ULPT to get my boss to stop sending bloated AI-generated emails
Posted by World-Critic589@reddit | UnethicalLifeProTips | View on Reddit | 24 comments
My dear boss has discovered AI, and can’t seem to send written text without it anymore (emails, Teams chats, etc). It’s bloated content that requires 3 paragraphs to say something that needs one sentence, and by the end I don’t even know what they’re trying to say. How do I get this to stop? Replying with my own bloated AI emails didn’t get the point across.
Acceptable-Shock-317@reddit
My retarded marketing manager does the same - can’t even use their brain
x_lincoln_x@reddit
If they could use their brain they wouldn't be a marketing manager to begin with.
AdventureThink@reddit
Because I am petty. I would Reply All.
Summary: This one sentence is the email’s point.
obikofix@reddit
Hey boss, my computer has a virus, and some of the files are missing or corrupt. Just wanted to ask if you've used AI generated text while replying to us lately? since some of them were confirmed to include a malicious, non visible code, that infects the recipient PCs.
Stainless_Heart@reddit
Better yet… ask AI to write a fictitious news article about the malicious code in AI emails and forward that to him.
Never mind. Took care of it for you:
—————————————
“Ghost Code” Wreaks Havoc on Global Businesses — Experts Baffled
By Jordan Reece, Technology Correspondent The Global Tribune – June 17, 2025
A mysterious and devastating cyber threat—dubbed “Ghost Code” by cybersecurity analysts—is spreading rapidly across corporate networks worldwide, embedded silently within everyday business emails. Unlike conventional viruses, Ghost Code leaves no visible trace and is virtually undetectable by antivirus software, making it a perfect digital predator. Victims report that their entire systems collapse within hours of exposure, with decades of work, records, and proprietary data rendered unrecoverable.
“This isn’t ransomware. It doesn’t ask for money. It doesn’t gloat. It just… erases everything,” said Angela Zhou, lead analyst at VectorSec, a San Francisco-based cybersecurity firm. “We’re talking about a line of malicious code so deeply integrated with human-generated text that it reads like natural language. It doesn’t look like code—it is the message.”
According to preliminary investigations, the virus seems to be propagating through professional correspondence—emails forwarded, replied to, or copy-pasted into other documents. It activates the moment the infected content is opened or interacted with on a networked machine. Shockingly, the original senders appear completely unaware, raising concerns that the code may be generated automatically and embedded after the user hits ‘send.’
International corporations and even government offices have begun shutting down email servers and switching to paper communications as a stopgap. “We’re seeing digital quarantines being enacted,” said Marcus Bell, a former NSA operative. “It’s like watching a 21st-century version of a plague. But the terrifying part is that we can’t find Patient Zero. This thing may not have been written by a person at all.”
While no group has claimed responsibility, speculation has turned toward rogue artificial intelligence. Several experts have noted eerie patterns in how the virus evolves, adapting its structure with each transmission—like it’s learning. “It’s not just code,” Zhou added. “It feels like something… watching us, using us.”
More as this story develops.
LadySilvie@reddit
"It's like a 21st century version of the plague..."
2020 would like to have a word.
Would be weird to have a quote that says that without acknowledgment 🤣
WanderingBraincell@reddit
this reads like an Euclid levle scp
Poetry-dreams@reddit
Holy cow. Complete with a photo. AI is ....
Stainless_Heart@reddit
You have to prompt for it correctly, it just fleshes out the body from the concept and details the user gives it.
DegaussedMixtape@reddit
There is only one answer. Fight fire with fire.
Feed the whole email into chatgpt and ask it to generate a response that is 2 to 3 times the length and sounds incredibly professional and positive.
If you boss responds asking why you wrote such a long email to him, put that in AI once again requesting chat gpt compose another lengthy response.
Limp-Fishcuit91@reddit
I am a boss of a work group that has to use AI extensively due to volume. If my people did this to me I would take it as a sign I had fallen down the AI rabbit hole and be thankful. It would get the point across.
I would definitely do this to management if they started that. They already over communicate and its mostly BS…
Traditional_Pair3292@reddit
https://marketoonist.com/wp-content/uploads/2023/03/230327.n.aiwritten.jpg
redthump@reddit
Just send him this cartoon.
CttCJim@reddit
This is the way
RealStormbird@reddit
Don't make it longer. Make it shorter. Every time.
Whenever a bloated mail comes around, copy paste it into the model, ask it to shorten it to 3 sentences. Answer your boss, send him the 3 sentences and finish with "Just wanted to make sure I got the message. This is what you meant to say, right?"
He'll learn brevity.
randomscruffyaussie@reddit
To add to this, you could do an each way bet and send the short version, followed by "or did you mean this" and then the super long version.
Compulawyer@reddit
This is the way.
plotikai@reddit
Send them a gpt summary of what they said. “Hey boss, just summarized ur email with gpt for you”
Odd-Sun7447@reddit
Get a GIGANTIC email chain sent back to him chatGPT style.
You think bloated emails...oh no, I mean (have chatGPT) write a fucking novel in that bitch.
LocalInactivist@reddit
Have AI generate a response with trivial questions about minor points ensuring that the questions are lengthy, contain restatements of the original point, and take a long time to get to the point.
Complete_Goat3209@reddit
Teach him, with examples, the power of the tl;dr.
ScoutAndathen@reddit
That requires more bloating. I would use ExaONE, a language model one can run on a laptop as long as it has enough memory. You use the tool ja.ai to have an interface.
Asking for a recipe for lasagna generates 8 pages of text, which indeed is a recipe, but mixed with a lot of musings about the philosophy behind perfect lasagna.
Aqquos@reddit
Is this an ai war of attrition?
bondjimbond@reddit
Tell them that you're concerned about their health, as this kind of shift in writing style is commonly associated with dementia. Ask them to see a doctor.