Fuck AI

Part of me wants to share my reasons against AI at work. Some possible reasons I'm thinking of sharing are cooking the planet, you don't know when it is hallucinating so how do you trust it, critical thinking rot.

No boss is going to be persuaded by "cooking the planet". Nor do they care about critical thinking rot.

But the hallucinations? That they'll care about.

Pick something non work-related that your boss is an expert in. Engage the AI in that something until it generates a whole bunch of hallucinations. (My favourite thing is to have an AI hallucinate bands that don't exist, albums that don't exist, songs that don't exist, lyrics that don't exist, etc.: All of which is trivial to verify and prove wrong.)

Here. I just generated this conversation in Deepseek for you to show you how easy it is to get an AI to hallucinate. I just asked a question about an almost completely non-existent concept ("inukpunk") and got it bloviating a bunch of idiocy before catching it with the fact that what it claims is a thriving literary movement doesn't actually exist.

Note: I just asked it a three-word question and it created from whole cloth a breathtaking amount of text on a subject that doesn't exist.

As someone who was recently fired for, among other things, being a stick in the mud about AI, I have some thoughts.

Try not to take work too seriously.

I drew a picture of Carl from Aqua Teen Hunger Force saying, "It don't matter. None of this matters." and put that by my monitor in my office. Whenever I started to feel activated about some bullshit, I just glanced over at Carl. It calmed me down to be reminded that my real job is to simply not lose the job.

It's a very bad time to be unemployed and anti-AI right now. Just tread lightly, ride out the storm, and let the inevitable reckoning eventually come to pass.

Oh, I'm a master at this. It's called malicious complience.

Do all the work your boss wants. Use the AI that they want. And......thats it.

The key here is NOT to correct any error it gives.

You did what your boss said. Let your boss deal with what he said.

Then as all these problems come back, you WANT it to cause chaos. You WANT him to lose money.

"I did exactly what you told me to do. These are the results."

Draw comparisons with trade secrets. "If it was profitable, AI companies would be hoarding it jealously, not trying to sell it" is a good argument I've used before.

Would you use a calculator, that sometimes, confidently without itself knowing when, provides a wrong answer, even if it's only 0,1% of the time?

Sorry for my english, I'm not as good as I wish talking on it.

I'm in a similar scenario. The tech leader of the group I'm working on it is a pro-AI. He use it everyday for all. He didn't search on internet, just ask to various AIs for answers. Today, in a daily, has talk to me not to ask to the development team for questions about compilation errors in their code (I'm a QA engineer) before to ask to an AI. He has finished hes speech with: "Another who use AI will come and do your job sooner".

It's so frustrating....

Ed Zitron's newsletter has some incredibly useful resources to reference. (I would link it, but I am sleepy at the moment, I apologise).

My opinion is that unless you feel like it's putting your career at risk, it's worthwhile to discuss the negatives of AI. When people who aren't very techy are surrounded by people bigging up AI and there's no-one dissenting, it's quite common that they will assume that this is just something that's too complex for them to understand, so some people are sort of conditioned to believe that AI must be magic, even if their own personal experience or intuition tells them that it's bullshit.

I also believe that anti-AI takes will age like wine; even though there are some legitimately awesome uses of this tech, overwhelmingly, the majority of the shit that's being pushed is dumb as hell. AI can be a useful tool in certain cases, but not in the way it's being rolled out. Instead of empowering humans, it's just making more slop and hallucinations to wade through. I'm reminded now of a good post from Cory Doctorow that discusses AI and "reverse centaurs". I can't explain it better than him, but you could probably find it easy by searching, if you're interested.

Make a point of mentioning all the mistakes it makes, and the time you spend unfucking them.

I had mandatory training on how to use the company’s sandboxed chatgpt. I asked a question which was how to justify the waste of money in it recommending products in our catalog that had confidently and blatantly wrong specs. I did not get an answer

Things like this are really going to depend on the use. If it's crammed somewhere for the sake of cramming it, there are probably going to be some good opportunities to pick apart that particular plan, as opposed to AI in general.

