As more and more dipshit managers mandate (or at least heavily encourage) the use of AI in the office, more and more workers are being confronted with a choice: do as they're told and help contribute to an economic, environmental, creative and cognitive disaster, or find ways to resist. Thankfully, plenty of folks are opting for the latter.
Luke Plunkett
Having heard an increasing number of anecdotal stories over the last year or so of people finding weird, wild and defiant ways to avoid using AI in their professional lives, tales I've seen told in fragments across social media and platforms like Linkedin, I decided to speak directly with as many workers as I could to hear their stories in full.
What I found particularly interesting was that while some employees are able to duck out from using the technology altogether, a surprising number of people without that luxury are still finding ways to minimise, misdirect or even sabotage their company's AI mandates by using the technology against itself.
Given everyone I spoke to was still employed at the place they're talking about, they've requested their names and employers be withheld. And while (since this is a video games website) there are some video game industry workers featured below, I also ended up speaking with many affected employees across a range of other industries, from utilities to finance to cloud hosting.
"I work for a company that produces luxury items", one worker tells me. "Any time I get prompted that my 'token spend' is too low (0), I go in and ask ChatGPT random gardening questions. So far it has failed to adequately build a rain garden, but I’ve gotten it to describe a sweet potato once."
A programmer at a small European studio, known for working on licensed games, says their CEO's rapid adoption of the technology has led to issues with their shipped projects. While their own team has been given an “opt-out” by their direct manager, so they aren't using the technology at all, they say a recent release's dialogue and text "was AI-translated, and the quality was so bad humans had to redo it, and it ended up costing more than just doing it by hand [in the first place]".
A team lead at a company that's a "competitor to AWS" says that upper management has begun telling employees they need to "find ways to integrate AI into our daily workflow", without actually providing specifics as to how, or why. It has since become a required objective for their direct reports.
"I have made it absolutely clear to anyone reporting to me that I will accept literally even the smallest thing, and I don't want anyone to actually start heavily using AI", they tell me. "If they lie to me saying they used AI when they didn't, I will happily turn a blind eye and still give them full marks on their yearly objective. My managers are on the same page."
Most cases so far have proven that there is too much risk in using AI unless the client pays for a service, so we just go back to doing things the old-fashioned way. Not sure how long it will last, but vive la résistance for now!
"When it was announced that we would start recording all meetings so that an AI assistant could take notes I was pretty concerned", a developer at a Western AAA studio tells me. "I decided to stop attending meetings until I could receive a formal statement from the company about how they would and more importantly would not use our recordings to train AI. I ended up skipping meetings for several months."
"There was a mandate for every employee to incorporate AI into their workflow and put it in official writing how they were going to do that. I wrote something sarcastic and ignored it". They were then forced to sign paperwork stating that all employees were required to use generative AI tools, which they say they also ignored.
"A lot of AI has been going into our tooling and I've been in a lot of meetings where I'm the only voice against it both morally and qualitatively. The vocal majority is non-critical, and has accepted AI as a standard part of their workflow. I see a lot of people on BlueSky saying that AI use is abnormal in the game industry, but from my experience and what I've heard from many contacts, it's becoming standard in nearly all Western AAA studios."
A software technician says, "My bosses have gone all-in on AI, and it’s immensely frustrating. Often they give me instructions on a task and the last step is 'run it through Gemini' and I just find some other way to do it and don't tell them I haven't been using it."
One example they provide is a task they were assigned, asking them to compile data from client websites and use AI to "add up the numbers". "No joke it was just straight up faster and way more accurate to copy that data into Excel and run two 'text-to-columns' functions to break out just the numbers I needed", they say. "Excel is a magical program and the fact that a lot of people's defaults for, essentially, a calculator is now 'the computer that is bad at math' is baffling to me."
"We've recently had an influx of 'agentic solutions' for basic tasks, ranging from annoying to counterproductive", a games studio producer says. "Generally speaking, contractors will try the tool so as to not rock the boat in this perilous economic environment, but a team recently got forcibly introduced to an agent that takes 600% longer (not an exaggeration) than their old tool to complete a task, among other sluggish changes".
