9,053 karma · joined March 7, 2019
Some people decide that sphere is really big, and they go on to be those historical figures. Others define it really big, and wallow in angst, aware and powerless to the suffering. Others still define it too small, and by the end of their life, find regret that they didn't try to help those within reach.
Would you rather take instructions from a ruthless robot or ruthless flesh sack?
Start at 1:07:00 in their announcement video. Craig is absolutely talking about "Apple Intelligence" as a whole in this segment.
Pragmatically, of course they'd need to add metering to any cloud available APIs that rely on large models. There's no way they will eat the cost of serving unlimited access to a cloud LLM to end users if they won't eat the cost of an image generation model.
Apple themselves have said there is usage limits, with a subscription upgrade for more usage. So clearly AI Labs are directly competing on that front, it's just a normal default/chosen decision. Considering there are defaults and still successful competitors (eg. safari v chrome), there's no reason to think that competition can't handle this too.
Edit: I want to add that Google is also probably willing to give the model away at a discount to its true value in exchange for guaranteeing that their primary competition (who has tons of cash) won’t have an economic incentive to enter the foundation model training arms race.
Most users who actually want these features for anything more serious than summarization and style updates will probably find value in a modest subscription or ad-supported tier of higher quality models, even if just for occasional usage. Apple can provide this, but once you're comparing features, for many Gemini/Claude/ChatGPT may be a better fit.
Oh, and I think there is an unfortunate but real risk that once again, apple totally over-promises here, and their AI models that they ship end up being pretty poor, and that drives users further into subscriptions.
The push by companies to incorporate AI into everything is (depending on the company) either hype and cargo-culting or it was an attempt by management to (1) try and discover if/what new workflows or tools could use it and (2) force the haters to use as it got better.
Where I work, there is an obvious split between people who have been willing to use AI, and those that hated it from day 1 and mocked the "stochastic parrots". Senior folks were disproportionately haters, and generally didn't see much productivity lift from early AI stuff. They strongly resisted the mandates to use AI, and completely missed the "agentic" inflection point that other colleagues experienced. The more willing users saw Claude Code/agents and were able to experience this as the genuine benefit it can be. Now that the more senior folks are using agentic programming, they're genuinely able to maintain code quality and see meaningful speed improvements in coding tasks.
Today, tokenmaxxing doesn't make sense because we found the product-market-fit of agentic coding. Now that most (?) employees are onboard with using it, the industry can shift focus to cost-effective usage and positive-ROI usage. For example, Uber shifting to a fixed per-employee token budget.
The coal is unreliable narrative has existed for many years because coal plants break down and have mechanical issues at higher rates than other sources, renewable and otherwise. Coal is less reliable than natural gas and oil based power, nuclear, to say nothing of the comparison to solar, which obviously doesn’t need mechanical repairs. The method of converting coal to power is actually less reliable because it’s more mechanically intense, and has more moving parts which makes it more fragile and repair prone.
If you google this, you’ll find that statistically across coal plants, they’ve historically had a reliability of 70-90% of target capacity due to mechanical issues.
The tax change impacted US company spend on their overseas businesses, so of course they’d be impacted, and it would indirectly impact overseas supplies of domestic business. Beyond that, it sets a tone and many in business are quick to follow the behaviors of others.
I’d assume they’re still building that profile while you use the product, but you won’t see any ads, and can still delete the data from the various points like you’ve mentioned.
Same with writing bad code. We’ve all seen sub-optimal decisions in code or technical artifacts that go on to be successful products or tools. Most people can’t/wont/don’t work at the Pareto-optimal workplace.
But it’s fairly visible that big companies (eg Meta) that are spending a lot on AI are actually changing spending on headcount and hiring to maintain margins. It’s not the efficiency of the workers, it’s the maintenance of margins with big new spending.
The data in this example was going to be made public anyways. All the examples of prediction markets are predicated on them becoming public. You not only need the info, you need the info before it becomes public.
I don't think we can discount this, frankly. Newer electronics are energy efficient, but older devices are more energy-intensive, and unless configured well, a gaming PC can easily use a few dollars a month in electricity, so now you're approaching subscription territory. A subscription comes with no upfront cost, higher reliability, no wasted space in your home, mobile apps, etc. (and less privacy).
In SF, you can’t evict tenants because you sell the house, and you can’t evict tenants with kids during the school year (without a just cause).
If you're trying (1), it's easier to say "I don't know, maybe <available ai> can answer". It doesn't save any time to ask an AI that the other party is equally equipped to ask. It just saves the responder time from being genuinely helpful.
If you're (2), at least explain this (or include the prompt so it's self evident and a teaching moment). Of course, if you're a SME, maybe you also have the knowledge to just answer directly - see 4.
For (3), why reply at all: see 1.
For (4), saying this associates your own authority and knowledge, and is valuable, but the omission of such disclaimer makes it indistinguishable from 1.
Let’s do the bog-standard obvious and sane thing and pick a single point in time, once a year and use the value then. Maybe, i don’t know, close of market on the last trading day of the year. At which point it won’t fluctuate again until the new tax year. Then, we can call it “mark to market” because we’re marking the value to the market at a point in time.
Finally, we stop with silly bad faith arguments because fluctuations in stock have been successful taxed for decades. This is how day-traders pay taxes, and it’s not even a little challenging to do.
Why is compensation not included in “politics” when it’s very clearly a political topic? Because when people say “avoid politics” they usually mean it as a derogatory term for “all the disagreements that I dont personally care about” - and conveniently exclude the issues they care about from “politics”. Unions don’t work unless they get enough members, and getting enough support sometimes means supporting the “political issues” of other members. It’s a team, and everyone has to contribute… but of course everyone will be better off in the long run
Negotiations for compensation is like the least life-impacting thing a union can do. Tech workers are well paid and capable of negotiating.
Things like work hours, quality of life, paid leaves, etc are important and can’t really be negotiated by the individual. Every labor victory from yesterday is the status quo but every future one is politics.
I don’t think it’s reassuring the way this goes away from AI layoffs. Just like in life, you need downtime and boredom. This is when you learn and reflect and develop taste, this is where your team steals time when an “emergency” breaks out and someone gets paged over an outage. This is where you take time to mentor the younger employees.
Relatedly, it seems that AI is entrenching cumbersome processes because it’s easy to write that extra doc, compile those notes, etc. When you make a team more lean, you’re supposed to cut out the extra process and let people work in a high-trust environment. That’s why startups are fast - everything is low process and the business has to trust employees. As companies grow, they develop all sorts of bureaucracy as scar tissue. But layoffs and cuts fail if you keep that stuff along the way.
They have a ton of other software in 2026. And they have a pretty diverse (and diversifying) income stream today. Like 30-40% from non-ads.
Is it worth it? That’s for them to say, but they can ramp up cloud services at scale pretty fast as a core competency.
It was never trumping skill. This is just a willful rewrite of history perpetuated for some political goal.
The goal was always to ensure that skill had adequate opportunity to be displayed without bias.
2015 had ~50k CS graduates.
2021 had ~100k CS graduates.
You can extrapolate the rest.
As an extreme example, maybe we’ll see long-running internships and trainings like doctors experience. Doctors don’t start their career until ~12+ years of prep and training.
Pragmatically, software development has a lot of examples of teenagers making apps and college students building software companies. In the 12 years it takes for training, low-knowledge workers could be vibe coding continuously replacements of most commercial software products they’d be hired to build. So I doubt we’ll treat software development as a rarified high skill job.
Plenty of people have a reason why they can’t do it, but plenty do it and are happier for finding their community IRL.