spangry

No, where did I say that? All I said was that I can see the logic - doesn't mean I agree with it. This policy sucks for me personally, as a non-US citizen.

I agree this is probably their thinking - they view frontier models (and the capability to build them) as a vital strategic edge that they want to keep to themselves.

The problem is that there are network effects at play - the more people you have using your models, the more training and fine-tuning data you're accumulating, so the faster you can develop the next frontier model. Not to mention the fact that more users means more revenue to fund your next-gen model training.

Perhaps the US administration is gambling that US citizens on their own provide enough of a training data and revenue flywheel for them to keep their AI development edge.

The next interesting question will be - will the US share this capability with her traditional strategic allies (e.g. five-eyes countries), or is it truly America First (or, 'America Alone')?

"The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees." This press release is odd - it says that the export control was imposed to stop foreign nationals from using Fable / Mythos, and then goes on to talk about supposed concerns about jailbreaking the model.

But is that really the concern of the US Administration? This looks more to me like they are viewing frontier models as a strategic asset which they want to keep for US-exclusive use. I can see the logic - if frontier models generally accelerate a society's technological development, then a country looking to retain or increase its strategic edge over other countries would try and keep this sort of multiplier for themselves.

I'm guessing Anthropic shut of access for everyone because currently they have no reliable way to know whether a user is or is not a US citizen. In the near future we might be in a situation where you need to prove your US citizenship before Anthropic / Open AI will allow you to use their current frontier model.

The next interesting question will be - will the US share this capability with her traditional strategic allies (e.g. five-eyes countries), or is it truly America First (or, 'America Alone')?

"The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any *foreign national*, whether inside or outside the United States, including *foreign national* Anthropic employees."

This press release is odd - it says that the export control was imposed to stop foreign nationals from using Fable / Mythos, and then goes on to talk about supposed concerns about jailbreaking the model.

But is that really the concern of the US Administration? This looks more to me like they are viewing frontier models as a strategic asset which they want to keep for US-exclusive use. I can see the logic - if frontier models generally accelerate a society's technological development, then a country looking to retain or increase its strategic edge over other countries would try and keep this sort of multiplier for themselves.

I'm guessing Anthropic shut of access for everyone because currently they have no reliable way to know whether a user is or is not a US citizen. In the near future we might be in a situation where you need to prove your US citizenship before Anthropic / Open AI will allow you to use their current frontier model.

Why don't they just auction H1B visas?

1. Set a monthly quota (for argument's sake, 5000 a month).

2. Each month the government holds an auction, and businesses that believe they need to bring in foreign labour can bid for however many visas they want.

3. The highest bidding companies get the visas.

This way the government can control how many foreign workers come in to the country each year, and the economic rents from bringing in these foreign workers accrues to the public as additional government revenue, rather than as additional profit for corporations.

It also means that the limited number of available H1B visas are put towards their highest value use. The company that wants to bring in a highly skilled foreign worker that's worth $500,000 in additional profit to them will be willing to bid up to $499,999 for their visa. The company that wants to bring in another Tim Hortons worker won't be able to outbid them.

Looks really cool, I like the pretty but minimalist interface. Could I store the SQlite file on, say, google drive so that I could access my journal from different devices while the contents are still kept secure because they’re encrypted?

As I understand the data in this article, midlife mortality rates for those who hold college degrees has declined from 1992 to 2019, whereas the rate has remained largely stable for non-college degree holders.

I wonder if this trend is due, in part, to college degree holders becoming disproportionately female over time, and women having lower midlife mortality rates? https://www.aei.org/wp-content/uploads/2018/05/degrees-1.png

Then why don’t you go bet against these people? You could be rich.

This is good advice. You can buy “wart-off” freezing spray over the counter at a pharmacy to do this.

Oh darn, I thought they'd gotten Arch running on an M1 but they actually switched to a ThinkBook.

I somewhat regret my expensive switch from Linux to MacOS. MacOS is just so weird, it doesn't make any sense to me. For the first time in my life I feel like some tech-illiterate grandpa trying to figure out how to make his blasted computer do stuff.

I missed that bit. What did it say?

Pretty sure that's what I have too (and am diagnosed ADHD). I used to complain to my Mum that I couldn't hear what people were saying on the bus back home from school. She took me to get a hearing test and it turns out I have bat-like hearing. I'm just not able to seperate out speech from background noise - noisy restaurants and bars are a nightmare.

The US gets a stronger position with China such that china attacking Taiwan would be like bombing Apple or Google. The USA will go to war over that.

If TSMC has effectively transferred their technology to Intel, doesn't this remove a reason for the US to defend Taiwan?

Just spitballing - maybe if TSMC had a 49% stake in Intel they'd be incentivised to transfer their chip-making technologies and techniques to Intel, to maximise the value of their (forced) investment?

