building a structured voice-information workflow," and it was even rated an outstanding AI application case; some deliberately sought out unimportant work scenarios, turned them into Skills, and then packaged them under an impressive-sounding name to show the boss; and some big-tech PMs (product managers) used AI to do the work of RDs (research and development engineers), only to have their RD colleagues redo it afterward.

This phenomenon is now becoming increasingly common in workplaces at home and abroad. Not long ago, a survey involving the business intelligence company Visier showed that among 1,000 full-time U.S. employees surveyed, 48% admitted to exaggerating their frequency of AI use or their AI expertise to colleagues or superiors.

01. To put AI content in my weekly report, I turned a ten-minute task into half an hour

Li Zhao | Born 1992, Hangzhou, product operations at an e-commerce company

March to May was truly intense. Our group had a usage ranking, and admin exported the data and sent it to the group chat every week. Back then, using AI wasn't about getting work done; it was about not falling behind on the leaderboard. I asked the woman ranked first how she did it, and she said she had AI write her annual plan before bed and never even looked at the result. I copied her: as soon as I got to my desk, I had AI run "competitor monitoring," while I scrolled short videos and caught up on entertainment gossip, and that was how I muddled through.

In June the wind shifted: no more looking at usage, but at output. Weekly reports had to show "AI efficiency cases," with data. I started figuring out how to pad the numbers. The most ridiculous time, I wrote "having AI transcribe half an hour of the boss's meeting recording into text" as "building a structured voice-information workflow," and this was later rated an outstanding case. Another time, early this month, I was drafting a channel price-adjustment notice. It was originally a ten-minute job, but to put AI content in my weekly report, I had AI draft it, and it made up a lot of data for me. In the end I dug out last year's Double 11 notice, revised it a bit, and submitted that. But in the weekly report I still wrote "AI-assisted drafting."

Padding is padding, but the work still had to get done, and I gradually worked out a method. Two monitors at my desk: AI running on the left, work on the right. For proposal-type work, I throw in the brief and wait for a framework, without repeatedly adjusting the instructions; whatever it asks, I answer. While it runs the first section, I write the requirements for the second; when the second starts running, I go back and check the first section's results, and while I'm at it finish refining the third part, spinning through several hours at a stretch. AI really can improve efficiency. Last Thursday, with more than forty customer service complaints, I had it sort them by issue and produced a table in ten minutes. And for organizing meeting minutes: as soon as the meeting ended and everyone had just gotten back to their desks, I had already sent the minutes to the group chat.

But AI doesn't run on a fixed schedule; it could be thirty seconds or an hour. I tried using that time to do something else, but halfway through an edit it would finish, so I'd switch back to look at the result, then switch back again, and often forget where I was. Later I simply scrolled my phone, keeping an eye on progress while scrolling. During that period I clearly felt my focus declining; switching screens back and forth, I was exhausted by the end of the day but couldn't say where the exhaustion came from. I had to categorize my work: when doing A, I resolutely would not use AI to do B, and I saved small tasks like polishing for later, setting aside one or two hours a day to handle them in a batch.

02. The demo I wrote was unusable, but my boss was very satisfied

Li Mo | 27 years old, Beijing, product manager at a big internet company

I'm a product manager at a big internet company in Beijing, 27 this year, and I've worked at this company for two years.

At the start of the year the company began requiring everyone to do AI and write Skills. The first to be affected were colleagues in R&D roles, and later it spread to every position. Actually, I had used AI on and off before, writing copy and organizing requirements, which was at the "give it a try" stage.

Around March or April this year, our group leader talked to me, saying that in the future PMs who couldn't write code would soon be eliminated, and required all of us to learn to use AI to write code, to solve simple requirements ourselves as much as possible, and not to go to R&D for everything.

After that I did actually write some small demos and did some small UI designs. To be honest, none of it was really usable; when it was handed to R&D colleagues, they still had to redo it, so no matter how impressive my work looked, in the end they were the ones who cleaned it up. But the boss was very satisfied. During that period I clearly felt that reporting and delivering was easier than before; anyway, whether the thing could be implemented was secondary—the posture of "you are using AI" was established first.

After June, our group changed a lot, with considerable adjustments on both frontend and backend, and my reporting line also changed. The new leader made the assessment of AI output and efficiency gains much more detailed; the old approach of casually writing something myself and handing it to R&D colleagues to revise no longer passed muster.

So I switched to a different approach to "AI efficiency." I wrote steps like requirement feedback, market research, and product frameworks into Skills—basically just some very simple page reading and organizing—and had AI handle each stage, with the deliverables essentially coming straight from AI. There wasn't actually much of real value in all this; it was more about "performance."

