.... Chemotherapy targets rapidly dividing cells (e.g. cancer, hair, blood). Most cells in your body don't undergo mitosis often enough to be affected by chemo.
Besides, we have a host of other treatments for cancer:
...to name a few categories.
So, no this is not even "basically" how we treat cancer.
This reminds me of learning a shortcut in math class but also knowing that the lesson didn't cover that particular method. So, I use the shortcut to get the answer on a multiple choice question, but I use method from the lesson when asked to show my work. (e.g. Pascal's Pyramid vs Binomial Expansion).
It might not seem like a shortcut for us, but something about this LLM's training makes it easier to use heuristics. That's actually a pretty big deal for a machine to choose fuzzy logic over algorithms when it knows that the teacher wants it to use the algorithm.
Yeah, but this reminds me of a line from game of thrones:
"If you're a famous smuggler, you're doing it wrong."
I'm not saying you should believe everything the media or government says, just that the specific lies they choose to tell holds valuable information about their motives, sponsors, allies, etc...
You are asking to be spoon fed the truth. I'm saying that if you learn to refine the truth, propaganda begins to have the opposite of its intended effect.
Let’s be clear: uncritically reporting the White House’s “nothing to see here” stance isn’t journalism — it’s stenography.
Yes, it is journalism. Access journalism specifically, which is still very useful, especially under a hostile government. It can serve as a launch point for investigative journalism (what Masnick is actually calling for here), without drawing fire from those in power.
Like it or not journalists don't have to prescribe anything specific to fill a role. Sometimes it is enough to let the ruling class feel heard, because no matter how opaque the rhetoric, even deceptive signals carry the truth.
But what do they connotate?
Root access is not given. It is taken.
xAI's Partners:
A16Z, Blackrock, Fidelity Management & Research Company, Kingdom Holdings, Lightspeed, MGX, Morgan Stanley, OIA, QIA, Sequoia Capital, Valor Equity Partners, and Vy Capital, among others.
If he could, he wouldn't need to say it.
Maybe you're right. Maybe it's Markov chains all the way down.
The only way I can think to test this would be to "poison" the training data with faulty arithmetic to see if it is just recalling precedent or actually implementing an algorithm.