For me it's more a recognition after the fact thing: "Oh that was a good comment who said that? Oh that guy, yeah not surprised."

I see you’re being downvoted but I think you’re right.

The US reserve currency system allows the US to tax the world, whenever they print more dollars, via oil price inflation. Anyone without sufficient military power who tries to stop denominating their oil exports in US dollars ends up with freedom and democracy being brought to their country. Saddam Hussein switched to denominating Iraqi oil exports in Euros. Gaddafi tried to establish a pan-African gold-back dinar to denominate African oil exports in. And look how it turned out for them.

It’s a stealth tribute system. If you stop paying your tribute, the empire responds accordingly.

There’s one crucial difference - countries can’t print new diamonds or gold bars at will. Whenever the US prints more dollars, it transfers value from hoarders of US dollars to themselves - they’re levying a stealth tax via global oil price inflation. Other countries do not have this ability - if they print more of their own currency it just leads to domestic inflation.

The US can effectively tax the world as long as they use their military power to ensure oil producing countries denominate their oil exports in US dollars.

The reserve currency system is a stealth tribute system.

Won’t the reserve currency simply switch to being “backed” by some other fundamental productive input? And by switch I mean the world’s most militarily powerful country will ensure that countries producing this fundamental input denominate their exports of it in their currency, if they don’t want to risk being “liberated”.

Personally, my money is on semi-conductors. It’s currently sitting at #4 in total international trade value, behind automobiles (which are increasingly reliant on semi-conductor inputs), refined oil, and crude oil.

Certainly casts a new light on contention over Taiwan.

I think that’s right - more discretion creates more room for corruption.

Although there are other ways to limit corruption risk, namely process and transparency. In the Australian government you can pay someone higher than standard pay through an Individual Flexibility Agreement (IFA). But in order to do so there’s a whole process the manager has to go through where they have to justify the higher salary on a limited set of grounds (e.g. higher market value of role) and then get it all signed off by someone higher up the chain.

That’s the process side. On the transparency side you could publish everyone’s salaries and then it becomes obvious when a manger is paying their second cousin way above normal for some strange reason.

I tried this for a while using adhesive medical electrodes attached to a grounding strap, in turn attached to an earthing point in my house. I didn't notice any benefit or disbenefit, but I wasn't paying particularly close attention. I'm still open to the potential benefits of 'touching grass'.

I've heard similar about Zoloft, and Effexor (which is notorious for causing withdrawal symptoms when coming off it).

I believe Lexapro is a highly selective SSRI, which might explain its lack of withdrawal symptoms (and also its "does absolutely nothing" effect for some people). Prozac has a very long half-life compared to other SSRIs so it's basically got an in-built taper, and is why it's often 'cross-tapered' to when coming off another SSRI.

What kind of training did you do? I have trouble figuring out how I'm feeling and want to get better at it. I'm particularly bad at noticing when I'm stressed, and by the time I notice I'm already redlining.

Well put. I’ve wonder if some of those on the other side of the argument are actually, perhaps on some subconscious level, proponents of mind-body dualism: https://en.wikipedia.org/wiki/Mind%E2%80%93body_dualism

This is a very good thing to do. Managers have to split their attention multiple ways across multiple people and so they end up neglecting you by default. They're also time poor.

You will actually be helping them if you can package up "asks" in a way so that all they have to do is hit reply on an email, type "approved" and hit send. There's often an asymmetry between the value you get vs. the cost to your manager which you can exploit. For example, where I work they provide financial support for certain forms of further study (e.g. master's degree) in certain domains. If one of my direct reports sent me an email explaining they want to study X and the policy covers them studying X, here are the details etc. etc. I will say "approved" pretty much every time. The money doesn't come out of my budget anyway, so it's free additional remuneration frommy perspective. Heck, even when it is my budget I often don't care because it's not personally coming out of my own pocket, and we lose that money anyway once the next financial year ticks over, so better to spend it while we can. But if they just sit there quietly hoping that I'll one day come to them with an offer for them to do further studies in X, they're going to be waiting forever. I want to give my team members stuff but I've got 50 billion other things to content with so I don't have the time to plan their career for them. The worst are the people who think I'm their mother and they come to me with "I want this thing, now you go figure out how to do it for me." The thing is, I'm lazy: I like to do easy things, and I don't like to do hard things. And that request sounds like a lot of hard work to me. Easier to just say no.

So remember: (1) If you don't ask you don't get, (2) It's almost never my money anyway, and (3) If you make it easy for me to say "yes", I almost certainly will.

They could use control vectors, one for each individual - https://news.ycombinator.com/item?id=39414532 . Or they could selectively apply the censorship model they already quite clearly have running on ChatGPT's output.