For one thing, a lot of this work didn't need AI at all—doing it myself wouldn't have taken much time. For another, many of the requirements AI produced were hard for colleagues to actually push forward once they got them; they'd often find logic errors, and we'd have to hold another meeting to go over everything again. On the surface the process was running, but the actual efficiency wasn't necessarily higher than before.

Still, doing it this way had its benefits: the Leader thought we were very busy, the AI workflow was set up, the process data looked good, and the reports looked good too.

My own judgment on this is that AI really can boost efficiency—especially in writing code and writing documents, where it genuinely saves time—but it can't always boost efficiency, and forcing it on actually creates chaos. And personally, I think the biggest problem is that once the Leader thinks your efficiency has improved, they want you to take on other people's work. The benefits of the efficiency gains don't land on you; instead they become the basis for piling on more work, which is a losing proposition.

03. Under 70% AI usage rate and your performance review gets docked—we ground out a pile of Skills nobody uses

Xiaonan | Post-95s | Shenzhen | Event Planning

Later they even folded the AI usage rate into performance reviews: if the usage rate didn't reach 70%, performance pay was docked by 20%.

At first, nobody thought much of it—after all, AI really could boost our efficiency. Making an event planning proposal used to take at least 3 days. Now with AI, a first draft can be done within a day. And for tedious work like competitor analysis and PPT creation, AI can help too, genuinely saving a lot of time.

But gradually, things started to go wrong. The company launched an internal AI competition, seeing who could use AI to cover more work scenarios, build more workflows, and create more Skills. As long as you could prove you'd used AI to connect up some part of the work process, it counted as an achievement.

At first, what people made really was useful—for example, Skills related to event planning could reduce some repetitive, tedious work.

But valuable scenarios are limited, after all, so to complete the tasks, people started competing on quantity. Some specifically sought out unimportant work scenarios, made them into Skills, and packaged them with impressive-sounding names; others used AI to make a mini-program and, in their reports, showed off what problems they'd solved and how much time they'd saved.

I feel like everyone gradually developed a subtle tacit understanding. Compared with whether AI actually helped, being able to show visible results mattered more. After all, how many Skills you made and how many work scenarios you covered are easy to "fake," but how much time was actually saved and how many real problems were actually solved aren't so easy to measure.

One colleague used AI to quickly generate an event proposal, paired it with a polished PPT, and took it to report to the boss. The boss was very satisfied after seeing it, thought the efficiency was especially high, and immediately told the people below to push forward according to the proposal. But when it came to the execution stage, everyone discovered there were quite a few holes in the proposal and many practical issues hadn't been considered. In the end it still had to be patched up and adjusted bit by bit by people.

What left me even more helpless is that AI's existence has also made the number of revisions to proposals increase. Before, once a planning proposal was done, the leader would look it over once, give some comments, and after revisions it could move forward. Now the leader, after looking it over themselves, also hands it to AI for review. AI often comes up with a whole pile of revision suggestions—some genuinely valuable, but some not in line with the actual situation—and we have to check, explain, and revise them one by one.

I find it quite contradictory. The boss wants AI to help the company save time, but we have to spend extra time participating in competitions to prove we really are using AI. We finally use AI to improve efficiency, and then we have to take on more review, revision, and delivery tasks. Seen this way, using AI actually becomes a bad deal.

04. We put on a show for AI: clear articulation, and smiles that can be heard

Wang Min | Post-95s | Shanxi | Sales at a software development company

Five months into this job, I've seen what "AI not adapting to local conditions" really means.

The company's main business is software development. It's been around for over a decade and has about 100 employees. The boss is over sixty, and his understanding of AI basically comes from marketing accounts. Last year he proposed an AI transformation, and at the start of this year it was directly upgraded to an annual strategy. Over the past six months or so, some business lines have doubled their output after using AI, but more often than not, it's absurd.

So everyone started using their own tricks to "scrape together AI." For example, the day before yesterday I made a skill, yesterday I did a round of data cleaning, and today I did a full-lifecycle plan for a client from 0 to 1—the AI share got pushed all the way from 30% to 80%. Whether AI actually did any work doesn't matter; as long as it "participated," that's enough.

Before, weekly reports were written for the boss to read; now it's become a looping process: I send my thoughts to AI, AI writes the weekly report, I check it and send it to the boss, the boss uses AI to raise a few questions and sends them to me, I send them to AI, and so on and on. The thing I use AI for most each week is having AI make up what I did with AI this week.