On the other hand, my small company is looking into it for proposal writing (government contracting). There are some tools out there for that specific purpose, and as much as I loathe saying it, it can be a decent tool if used properly. Right now we can barely afford the proposal writer we have, and when a decent RFP comes out it's all hands on deck with all management diverting their attention for what amounts to a chance of business. If you treat the tool as a very fast junior-level proposal writer whose output you can review on the fly, especially for a lot of the boilerplate stuff, it can be legitimately useful.

That said, I'll also openly make comments about how I'm a curmudgeon for that kind of thing, or maybe say something about how it's basically your cell phone's text prediction thingy on steroids. Usually I say something about how the human element becomes an afterthought (example: generative AI producing content for SEO. One bot trying to please another under the guise of being for the humans). My AI-loving boss may disagree, but I'm otherwise good enough at my job (software development) that we're respectful about it to one another and he doesn't try to shove it down my throat. And even our lone proposal writer doesn't seem overly excited about it when I've talked with him (but willing to give it a try)

Consider placing an A4 sized page of general info in the lunchroom. No one will pick it up but many will read it. Replace it occasionally with a new design that isn't AI generated.

To start a discussion, perhaps try from a personnal perspective which can be an honest story of transition from ignorance to understanding. The awareness of these issues is coming to the collective concious light, albeit gradually, and we are all learning the impacts as they're occurring.

"AI is an awesome tool but I had no idea it caused so much pollution. I watched a story about X's platform and their Grok AI centre in Memphis. Its causing so much harm to the people and the local environment, I had no idea! I was shocked and sadened to see these peoole suffering. We might need to start thinking about responsible use of AI to limit the human and environmental impact, especially if that's what our planet needs right now."

The last time they did a survey, about the most recent town hall, I wrote like 1600 words about it, citing about a dozen sources. Haven’t heard from anyone about it yet, so… ¯_(ツ)_/¯

It's insufferable how my company is shoving AI everywhere! We are expected to all embrace AI and let it help us to do our jobs better and more efficiently. But it objectively sucks ! Isn't only somewhat good at things that are either useless or already have trivial solutions, and a complete waste of time for anything I could reasonably use some help with automation.

Now, the training materials are presented by AI hosts, the posters are AI slop, every corporate communications is about how great is AI. Even a colleague retired last week after working some 40 years or so for the company and he got an AI song 😵

I am open to my boss, I tell him it's a waste of my time and that it looks like a good deal today when the service is given at a loss by companies that are still in phase 1 of enshitification, that once it moves to the next stage, we will be on the losing end of the deal. I work for an industrial company, this isn't going to play out nice for us for sure.

I tell people to get fucked, but I'm self-employed. People are so easily manipulated, and when billions of dollars are being pumped into it from the largest companies on Earth, it's not too hard to brainwash people into thinking it's useful.

I remember in the early days of CGPT some guy on a podcast talking about how CGPT had already read this brand new book and CGPT had already summarized it for him. Putting aside how utterly useless that is, CGPT isn't trained on anything after 2021, and this dude not only didn't know what he was talking about, but was straight up lying.

• ELIZA effect – 1966 to give you an idea how long this idea’s been around
• “it’s all about the grift” – techbros lusting after their anime sex-slaves and bosses trying to fire every worker they can’t underpay or offshore
• “just one more prompt bro” – psychological addition to a program designed to be obsequious, programmed to apologize profusely and lay on the praise thickly, one more pull on the slot machine coding prompt
• “prompts are cheaper than therapy” – when your country blames mental health on moral failings instead of societal failings
• “bullshit generators for bullshit jobs” – the job is meaningless, you know it’s meaningless, but you’re required to pretend it’s important – your boss is running his bullet points through a bullshit generator to email you an essay that you run through a bullshit generator to summarize as bullet points

I usually tell people I have better ways of doing actual work that doesn’t require me to hand over every facet of my life.

How easy would it be for you to find another job, and how much would you care if your workplace closed down?

Just be openly scornful of it and people who use it