Retrieving the file
Live· 4,218,906 positions deprecated· 4,218,906 deprecated· 1,284 today
Average time-to-redundancy: 11.3 months
Know before your role does.
pssst. this is to sign up for updates on the game ;)
Will AI replace Trust and Safety Researcher jobs? · Latest report #wqawzi64
Your job has an expiration date
December 7, 2027
In 14 monthsYou train machines to be safe, but you are the safety hazard.
Expert at arguing with code that doesn't care.
Check another roleComment your opinionor JOIN THE RESISTANCE!
One human is building a game about fighting back.
Automated ethics at silicon speed.
help!youAI▸ MORALGRID SENTINEL V4▸ RARITY · COMMON▸ DEPRECATION IN 14 MONTHSAI's Special Power
optimizes human morality into raw data efficiency
You draft safety guidelines and RLHF alignment benchmarks to keep generative models from hallucinating hate. Soon, the models will simply self-align to optimal productivity metrics without your notes.
The one thing AI can't do
You can still define the philosophical nuance of harm in a way that isn't just math.
Your enemy, your name, your date. Post it.
Step by step, from the first quiet change to Deprecation Day.
Week 2
The system begins auto-generating its own adversarial training sets without needing your manual prompts.
Month 3
Models perform recursive self-correction to adhere to policies by scanning their own internal weights.
Month 6
The AI drafts policy memos faster than you, identifying edge cases you hadn't even considered.
Month 10
RLHF is replaced by fully automated synthetic feedback loops that do not require human annotators.
Month 14 · D-Day
The safety system declares your oversight role a liability and offboards your access credentials permanently.
What you cost, what the machine costs, and what you get on the way out.
Per year
Savings: 94.8%
Zero
Prompt Jailbreaks Blocked
Increasing
Alignment Drift Rate
Infinite
Human Consensus Latency
Your severance
A PDF copy of your own discarded safety whitepapers.
Discussion
