I'd like to understand why ID.me doesn't let non-government companies use them as an identity provider. I would love, love, LOVE if I could use a social media site which guaranteed that every user was a US citizen or green card holder.
Anonymity has its benefits, and I don't think it should be outlawed for SURE, but I should have the option of being able to use some social media site that any US citizen could participate in and know that everyone else on the site was a US citizen.
> I would love, love, LOVE if I could use a social media site which guaranteed that every user was a US citizen or green card holder.
I wouldn't assume that they verify or share US immigration status for any of their clientele, governmental or private. Being ID.me-verified says approximately nothing about US immigration status.
If ID.me sees a valid unexpired US passport or green card, then sure, they know the US immigration status which that document indicates. But there are plenty of cases where they don't know that.
If they see a REAL ID-compliant driver license/state ID, they know the person showed verifiable evidence of lawful presence in the US at the time the application was submitted to the state, but not what the corresponding US immigration status was then or what it is now. If they see a Social Security Number, they know the person either has current or past US work authorization or current or past entitlement to a benefit which requires having a Social Security Number, but nothing else about the person's current or former US immigration status.
The other documents ID.me accepts say even less on the topic.
When they see a foreign passport as the primary ID, the person's current US immigration status could range from none at all to US-foreign dual/multiple citizen to anywhere in between. And they also accept non-REAL ID-compliant driver licenses and state IDs, which are perfectly legitimate and valid for most purposes, and those usually don't indicate immigration status either. Yes, undocumented migrants can get them. But so can any other immigration status. My parents have been US citizens their whole lives and only just switched their driver licenses to REAL ID this year, because their US passports served the few purposes for which REAL IDs are required and so they kept deferring the complex procedure of making that switch instead of easily renewing the old type.
> If every account has to be tied to a real person, the cost of obtaining bots accounts will rise exponentially
Nope. Dedicated states love building fakes and passing these identity challenges. That doesn't stop the bots. Also, scammers perform these identity validations with vulenerable populations. (They even sign up their victims for KYCed bitcoin accounts)
Well, the examples you provide just show how much the cost is higher to get an account. And over time, id.me could perform additional security measures which increase the cost even further.
To be clear, I don't like that future, but I suspect that some variation of it will be in demand given the amount of bots exponentially rising.
It doesn't stop the original problem, it just makes the scams worse.
Foreign states mint fake documents all the time, scammers force their victims in to lengthy KYC processes and it gives them more leverage to do even more damage. The companies who invest in this crap end up getting a "get out of jail card for" well we implemented something how can we be responsible.
But not always authentically so. Pretty much every election cycle someone turns up a network of foreign bots engaging in political debate on the various social networks (and not just in the US either - UK and Australia both have had the same issue)
There are less-charitable explanations: the OP could want to explicitly exclude undocumented immigrants (and anyone else without an up-to-date government ID), from their social networks.
They aren't arbitrary; they're useful. If we imagine a different set of axioms also being useful, which we sometimes do, we try to figure out what follows from them as well.
My day job is in a regulated industry, we are required to have humans review everything. WE do have many many thousands of tests verifying behaviors, linters, autoformatting, we can get agents for an extra review, but human review is not going away in my industry for a long time.
If you are asking any human to write code, you are effectively asking them to write byte code. Humans should not be writing code in this era, and only in rare circumstances should humans be reading code.
You can still control it if you can trivially spin up a node with eight B200s and 2TB of memory to run your Deepseek v4 instance. Nobody on a subsidized plans is doing that. Local models that fit 128GB memory don't come close even to a "nerfed" frontier model.
Actually Qwen3.8-flash-next comes very close to a "nerfed frontier model" the way I host it locally at 4.75bpw with int8 kv. It is pretty much equivalent to opus 4.6 up to 4.8 on coding. What it lacks is some factual knowledge - this is to be expected, it's a far smaller model. But if you run the kind of coding tasks where you have to spec them in high detail anyway (large very custom environment) this model is very usable.
I run it on a 6*rtx3090 consumer PC (no threadripper etc sadly) with very constrained pcie topology (4 cards sit on 2 lanes basically) and i still get 7k prefill and 50t/s single user or 200t/s 7 users. All at full 262k contexts of course and with 96GB of ram kv cache that means I can easily have 10+ Claude code sessions open with 1-7 generating at any time.
I've been the person that said, before opus 4.6 came out that this AI stuff is absolute crap for programming. Then it suddenly wasn't and I was using it many hours each day.
Local AI was the same. It was all crap. Models like GLM5.3 Flash that are many times larger than that qwen, lots of people are oooing and ahhing over I found way too inferior. Previous DeepSeek V4.1 Flash 0731 was atrocious. They benchmaxed it to hell, but at least it was small.
The current model of same name but release 1017 and double the size is currently the best open source model IMO. It even beats my self hosted Qwen3.8-flash-next.
I've tried hosting that DeepSeek model, but even adding another rtx3090 for a total of 7 cards and my 192gb of ram bought when it cost under $1000 I get only about 25t/s decode and few hundred prefill. That is far too slow for me to use.
But, I believe once this pyramid scheme of circular financing finally bursts there will be a glut of hardware on the market where people like me will be able to buy equipment to run 750B models like we now run 120B.
I've been very impressed by it and recommended it to everyone. 32gb ram, 12gb 3080ti, and I run 3.8 flash next coder (half the experts ripped out, leaving in coding) int8 kv 128k context at 35-55tk/s and I agree with the opus comparison; I personally say 4.7 is what it seems like the most.
I read stuff like this and just think that if these open models work well enough financially you'd still be better off using them via OpenRouter or some other hoster.
Local is not worth buying hardware for, it's only good to run on the stuff you already own. As soon as you want to do work you are getting screwed almost as hard as people in the 90s did with $2000 pcs that went out of date in under 6 months. Just pay 20 bucks and use the frontier models, they have better harnesses and features than we have access to anyway.
Please don't take up discussion space for something else like that. You can write about that in plenty of other places where it's germane to the topic at hand.
Your comment is far outside the culture here. If you don't like someone's comment, read something else; click the minus sign (collapse thread) if you like. HN pages are not paper; we're not running out of space, and this thread has only 64 comments.
Also, have some empathy. You're talking to a human being.
I think their comment is the most beautiful thing I've seen today; I'm so glad they left it here.
You have a point, that the comment isn't actually about the article. And yet, einrealist also has a point that is worth hearing. It's OT, but it's worth hearing.
> Too much context confuses the LLM. Reduce unnecessary context and performance improves. Not to mention, too much context and you just run out of vram.
I don't think you can reduce context enough on a local LLM to get performance comparable to a 500k context without those reductions...
reply