The EU probably does want worse relations with the US... the narative that "The US is steamrolling all you helpless tiny European countries" bolsters the EU's power even further. Honestly, it's all heading towards authoritarianism at this point; dark times.

If/when the AI subsidies get pulled the cost/benefit is going to take a nose dive. Things could get really interesting... https://thehill.com/opinion/te...

Apple's worldwide marketshare in mobile phones is ~25%.

Epic has ~25% of the market and YES the studios that built on their engine a very locked. https://www.reddit.com/r/gamed...

The "actual costs" are all Apple's servers... so if Apple needs to segment these people into sandboxed physically separated servers for "security" then "reasonable" could be easily $100K / month.

The ruling explicitly says Apple should get a cut, and that current cut is too big, but not even any real guidance on what "reasonable" is... the ruling is garbage

I guess Apple should try 26.9%... this ruling is garbage and the judges should be embarrassed. If it's completely up to the judges gut whether a certain cut is "reasonable" it will mean this stays high and Apple will eventually find a judge that agrees with some high number.

If the AI crash happens in the next couple years these HUGE companies will likely hold onto the accelerators for a couple more years after that and by the time they are sold at rock-bottom prices they will be truly worthless... which at least saves us from them being used by miners...

The obvious thing to question is the thing he says is gospel... that if not true would mean he and his company are irrelevant.

• "But if you define console as the property, not the system, then the notion of a very rich game that you engage in for many hours that you play on a big screen -- that's never going away."

If he's saying this... he is worried that big AAA games are in danger of disappearing.

Not sure these sources are any worse than other country's self-reporting. . . I don't trust any country's self reporting. What would make these more numbers convincing to me is if these sources jived with independent sources.

One reason these numbers do seem plausible is China is FAR AND AWAY the biggest polluter by country and has nearly tripled it's CO2 from 2000. Leveling off should be congratulated... but in the context of the incredible increases over the last few decades it's not as unbelievable. https://en.wikipedia.org/wiki/...

If this new laptop has the same ancient M1 SSD/CPU performance and a gimped version of the GPU and Neural engine of the M5 it will be pretty darn amazing.

The M5 SSDs are MUCH faster(2X+) than the M4, which makes them faster than anything else out there now and approaching DDR3 speeds. This would let Apple differentiate the "entry" MacBook with a regular fast SSD and reserve the new ludicrous speed SSDs for their other lines. https://www.reddit.com/r/apple...

Or Facebook(and all advertisers) need to get out of hoovering up data from anywhere they can... arguing that Meta isn't really eavesdropping because they record data from too many places to notice is a weak argument.

Google kneecaped alternative app stores where Apple doesn't allow alternative app stores. TFA says Google paid app developers and phone makers to hinder alternative app stores.

What is the first moved advantage for AI companies to train their own models?

If I accept your argument that making work indexable makes that work fair game for including in LLMs... then the LLMs also fall into the same category. It's relatively easy to index an existing LLM and use that to train a "new" extremely similar LLM. The hundreds of billions spent curating these LLMs seem like a boondoggle for the investors. I think a couple more DeepSeek type events will see the massive money pulled out of training models.

While I see little value in companies training models, I do see opportunity in creating systems to run models. All the current server-based LLMs will be able to run locally in the next couple years and the first companies to enable local low-latency hundred-billion-parameter LLMs will have a true first mover advantage.

Considering most of companies making models are disregarding copyright to create them, what advantage would being "first" have? The first mover can't claim copyright. The good model would be used to train another nearly identical model fairly quickly by a competitor even if only allowed to be run on their own servers.

I see server-based AI as having a very time-boxed appeal; local hardware is handling things better and better and buying LLM as a cloud service seems like a huge boondoggle from a latency and cost perspective.

What prevented you from installing what? I get you hate Apple... but seems you are just confused as the current version of Photoshop is still supported on every 2015 Mac(Photoshop 2025 works on MacOS10.12 released in 2021 which all 2015 Macs are compatible with). https://helpx.adobe.com/photos...

Maybe you couldn't figure out how to install system updates?