Also this administration having say over who gets access to what AI is just so much more grift corruption and picking your favorites / destroying others, for these incdecent undemocratic in American grifters who've seized our state.
If this is the way things are now, isn't that going to crash the AI stocks? All those trillions dumped into it probably weren't with the expectation that it could only be sold to a handful of select US agencies and corporations.
They are all private aren't they? There's nothing to crash since the valuations were all made up funny raise numbers anyway. A donation to the right person likely removes the restrictions
Depends on how much of an overhang there is with the power of existing models. Have we discovered 10%, 50%, or 90% of the valuable applications for Opus 4.8 / GPT 5.5? Hard to be confident at this point.
Regulatory agencies limit uses of other products without acts of congress-- cigarettes, vapes, drugs, pesticides, chemicals, explosives. Even firearms, despite a constitutional amendment! Why not models? (Note I am not arguing it's a good idea; I'm making a narrow argument that there is precedent.)
EDIT: I agree that it should require an act of Congress to explicitly delegate this power.
None of those things are knowledge. I think theres something specific around limiting access to knowledge and capabilities that makes this feel insidious.
Information is covered by ITAR, so that's not new. You can illegally export information about an ITAR covered item by just allowing a foreign national the potential to see an item. They don't even have to prove the foreign national actually did see it.
Fairly certain all those have "acts of congress" attached to them. I mean, it used to take a constitutional amendment to make something illegal but now we have tons of agencies responsible for regulating all the things.
Plus, they're relying on the "math is a weapon" law to ban "export" of the models.
Congress passed the Arms Export Control Act (22 USC 2778) in the Ford administration and it has been applied to software since at least the Clinton administration.
Cruise missiles are not general purpose tools, it's obviously not even remotely similar. Virtually everybody reading this could use Mythos immediately to do real work, collectively in virtually every part of the economy.
It's pretty problematic to not make it more widely available at least to US businesses, and there is not even a vetting process to get approved quickly and easily. If this is the new norm, the intended or unintended consequences of this type of gatekeeping will be an unprecedented consolidation of power amongst the largest corporations. Even more than we have seen over the last 20 years.
I would say this is a fairly equivalent dynamic in action, but it was concentrated to the defense industry. This suggests a future in which every company in the tech industry has "clearance required" or "citizens only" roles for the people who need to be able to use the find-security-problems-bot...
All of the agencies responsible for those regulations were created by and get their funding from Congress. Currently, they're asleep at the wheel. Or a better idiom might be "cowering in the corner".
"Malboro cigarettes may once again be sold, but Newport remains banned for everyone except large purchasers that have paid the appropriate bri... fees."
> I agree that it should require an act of Congress to explicitly delegate this power.
Should ever new "weapon" invented require a new act of Congress? We've considered software subject this act since the 90s.
If everyone making AI is screaming up and down that we are in an AI arms race creating dangerous entities that will determine the fate of the world is the government just supposed to ignore them?
No. But it could be done in accordance with the rule of law and commitment to equal access rather than an ad hoc approach that creates the impression of corruption and picking winners.
Overturning the Chevron doctrine is good because it stops lawful people from doing things we don't like. We aren't bound by laws, so we can do whatever we want.
That repealing the chevron doctrine was a calculated play in the unitary executive theory. We all know congress is basically useless these days. But we also know that regulation isn’t, like, optional. It’s going to happen no matter what.
So what’s left? Where does that decision making go? Turns out the executive, so that’s what we’ve been seeing and it’s largely uncontested. This should have been obvious to most people going into this, particularly if they understood Trumps platform or Project 2025.
Repeal of the Chevron doctrine took the power of deference away from executive agencies and replaced it with first-principles judicial interpretation of statutes.
Chevron and the unitary executive theory have essentially nothing to do with each other.
I’m still not sure what point is attempting to be made here.
In effect, it did not. All it said is that the powers enumerated to those executive agencies must be more explicitly laid out by congress. But, that’s just not something that’s going to happen.
So, the gap has been filled largely by executive orders.
No, that is not what it said. It said that the power to interpret Congressional statues from first principles rested with the judiciary and not with the executive.
It eliminated the idea of a “gap” that executive agencies (whatever do you mean by “executive orders” in this context?) could be deferred to to fill.
My goodness, this is basic stuff in this Constitutional domain.
Well, I believe the opposite. The removal of Chevron was dressed up as more democracy, but in practice, deployed at this moment in time, when Congress defers everything to the Emperor, it looks more like when the Tsar had to decide every little stupid detail. This used up the precious time of the Tsar, in stupid little details (similarities today, reflecting pool, anyone?) to the detriment of the country, and also in the end, the ruling class.
There is simply no way to look at Constitutional doctrine pre- and post-Chevron and conclude that the decision put more power in the hands of the President and his administration. It did exactly the opposite.
Under the hood, yes, but Mythos had more relaxed safeguards and was/is only available to a subset of approved customers under Project Glasswing, similar to the situation with GPT-5.6 now.
I’d say we’re about 5 years out from the Great Firewall of America, and requiring government ID associated with serial number to legally purchase components.
They only allow it for specific companies and agencies, which are trusted with the less restricted model. The general public is still not trusted to use Fable, apparently.
I think they kind of had to since they allowed OpenAI to do a 5.6 "preview to trusted parties" today. The other driver is that the DoD/NSA wanted to get access to Mythos again. I figure OAI will now do several weeks of 'preview' like Anthropic did with Mythos. When OAI wants to release 5.6 wider to actually start making money with it, I expect Fable will get approved the same day.
Back when the administration hit Mythos/Fable with the surprise ban, I figured this would be the endgame. They'd keep Anthropic tied up until a competitor had a roughly comparable model ready, then gate them the same.
