It should be obvious that these services are operating at a loss. The monthly subscriptions especially, but I’m even skeptical that the linear API pricing is sustainable.
It feels like a classic “drug dealer” model to me. Get everyone hooked with cheap access, then raise prices later. Unless there’s a major breakthrough in the underlying technology, I don’t see how a significant price increase isn’t inevitable once adoption is locked in.
This seems unlikely while we have open weights models available that are ~as decent as the frontier ones.
Given the API prices for open weights models of similar size are 5-10x less than the frontier models the APIs are very profitable on a pure unit economics approach. I strongly suspect they make money off their monthly plans as well.
Did people learn nothing from the rise, stall, and now fall of social networks?
Yes, AI can do some incredible things. But we’re also running full speed into an ecosystem controlled by 2 or 3 major companies. Running at a loss. A reality check is coming.
It’s not a technology problem. It’s an economic problem. People are too busy looking at the tech to notice.
I'm not concerned, they're accelerating research and development into hardware and more optimal models. People forget that you can locally host some of the early models quantized to 4 with reasonable inference with a 4080 and 64gb of ram. There are daily tools being released that are a simple click and run, without much hassle other than downloading the model and you're off and running.
Yes there is mad dash by Google, Oracle, Microsoft, Meta, and China not to cede their position to each other - it actually isn't about who will buy or pay for the service its more of a Business Strategic position to obtain critical mass in a new market using their massive reserve of cash. The users right now are insignificant to their goal - they probably aren't even given a second thought.
> But we’re also running full speed into an ecosystem controlled by 2 or 3 major companies.
We aren't, though. They think we are :-/
The reality is that tokens are the second-lowest value link in the AI value-chain (the lowest-value item being electricity).
These providers are operating low down in the value chain; they are trying to sell a fungible, easy replaceable and (if hardware price trends is any indication) easily self-hostable.
They have no secret sauce, no moat. If they jack up the prices, their users will simply move to the next provider, and repeat ad nauseum as long as VCs want to subsidise in the hope of a landgrab.
Their bet is that most people will not fill up 100% of their weekly usage for 4 consecutive weeks of their monthly plan, because they are humans and the limits impede long running tasks during working hours.
Skills is a generic construct.
System prompt is generic as well.
Subagents, AGENTS.md, CLAUDE.md etc. these are generic, "please care for my instruction" kind of constructs without any real guarantee to close gaps.
Tool is generic (CC vs OpenCode)
Ecosystem is already same everywhere.
The point is that wrappers matter. Orchestration, tool calls, reasoning loops, system prompts, agentic capabilities. Output is different, quality is different.
On [2] he said that natural immunity from getting covid-19 is better than getting the vaccine alone, which is factually correct, as many studies demonstrated (note: may vary by strains, but was particularly the case in 2021/2022). There's nothing crazy about this, and it's very reasonable to say you prefer to evaluate the risk/benefit and take the vaccine accordingly, instead of mandating this for every demographic.
People tend to fall back on tribalism and slap labels on others instead of engaging with nuance or complexity.
> On [2] he said that natural immunity from getting covid-19 is better than getting the vaccine alone,
He was more on the anti vax side than this statement implies, at least that was my take away from the [2] article:
> For unvaccinated people who got COVID-19 and recovered, he said, "Now you’ve got natural immunity and you’ve got no vaccination in you. Can we all agree that that was the winning path?"
[a]
> better than getting the vaccine alone, which is factually correct
You are not giving a metric here so I can not tell why you think it is better. Everything I have read indicates there are more risks, death or long term complications, with covid-19 exposure before vaccination than the other way around. The conclusion of [2] is similar to this.
The original Scott Adam's post not longer exists, is there another place where he recorded why he believed contacting covid-19 before vaccination was the winning path? Without that the quotes look damning against his view point.
Apparently politifact reached out for comment and did not get any:
> We sent emails to an address listed on Adams’ website and at Dilbert.com and an address on his Facebook page. We didn’t get a reply.
Several 2021–2022 studies, especially Delta-focused, suggested natural immunity provided robust or superior protection against reinfection compared to two-dose vaccination alone.
I read the abstract and conclusion of all three, none of them talk about natural immunity with no vaccination being the "winning path" like Scott Adams did. None of them talk about getting covid before getting vaccinated(maybe only optionally) as a better or safer path, not in the abstract or conclusions at least.
[1] essentially says that there is no value for people who got infected by SARS CoV-2 to get vaccinated:
"our findings suggest that once an individual has fully recovered from initial infection, prior SARS CoV-2 infection protects against subsequent SARS-CoV-2 infection and its related negative outcomes. Moreover, the level of effectiveness seemed similar in both the recovered and fully vaccinated cohorts. With a paucity of vaccine doses, this should be one of several aspects that should be considered when deciding whether or not to prioritize vaccination of previously infected adults."
None of that is advise to not take the vaccine and try for natural immunity before getting a vaccination.
In fact the advise here is conditional on "a paucity of vaccine doses" so they may(not clear one way or the other from your quote) recommend vaccines for people who have natural immunity if there were enough vaccines to go around.
"All of the included studies found at least statistical equivalence between the protection of full vaccination and natural immunity; and, three studies found superiority of natural immunity."
> "The anti-vaxxers clearly are the winners at this point, and I think it would probably stay that way," Adams is seen saying in a video clip posted on Instagram. "And I don’t want to put any shade on that, whatsoever; they came out the best."
Please actually read the linked article instead of creating some false narrative about people falling back into tribalism. Additionally, his claim from his quote is predicated on ignoring the fact that someone who has natural immunity from past exposure didn't die. It also overlooks those who may suffer long term side effects from the virus that a vaccine would help avoid.
> [1] The best the fairly obvious house republican "investigation" into joe biden could manage was some vague statements about his son getting paid for having the last name biden, which may or may not be illegal, but certainly seems unethical, but more importantly, ISN'T THE SITTING PRESIDENT OF THE UNITED STATES. Like, it is so incredibly obvious that words fail me that the president being corrupt matters A LOT MORE than his son being corrupt. Like, a lot a lot a lot more.
reply