He thinks employees typing away at computers making proposals all day is slacking off, and always says you can't type your way to results—you need to make more calls and communicate more. So an AI customer management system went live: when a call is dialed it records automatically, AI listens in, summarizes, gives suggestions, and generates a daily report when it's over.

Now employees on the phone at work look like they're "on camera." Say one wrong thing and the backend logs it; use a bad tone and AI gives suggestions; fail to close a deal with a client and the daily report is automatically flagged in red. The boss idly scrolls the app replaying communication records, and whenever he hears something he doesn't like, he screenshots it and tosses it straight into the big group chat. Every time we see a group message pop up, our backs go cold.

One day the boss came out of a meeting and said he wanted to build an internal AI bot, with employees at the center of an AI-driven transformation, to solve document sharing and boost efficiency. The HR colleague followed online tutorials and uploaded all the attendance, performance, and payroll spreadsheets, and it was built in a day. Then someone in the company group chat asked about salaries, and the AI just dumped out the details.

That set off an explosion. Some people discovered unequal pay for the same role, some found their performance scores had been suppressed, and others dug through the entire company's salaries and found three names that were still being paid but had never once shown up for work. To everyone's surprise, the boss didn't publicly criticize HR at the next day's summary meeting—instead he praised her for her high execution efficiency. But after the meeting, the boss contacted the AI company's technicians to add permission restrictions, and the system went back online.

AI is clearly a good thing, yet the more we use it, the more off it feels. We originally counted on it to save effort and boost efficiency, but instead we have to put on a show for it. In the end, what's exposed isn't a problem with AI—the payroll leak is because the company lacks basic common sense about data permissions. Employees putting on a performance of working is because the boss has turned AI into a new surveillance tool. AI amplifies management itself: if processes are smooth, it amplifies efficiency; if there's a lot of suspicion, it amplifies internal friction.

Once I understood this, I actually felt relieved. Even without performance reviews, I still can't do without AI—making PPTs, writing planning documents, organizing files, it really does save me a lot of effort on these odds and ends. But anything that requires professional judgment, I absolutely will not hand over to it.

05. What leadership wants isn't efficiency gains, it's the look of "AI for everyone"

Lin Zhou | post-00s | Guangzhou | software R&D engineer

A few days ago, a colleague I didn't know suddenly reached out to me, asking how to use a plugin I'd published on the internal plugin marketplace. I was actually pretty happy. After all, this was something I'd spent time researching, and someone actively using it meant it really solved a practical problem.

But my +1 (direct supervisor) didn't seem satisfied with just that.

I was a bit confused. The tool I'd figured out on my own really was useful, but everyone's work content and habits are different—why force everyone onto the same process? What was even stranger was that the veteran employees in the group who'd been working for more than five years didn't seem anxious at all. One veteran told me privately that he didn't really care.

Later, I figured it out. The core competitiveness of veteran employees lies in their business experience and professional knowledge; to them, AI is just an auxiliary tool—use it however is convenient, since it isn't tied to performance anyway. What's really tied to AI is the +2 who called for the department's AI transformation, the +1 who actively responds, and the unlucky new employees assigned concrete tasks—like me.

This also explains why +1 is so fixated on building SOPs.

Recently a few outsourced colleagues joined the department. I have a vague feeling that once this version's development wraps up, they'll probably leave. When the next project starts, new people will come in again. With turnover this frequent, the ideal management approach is naturally to standardize workflows so new people can get up to speed without much training. People can leave, but the skills stay.

I think the direction of AI development is personalization and customizability. In the past, people adapted to tools; now tools should adapt to people, and everyone can build their own workflow according to their needs. But +1 seems to think AI can turn all business into SOPs. To put it bluntly, I think his idea is half wishful thinking and half performance. How much efficiency is actually improved may not be the most important thing—what matters is letting superiors see that the whole department is undergoing AI transformation.

It's just that I originally thought I was helping colleagues reduce repetitive work, and only later realized I might also be helping the company reduce its dependence on employees. What's more subtle is that this doesn't conflict with my personal interests—it may even benefit me. I can learn new technologies, accumulate project experience, and gain real sense of achievement.

I still enjoy doing AI efficiency work; it's just that after suddenly seeing through management's possible motives, there's an extra layer of unease in my heart. Now I finally understand why some people say a high moral sense makes you unfit for this company. The most painful thing isn't being forced to do something bad—it's clearly doing something you love and discovering it also serves a logic you don't agree with.

*At the interviewees' request, Li Zhao, Li Mo, Xiao Nan, Wang Min, and Lin Zhou are pseudonyms.