paxys

No, because it has never meant a specific thing that everyone agreed on.

The model said it was perfectly aligned.

Every frontier release nowadays is "we've launched*"

* for a special group of customers that you're not in. Keep waiting peasant.

Why is this flagged ?

Shows the weirdness of online journalism. News outlets were briefed about an upcoming event and pre-wrote and scheduled articles. When the time came they were all triggered. Except...the event didn't actually happen.

That's how every AI launch goes. 5.6 was the same, as as Mythos etc.

Boring answer – all these services are individually down a lot, and the downtimes were bound to sync up. Similar to the pendulum synchronization effect.

Especially considering memory/gpu/compute are scarce so these services are likely running with very little buffer.

Willing to bet it's because they can't handle the surge of traffic they are getting from ChatGPT/Claude users.

They're going to deadlock their own productivity as well

I know they're not all hosted on AWS, so Cloudflare?

Models have all kinds of garbage from all corners of the internet in their training data. The key is alignment. You feed it bad data but also teach it right from wrong.

Can’t imagine the stress of the researcher who had to run exploitbench again knowing what happened last time around.

I personally find that Scratch has very limited utility because kids can start to code in a "real" programming language pretty early on, by like 11 or 12. Would it make a difference in outcome had they started even earlier? I doubt it.

Why does anyone need to "do it"? Who did God?

The "battle" for Pluto is an interesting case study in indoctrination. A few generations of children grew up being taught a specific fact, saw it in textbooks, learned mnemonics about it, had memorable cartoon characters named after it, and then one day they were told – with sound reasoning and plenty of scientific evidence – that they should now update their views on the topic.

You would think this group of well educated, critically thinking adults would go "sure", but something fundamental in their brains just refuses to let them accept this meaningless change to their lives.

Now scale this up to actual impactful topics like religion, politics, relationships, race and more and most conflicts around the world start to make sense.

Trusting Lenovo as an auth provider is certainly a decision..

Handling SSO related security-holes in SaaS apps has made up a major chunk of my career at this point.

The most common one – app adds SAML-based login and treats the returned email from the SSO provider as authoritative. What the developers don’t realize is that you can set up your own tenant with most major SSO providers and assign users any arbitrary email – no verification needed.

And if an account with the same email address already exists in your system – congrats, you have just granted full access to it to basically anyone.

Half the job of a map maker is to figure out correct borders in disputed areas while keeping both sides happy. In the internet age they can do that by simply showing a different map in each country.

Has writing ever been a safe job, even before AI?

Local setups aren't going to make sense purely from a cost perspective, and definitely not when you are buying Apple hardware. AI subscriptions are too highly subsidized right now.

I really hope with Ternus taking the helm Apple starts to remember that it has products outside of iPhone.

ChatGPT Work = Codex rebranded for non-coders, with the option to run it in a cloud VM.

Opium

> I don’t use social media

You are using social media right now.

"You are not allowed to use our services to develop a competing product" is a standard clause in pretty much every contract of this nature. There's nothing antitrust about it. Companies are allowed to say no to bad deals.

Musk was forced to admit under oath during his OpenAI lawsuit that xAI distilled GPT models to train Grok. No reason to believe he won't use Cursor data for the same thing, if it isn't happening already.

They already publish gpt-oss which is several generations better than gpt-3

Sure, and what are you going to do about it? The current government is above laws.

The concept behind it isn't going away. There are plenty of companies that hire people in bulk to do manual data labeling, transcription, RLHF, moderation and lots more. It's just that they are now catering to large AI companies, not regular people looking to get some repetitive work done (since that can now mostly be done by AI).

The impact on academics and research has already been felt over the last decade. US used to be the default destination for academics and skilled workers from China ever since the cultural revolution. They engrained themselves in the country, studied/worked at top univerities, taught multiple generations of students, did key research, worked at top tech companies, and overall were major contributors to American technological superiority.

Then starting ~2016 Trump and his DoJ made it borderline impossible for Chinese nationals to get visas, specifically targeting research universities. Since then new Chinese skilled immigration has slowed down to a trickle, and there has been a massive "brain gain" of existing talent in China's favor.

The result - China now has a booming tech industry, Chinese universities have overlapped Amercian ones in research output, China is a very close 2nd in the AI race. And as skilled immigrants continue to be driven out of the US China in particular will enjoy the benefits.