Everything about frontier AI companies relies on secrecy. No specific details about architectures, dispatching between different backbones, training details such as data acquisition, timelines, sources, amounts and/or costs, or almost anything that would allow anyone to replicate even the most basic aspects of anything they are doing. What is the cost of one more secret, in this scenario?
Great for him, but when you mention research and fun, I have to say I'm not aware MJ published any research whatsoever.
And on the topic of fun, while it's certainly highly subjective, I remember that the moderation with the MJ tool was at one point so strict that you could not generate an image containing a "treasure chest" since they censored the word "chest".
I'm happy that state of the art models are now developed by actors who publish comprehensive technical reports and open-weights.
Would that not create the issue that you would only need to find one bypass for said official anti-cheat that then works for all games out there?
I heard with Denuvo reverse engineering work needs to be done for each individual target to unprotect it, but I'm not sure how this will be the case with a first party anti-cheat driver.
When I tried 5.2 Codex in GitHub Copilot it executed some first steps like searching for the relevant files, then it output the number "2" and stopped the response.
On further prompting it did the next step and terminated early again after printing how it would proceed.
It's most likely just a bug in GitHub Copilot, but it seems weird to me that they add models that clearly don't even work with their agentic harness.
I paid 150€ for a Mini PC with an Intel N100, 16 GB of DDR5 memory, and a 500 GB SSD.
While I have no intention to scale up low spec hardware like this, it at least seems to beat the Azure VMs we use at work with "4 CPUs", which corresponds to two physical cores on an AMD EPYC CPU.
And that super slow machine I understand costs more than $100 per month, and that's without charges for disk space slower than the SSD, or network traffic.
Renting at Azure seems to be a terrible decision, particularly for desktop use.
I strongly oppose the constant slander and the litany of lies partisan commenters post about Musk.
You don't get to throw out "fondness for throwing Nazi salutes" slander, based on an hoax immediately debunked at the time, and then act like you're doing democracy a favor. Try to stick to the facts.
Regarding the journalist discussed here, I had a look at his X account, and he posted no less than 20 posts attacking Tesla and Musk in just the last day. It's virtually all he posts, and it indeed appears deranged. The flagged comment was fair enough.
Seriously what is up with all the Electrical n apologists?
Dude's a nazi. Weaseled his way into everything digitally related to the American government and should be treated like foreign intelligence agent. He has oversold and under-delivered everything he has bought from other people to claim for himself.
Weird he's got so many dickriders on HN.
> You don't get to throw out "fondness for throwing Nazi salutes" slander, based on an hoax immediately debunked at the time, and then act like you're doing democracy a favor.
Are you claiming that this is not an accurate depiction of what happened on stage? (That is the video is in some form fake. A deep fake, or special effects, or an Elon impersonator or whatever.)
Or are you claiming that the gesture seen is not a nazi salute?
Probably not that you support the Nazi regime, as that would be a ridiculous thing to think.
Particularly so if a year before you visited Auschwitz and stated it was "tragic that humans could do this to other humans", and told us how you attended a Hebrew preschool and have a lot of Jewish friends.
There is a difference between paying 30% and 0.1% that goes beyond "precise maths".
It's an egregious share, and Apple is making an estimated $30 billion a year with this, at a margin perhaps more than twice as high as on iPhone sales.
It's not so much that I love giving 30% to Apple, and more that there is no way to move your business elsewhere because Apple monopolizes mobile app distribution.
And the other half of the mobile app market is monopolized by Google who copies the pricing model while delivering even worse (if any) service to developers.
It's either getting out of mobile apps or paying up.
This is not going to change without drastic steps by regulators, which both Apple and Google fight tooth and nail.
You know some of us remember Mac System [7|8|9] and how MSFT pretty much ruled everything (Apple had low %).
We kept working on the platform and developing tools and things changed. Of course Apple is a lot more powerful than MSFT back then and the general population is their target.
Cheating on the benchmark in such a blatantly intentional way would create a large reputational risk for both the org and the researcher personally.
When you're already at the top, why would you do that just for optimizing one benchmark score?
reply