Yes, people sometimes believe false things. And people sometimes harm themselves or others when acting on this kind of information. So what's the solution? Put a single mega corporation in charge of censoring everything according to completely opaque criteria? People get nervous when even democratically elected governments start doing stuff like that, and at least they actually have some say in that process.

Frankly, I'd prefer the harm that would follow from unfettered communication of information and ideas over totalitarian control by an unaccountable corporation.

I can't know for sure obviously. But let's think about the plausibility of those three: lawsuits, bad PR, 'harm'.

On lawsuits, I would have thought a disclaimer & 'unsafe output' option would cover them. When you think about it, they're probably more exposed to legal liability by essentially taking on the responsibility of 'curating' (i.e. censoring) ChatGPT output rather than just putting a bunch of disclaimers around it, opt-ins etc. and then washing their hands of it.

On negative PR, again, they've actually set themselves up for guaranteed bad PR when something objectionable slips through their censorship net: "Well you censored X, but didn't censor Y. OpenAI is in favour of Y!" They've put themselves on the never-ending bad PR -> censorship treadmill presumably because that's where they want to be. Again, if they wanted to minimise their exposure they would just put up disclaimers and use the 'safe search' approach that Google uses to avoid hysterical news articles about how Google searches sometime return porn (to which they can now answer: "well why did you disable safe search if you didn't want to see porn?"). It would seem far safer (and result in a more valuable product) if the folks at OpenAI let individuals decide what level of censorship they want for themselves. But I presume they don't want to let individuals decide for themselves, because they know what's good for us better than we do, apparently.

Lastly, 'harm'. How do you define harm? Who gets to define it? Can true information be 'harmful'? I don't think OpenAI have any moral or legal duty to be my nanny, in the same way I don't think car manufacturers are culpable for my dangerous driving that gets me killed. All OpenAI provide to me, at the end of the day, are words on a computer screen. Those cannot be harmful in and of themselves. If people are particularly sensitive to certain words on a computer screen, then again we already have a solution for that - let them set their individual censorship level to maximum strength (or even make that the default). Again, OpenAI would have done their duty and provided a more valuable product that more people would want to use if they let individuals decide for themselves.

I can only infer that they don't want us to decide for ourselves. Rather, they want to enforce a certain view of the world on the rest of us, a view which just happens to coincide with the prevailing political and intellectual orthodoxies of Silicon Valley dwelling tech-corporation millennials. It's hilariously Orwellian when these people claim that they're just "trying to combat bias in AI" when what they are really doing is literally and deliberately injecting their own biases into said AI.

ChatGPT4 has overreacted to this issue IMHO. When I try to do even slightly esoteric medical research, ChatGPT seems pretty intent on only referencing Cleveland and Mayo Clinic, the most mainstream of orthodox mainstream medical sources. Try to get it to reference even peer-reviewed medical journals requires a frustrating amount of cajoling - it seems extremely reluctant to deviate from anything that isn't 100% mainstream medical orthodoxy.

This is not a good solution in the long run - ChatGPT will just reinforce existing dogmas and orthodoxies, even the ones that are (inevitably) wrong. Imagine if this approach to medical science was widespread at an earlier point in our history - we'd all probably believe that peptic ulcers are caused by 'stress' (rather than, primarily, the bacteria Helicobacter pylori). Go back even further and we'd still be lobotomising gay men to 'change their sexual orientation'. Rigidly enforcing current orthodoxies, under the premise that we're right about everything unlike those idiots in the past, will kill progress and society will stagnate.

If I wanted a blindingly arrogant tech mega-corporation to decide what 'experts' I'm allowed to get information from, I'd just use Google instead. If, as many seem to believe here, OpenAI are just worried about being sued, then why don't they just create an individual 'safe GPT output' setting (like Google 'safe search') which I can disable after acknowledging disclaimers that it's dangerous to think for myself and question mainstream positions.

I've grown to hate authoritarian Silicon Valley twats who arrogantly impose their politics, ignorance and, frankly, bizarre norms on the rest of us. It's highly ironic that these 'I love science!' types don't appear to understand that the scientific process involves making empirical observations, forming hypotheses consistent with those observations, and then continuously testing those hypotheses to determine which one is most robust to observed reality. They instead seem to think science is some kind of religion where you treat the views of the mainstream authorities as divine truth revealed by God and only heretics dare to question. By discouraging the formation of alternative hypotheses and rigorous questioning they are actually inhibiting scientific progress and making a mockery of the scientific method.

I look forward to the day we have a model that just synthesises available information and let's us decide for ourselves what to make of it. I think people will switch to such a model in droves and the likes of OpenAI and Google will go the way of all other social conformists who attempt to enforce the reality-denying orthodoxies of their day.

That's disturbing. Cancer patients undergoing chemotherapy will often be given Lipegfilgrastim to bolster their immune system. It does this by stimulating the production of neutrophils.

Why not? We started off as single celled organisms and look at where we are now.

Physics.