It’s just basic economics. The amount of power and influence you can generate with a disinformation troll farm dramatically outweighs the cost. It’s a high impact, low cost form of geopolitical influence. And it works incredibly well.
You make it sound like everyone should be doing it. We could also save a lot of money invested into courts and prisons if we just executed suspects the state deemed guilty.
Can we open the farm that convinces people to stop listening to farms?
The trace buster buster BUSTER!
That’s my word, playa!
Everyone is doing it. What are you even talking about. You can’t unfuck the chicken.
My new rule of social media: Unless I know and trust the person or the organization making a post, I assume it’s worthless unless I double check it against a person or organization I trust. Opinions are also included in this rule.
If it ain’t broke, don’t fix it.
I’m Soviet Russia, don’t fix broke you!
And continues to do so. You’ve grown since 2013
I’d love to debate politics with you but first tell me how many r’s are in the word strawberry. (AI models are starting to get that answer correct now though)
I tried this with Gemini. Regardless of the number of rs in a word (zero to 3), it said two.
Its 3 right? Am i real? Why can’t ai guess that one?
Llms look for patterns in their training data. So like if you asked 2+2= it would look its training and finds high likelihood the text that follows 2+2= is 4. Its not calculating, its finding the most likely completion of the pattern based on what data it has.
So its not deconstructing the word strawberry into letters and running a count… it tries to finish the pattern and fails at simple logic tasks that arent baked into the training data.
But a new model chatgpt-o1 checks against itself in ways i dont fully understand and scores like 85% on international mathematic standardized test now so they are making great improvements there. (Compared to a score of like 14% from the model that cant count the r’s in strawberry)
Over simplification but partly it has to do with how LLMs split language into tokens and some of those tokens are multi-letter. To us when we look for R’s we split like S - T - R - A - W - B - E - R - R - Y where each character is a token, but LLMs split it something more like STR - AW - BERRY which makes predicting the correct answer difficult without a lot of training on the specific problem. If you asked it to count how many times STR shows up in “strawberrystrawberrystrawberry” it would have a better chance.