To get around this, someone on the team figured out that they could just ask the agent to...take them back to their old tool. But even that discovery had to be kept secret from their bosses, requiring espionage-like workarounds to share the knowledge (like making direct phone calls), since their keystrokes are being tracked and their video calls monitored by AI agents.
A worker at a games industry support studio says that "While my company does not force us to use AI (since we work with lots of different IPs we don't own), when someone in Slack mentions using AI I react with 👎🤖 🤮 which has caught on with co-workers."
A video game artist says that while both clients and bosses are "very hyped" with the technology, it's hard to make them understand that they personally don't want to use AI in their workflow. But rather than counter with ethical concerns, they're taking a legal approach, telling management that they're risking fines for breaking copyright laws, endangering the business by allowing clients to be provided sub-par work, risking confidential content by feeding it into AI, wasting money on tokens and clogging up art pipelines because they work faster without using AI.
They say this approach, acting as a consultant would and framing the issue in terms that business people understand, "works like a charm".
I built lots of inane, rambling prompts, like having the AI write (bad) book reports. I did all the real work myself, but my usage showed up on the dashboards and made me an 'AI Champion'
After "getting a ton of pressure at work to start using AI", a marketing writer for a tech company says, "I just started describing my normal workflow to my bosses as if AI were part of it".
"I keep a to-do list in a paper notebook, but my boss believes I have an AI task-keeping assistant via Claude. I no longer say I Googled things (I actually Kagi things), I say I asked Claude to explain it. When I copy-edit writing I take an extra twenty minutes to double check, then say I had Claude copy edit it as well. Everyone’s really happy with my performance".
A manager at a tech startup (bleeding cash because of the rising costs of using AI) is another who was forced to use AI by employers who had absolutely no idea how or why, exactly, worker's AI usage was going to be tracked.
"As a manager, I knew that there were internal dashboards at the company showing AI usage, so I built lots of inane, rambling prompts, like having the AI write (bad) book reports. I did all the real work myself, but my usage showed up on the dashboards and made me an 'AI Champion'", a title for which they received both company-wide recognition and a cash prize.
"No one had ever bothered to check if the things I built on a daily basis were AI generated".
A worker at a major AAA games studio says that while AI use hasn't been mandated, things are becoming increasingly dire. "A couple of weeks ago we had some disastrous 'AI vision' meetings", they say, "where people who were pushing back against the use of AI got really agitated, and it led to at least one person being let go from the company, and a few other got a talking to from HR".
"Some of us are vocal about not wanting to use AI tools, and we've basically held our ground by asking for meetings to not be automatically transcribed or have notes taken by AI agents. But it definitely feels like we're swimming against the current".
"Externally and internally, upper management is pushing hard on AI, but around our team adoption is slow at best", a cybersecurity worker at a major entertainment company says. "Infosec management hates it, and resents the amount of time they need to spend on it. They're being told to find use cases for AI, but none really exist".
When they're forced to use the technology, they engage in 'malicious compliance', where they use AI exactly as it has been requested by management, and will submit the results directly and unedited, including every inaccuracy and inconsistency. When management points out these issues--which can be serious when it comes to things like incident reports– "we point out that we did exactly what was asked of us, and they get real quiet".
"Because I work in cybersecurity, I also have free rein to use the company's internal AI as I see fit, and so far that's been showcasing to others all of the ways it can be abused, to the extent where it's clear it can't be used for anything even remotely useful".
A senior artist in the video game industry says that "at every step I've made it extremely clear that I won't use it, will never use it and have given all of my reasons why; to the point where my boss became a little upset about it all. I have a zero-tolerance policy towards it because I know that if I give it an inch, it'll slowly work its way into everything."
They say that "certain folks" higher up at the studio keep trying to implement AI tools into the artists’ workflow "for the sake of it", even when--as we've seen before when it comes to AI and art--there's no problem to solve.
"They always resort to the 'time is money' argument, and that they need to speed up production", they say. "Thankfully, I've been able to keep firm about my position, as well as safeguarding my role and those of my fellow artists, while maintaining the integrity, authenticity and accuracy of our art intact".
