On one side are leading A.I. companies like Anthropic and OpenAI, which claim that some A.I. models are too dangerous to be developed in the open and must be tightly controlled — by businesses like themselves — for safety.
And there's the reason these giant companies keep saying their products are dangerous over and over.
Sounds more like the Indian government wanted an apology for Facebook removing a state propaganda video, and added the other apology requirement for PR.
Earlier on Wednesday, Meta representatives also met IT secretary S Krishnan, who gave them three days to apologise for removing PM Narendra Modi's video on exam reforms for five hours and for hosting sex abuse content on its platforms.
If you're using an online chatbot, there's a roughly a 0% chance that there's any encryption keeping your conversations provably unreadable. This part of the conclusion is inevitable.
Still very glad there was an article with insider information though, because exposure like this is important.
Words are meant to convey meaning, and I bet you would lose a lot of non-technical people if you tried to explain the intricacies of what makes large language models worse than content recommendation systems, and doing this would be redundant for technical users.
So Corey can safely use the two words interchangeably and communicate with technical and non-technical alike.
I remember a candidate talking about shooting a person in public and still winning, and he did. Call this a hot take, but I'd rather my politicians brag about shooting cheap surveillance tech than a living being
I can already hear jingoist Dario Amodei shrieking about how we must oppose the Russian devils' laws by allowing copyright on this side of the Atlantic.
Employees are given a 2TB iCloud plan and are able to merge the storage with their existing Apple Account or create a new account. Since only one primary account can be signed in at a time, most opt to use their existing account to avoid having to carry two iPhones.
This makes Android look like a sophisticated OS by comparison.
Some former employees said they were petrified to delete the files for fear that doing so would attract Apple's attention.
I can't blame them. Imagine having a sensitive document shared with high-ranking employees popping up in your (aforementioned very well designed) cloud storage account. What are you going to do, sent it to Steve Apple's trashbin?
While this sounds like cope, I'm leaning towards believing it. Firefox also just saw a 50% increase in users... Or the data got messed up in some way that nobody else (including Mozilla themselves) is able to detect and report.
Organic Linux usage jumping a full 100% seems even less likely.
I don't associate these two different events as evidence of a shift in personal stances, because they can both be things he believes at the same time. That also means that hopefully people can empathize with him even while disagreeing with his stances about some things.
(If his stance did change, that would be great and I wouldn't want to deny it, but his video promoting AI lobbyists is still unfortunately up.)
Right after posting this, I stumbled on basically a text version of the same sentiment. I can understand not wanting to trust a random YouTube video, even if it is about a guy who most people know due to his own YouTube videos.
The article (Hank Green understanding himself) and the comment I was responding to (Hank Green misleading people about AI) are two similar different topics that can and should be judged separately.
I was responding to the person who said he was generally trustworthy about the topic in general, with what I think is a significant counterexample.
And there's the reason these giant companies keep saying their products are dangerous over and over.