@yacineMTB playing devils advocate here. Wouldnt meta / tiktok be able to detect unique signatures of a human vs llm scrolling? Then they filter the llm scroll data out of the training set
@moultano nobody disagrees that the US needs top AI talent. The problem is that body shops like accenture fill positions with indians that my 79 iq cousin from new jersey could have
@ReaperCapital some of them act as bridges to offshore teams. multiply those numbers by 2-3x to get the true number of positions held at American companies by foreigners.
Chris, the reason why most phds are foreign born is because it provides a 3-8 year extension of their student visa. Not because they are better at it.
Then, they enter the h1b system which holds their citizenship hostage in exchange for their labor. They work for lower wages because the alternative is deportation. Its indentured servitude which suppresses wages for native born Americans.
Native born Americans are plenty capable of high level research. Your implication that enforcing immigration law will stymie academic research in this country is insulting to those who built it. Youβve been duped by corporate propaganda; this is one of the issues you must use first principles to derive a stance on.
Quite possibly they looked at the questions and load you were bringing to the table and decided you were a βlow tierβ user, and are A/B testing a simpler configuration for users that fit your usage profile. They are basically saying you have no use for extended reasoning based on previous interactions. Whatever questions you bring they think can be solved with this set of model parameters.
Here's what happened: they were swimming at a responsible depth and then a powerful current dragged them down and they all drowned. The Maldives doesn't want to say that's what happened because it would be bad for tourism. Better to say it was a freak caving accident.
Sometimes the masters of the universe inadvertently disclose that they know very little about what the masses infer they know much about. This is one of those times.
We should federally tax Tokens at the Provider level.
Not a lot. Less than 50c per million tokens.
It will accomplish 4 things (at least )
1. It will push the big AI players to optimize tokenization, caching , routing and localization
Which will
2. Reduce energy usage. Saving them in energy costs more than what they paid in tax and reducing strain created by the growth in energy consumption
Which will
3. Generate maybe 10 billion dollars a year to start, but over the next ten years could grow 30x to 100x
Which will
4. Create a source of funding to pay down the federal debt or deploy, in response to the things AI brings that we donβt expect or donβt like
At some point the models will pass it on to customers. Of course. Thatβs ok. Customers will have the ability to choose between providers. Or to do everything using open source models locally.
Thoughts ?
@mcuban@jfeldis Sometimes the masters of the universe inadvertently disclose that they know very little about what the masses infer they know much about. This is one of those times.
@SofiasBio Its a marketing scheme to sell tokens. They donβt care if the products work. They care about founders excitedly posting about the tasks they do with ai.