"Anyway, things are looking up for now, I think leadership are starting to see things my way!"
"At work they've been pushing AI use for the last two years, and this year it was shared that bosses will be looking at AI usage as part of our performance evaluations", someone in the IT department of a utility company says.
"I did some looking into how they're evaluating that, and they pull numbers for submitting prompts.... but they don't collect the prompts."
"Which is why earlier today I made sure to ask Copilot 'What year was Star Wars released?' And yesterday I asked Copilot the weather. Essentially, I make sure to prompt at least once a day, but the few times I've ACTUALLY tried to use it for work, it either is extremely basic and unhelpful. Or it's flat out wrong".
A worker at a games localisation studio says that while some of their colleagues have embraced AI usage, "using ChatGPT/CoPilot to write all their emails and effectively sound like a machine in all their responses", the majority of their team keep their use to a minimum, preferring to "remain human, typos and all".
"The best tool I’ve found to resist enforced AI use on projects is to raise the security concern over feeding sensitive materials to AI", they say. "Our company doesn’t want to invest in their own AI instance for every project, so we warn employees to not include project-sensitive materials and to ask leadership for permission before they use AI for any of their work".
"Most cases so far have proven that there is too much risk in using AI unless the client pays for a service, so we just go back to doing things the old-fashioned way. Not sure how long it will last, but vive la résistance for now!"
A veteran comms director says that while their workplace doesn't have a mandatory AI usage policy, the expectation has been set by leadership that everyone on their team needs to "be open to, and work to implement, AI-assisted workflows moving forward".
"Over the course of the last six months, there's been a sharp increase in vibe coding by non-developers, and some members of leadership have praised the increase in AI adoption across the company. As you can imagine, it's been a polarizing issue", they say.
Handily, given they work in comms, they've been able to "lean into soft skills", and have provided a breakdown of the ways they've able to object to the use of AI at the office
When leadership presents documentation that was clearly (or majority) generated by AI, I point out inaccuracies in real time. This is most effective when done in a setting with multiple people (Slack channel/group DM/meeting).
When I share a draft for comment, I disclose that it was not generated or modified by AI, and I ask my colleagues not to run it through AI as a part of their review process. There's no way to know for sure if they don't, but at the very least I'm establishing a clear boundary, and hopefully making my colleagues pause before dumping my work into Claude, GPT, or whatever.
My favorite: just ask why. If a colleague or manager suggests that something can be easily achieved by using GenAI, a simple "why do you think that?" can really derail the conversation. More often than not, they have a flimsy response and realize it's not more efficient or producing a better outcome.
It ultimately saved us time to just sit together and make them right the first time, without AI...
A worker at an American games studio says that after fumbling along for a few years their company finally implemented an AI policy in 2026, which "strongly encouraged" the technology's use, and asked employees "not to shame" those who chose to make use of it. Which management hoped would be everyone, because despite the policy falling short of enforcing AI's mandatory use, workers are also being told that they "have to be open-minded if they want to stay at the company".
"I've embraced professional hater status by trying to educate any co-workers who will listen to me, including senior ones," they say. "I am angry enough to keep fighting; not just for my own job, company or even the games industry. AI is a rot at the heart of the world right now".
"I just desperately hope the bubble bursts before all these companies destroy themselves chasing the fantasy of 'number goes up forever'".
More tips! A data scientist at a consumer tech company, working higher up in management than most others I spoke to, was able to share their personal tips on dealing with executives and others obsessed with a seeming need to employ AI, without actually being able to explain why.
- Relentlessly point out how the "positive" examples of AI use internally are not generalizable. "Pilot A worked because the project was simple, and you showered them with extra support and attention AND took away all their other obligations for the duration and shifted them to others. That doesn't generalize."
- Change the conversation from "lines of code produced" to "does what we're building actually matter to anyone?" We supposedly are producing a lot more code (not actually true – we've looked at the data every which way, but everyone THINKS this is true), but nothing has observably changed about the overall business. This is causing some serious crises of confidence in engineering leaders, because it is shining a light on the reality that more 'features' does not mean more money, and that actually they are not necessary for growth.
- Try to appeal to their professional pride. "Do you want your job to look like a factory role where you have a code quota and are told precisely what to do and how to do it?"
- Fight against the inclusion of AI usage mandates in performance reviews. For example: "We have to have an independent perspective on whether people are performing their jobs well, otherwise you will not be able to fire someone who burns a lot of tokens, and you'll be forcing your top team members into performative AI usage just to avoid consequences. Your jobs as managers will get very stupid if you start telling people there is only one acceptable way to do the work, and it's impossible to do good work without using these tools."
- Oh, and finally: give them what they're asking for. I'm extracting my teams from these AI experiments and saying "If you really think you can deliver entire products without us, have fun." If you think an engineer with an LLM can do design, product, research, and data work, show me. I'm very confident engineers do not want to do this stuff. And if we stop catching them when they fall on these topics, they will revolt.
A lead engineer at a major UK games studio says, "Our CEO has made a big song and dance internally and externally about how no AI is used on our game content. And he's right, the problem is everything that's not considered 'game content'..."
While they and their team are "pretty much all anti-AI", and they're able to shield their colleagues from most prodding coming from above, as time goes on this is getting harder and harder. "At first I tried the whole 'oh but it doesn't work for us' approach, and showed where AI fell down in some examples. This worked fine until third-party vendors like Amazon kept mentioning to executives how they just launched a new AI tool that does our jobs for us, or Cloudflare puts out a blog post showing how they used AI to do something similar to what we do, etc".
"Recently I heard about an executive asking why a dev team will take eight weeks to make a prototype, when Claude could do it in a few hours. I'm not sure how the lead of that team handled it, but I know it's absolutely killing morale of developers", they say.
"My organization is full of people who chase trends", a worker at a health insurance company says. "This includes finding any possible way to adopt AI into workflows, of course, but it has fortunately also meant adopting an AI strategy that includes some constraints".
Specifically, this strategy includes a policy that anything developed using AI must be reviewed by a human before it's used. "I was recently tasked with creating a massive pile of instructional documents for a new HR system we were implementing, and was specifically instructed to use Copilot to streamline because of the volume of work", they say.
"As a junior employee, I realized that all of my work, written by AI or not, would have to be tested and edited for accuracy by people higher up than me. So I was able to refer to the AI standard of human review and convince my supervisor that we would be duplicating work to generate, review, and then re-review the AI-generated documents. It ultimately saved us time to just sit together and make them right the first time, without AI".
One thing that struck me speaking with so many workers across so many industries was just how quickly AI usage has become thoroughly embedded in their daily professional lives. We're only a few years into the supposed 'boom' of this technology, with questions over everything from its legality to its effect on our cognitive ability still very much being asked, and yet its adoption has been swifter than that of the computer itself.
No wonder we're left with so many absurd situations like those described above! Managers mandating the use of a technology that their workers are actively conspiring to avoid (or are even turning against itself) should be the stuff of fiction, a workplace satire for the ages, and yet it's becoming an increasingly dire reality for more and more people. When we have games industry employees resorting to Soviet-era tactics just to share tips on dodging AI usage--avoiding an overseer that's literally recording everything they type at their keyboards--what are we even doing here?
When the bubble inevitably bursts, what happens to everyone at these companies when they're caught up in the middle of an economic meltdown? Will the short-sighted owners, executives and managers who became so captivated by the latest set of keys being jangled in front of their faces, to the point where it left the entire global economy's ass hanging out, be the ones to take responsibility? After ignoring the repeated protestations and warnings of workers like the ones I've spoken to here?
Don't answer that. Of course they won't.
Anyway, rather than end on that downer, I hope the tips and stories that have been shared here are useful to people. Both because they show that the increasing fervour with which AI is being adopted is still not resonating with your colleagues in the trenches, and also because there are some great tips to be found here! Asking a boss "why do you think that?" and making them show their ass is some all-purpose advice for the ages.
It's Hater Week here on Aftermath. This week we’re letting the hate flow through us by examining what it means to be a hater (and… a lover?), venting about some of our biggest pet peeves in games and life, and giving others a space to do the same.