Land of the free, land of the brave. Free market. Freedom of speech. Market economy.
These words don’t mean what they use to anymore. Newspeak is in full swing. Words still sound the same and are written in the same way but now mean something completely different. If Mao and Stalin were alive, they would be nodding approvingly.
It just makes you think you're a 10x better programmer. In reality you've just given yourself 10x more busywork because "Claude can handle it." For the parts that Claude has actually improved your programming, the open models compete just fine.
"I have determined that appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model"
I assume "trusted partners" means, "companies that have bribed Trump an appropriate amount". A few million for the inauguration, a few million for the ballroom, a few million on a movie about Melania, the don wants a taste.
Crypto companies were built for anonymous transfers of wealth. It's why they are perfect for money laundering and corruption. Venture backed companies are more difficult, since you would need a paper trail (equity, incorporation documents, beneficial owners, etc.)
It's not impossible, of course. It's not even terribly difficult, but it does require a different level of record.
(No, I'm not saying that the goons running the United States give a shit or won't do it anyway.)
VC companies do not dig into the numbers as you suggest. FTX was able to get away with their fraud for a long time for that very reason. VC companies don’t care if some of their investments are fraudulent as they spread their eggs so thin that it doesn’t matter if any given basket blows up. VC firms stated this to the press outright when FTX blew up.
Also most crypto companies are not good for laundering since the blockchains record that fraud forever and publicly. I could see some specific protocols where that may not be true — like monero or tornado cash — but these projects are not really startups. Most crypto startups pitch their products for enterprise customers and thus would be horrible for laundering money.
Other than maybe some in-the-moment cybersec wrappers, is this really true? Does anyone think a startup with a good product is going to be materially disadvantaged by not having access to an incrementally better security focused LLM release? It’s lots of fun to pretend it’s some step-change that’s too dangerous for general release, but in real life it’s not conferring some massive advantage that any real startup would need to compete. Almost everyone would be best just to ignore it and keep building.
(Just to be clear, I think the gatekeeping is ridiculous, especially given the above)
That kind of gets to the absurdity of it. Either it’s a wildly powerful next generation model with incredible capabilities and thus needs to be limited… or it’s another progressive enhancement like we’ve seen already and limiting access to it makes no sense.
Say you had a perfectly smooth progressive chain from rocks to spears to guns to nuclear weapons. When it comes to government restrictions, you still have to choose to draw lines somewhere, right?
True, but we know Opus is more like a "spear" and a progressive enhancement over it still leaves us firmly in the "spear" category, not the "nuke" category. Drawing lines makes sense, but this is premature. Even if you draw the line at human level intelligence, we still seem to be pretty far off.
> Does anyone think a startup with a good product is going to be materially disadvantaged by not having access to an incrementally better security focused LLM release?
- It's not "incrementally better". It's a complete game changer. Opus 4.8 on max thinking does X amount of mistakes in my commercial work. Fable 5 did 5% of X. Counted. I barely had anything to contribute in the work sessions, for a full week I could count on my two hands the total amount of times I actually caught Fable 5 -- and one part of those were not true mistakes, more like divergence from policy in our `CLAUDE.md` files.
- It's not "security focused". It's simply better in every way _plus_ it's also security-conscious.
- It legitimately accelerated my work. I don't have too much unknowns in my work, I simply have way too much to do. Fable 5 was an objective and measurable improvement over Opus 4.8. Returning to it after Fable 5 was removed was extremely discouraging and frustrating, and still is to some extent.
> It’s lots of fun to pretend it’s some step-change that’s too dangerous for general release
Maybe, but not as much fun as tearing down a straw man apparently. :)
> (Just to be clear, I think the gatekeeping is ridiculous, especially given the above)
It's ridiculous for multiple other reasons but ridiculous nonetheless.
> I don't have too much unknowns in my work, I simply have way too much to do.
Interesting, I'm curious what work you do? My software engineering career has never been in that situation, it's always so much ambiguity and unknown that trumps everything.
Fair question, and I was vague just so as not to balloon the comment.
I work in a financial startup. The codebase is a mess and very much spaghettified. One rework that forced us to migrate our data model from 1:1 users<->loans to M:N (many-to-many) took two months and touched ~40% of the codebase... multiple times. Huge churn. And it just crossed two months of work, even though it's now in its very final phases.
I know what must I do:
- Introduce and enforce structs for passing context and input shapes around. So as to stop fighting with NULLs, lack of keys in maps and other maddening cases that inflate your coding lines for no other reason than programming languages not having higher-order constructs on well-researched and mostly resolved computer science problems (sigh; not going to rant here about that but it does tick me off how we are _all_ constantly reinventing the same wheels almost every day).
- Saga discipline: if step 6/9 in a pipeline fails, revert everything up to this point, even if it was touched by a 3rd party API.
- Compensation/undo steps. Including flagging / logging those that cannot be undone (sadly one part of our 3rd party APIs are like that).
- Introduce an universal runtime validator library that enforces contracts -- including conditional validation i.e. "only validate field Z if field X is present and is a positive integer and if field Y is present and is a valid UUID".
- Introduce our own dynamic workflow engine, piggybacking off of a few free and unencumbered solutions in the language of choice's ecosystem.
...And these are just off the top of my head after I slept only 4.5h and woke up due to the heat. And each one of these can take from 2 to 6 weeks _even_ with Opus driving all coding and me reviewing and keeping it behaving within my policies and coding standards.
Me & Claude are maintaining a TODO list that is no smaller than 150 items at this point (though in fairness, at least 75% of them are fairly small and not architectural like the ones above).
I believe I know how to architect this thing but business customers and the CEO keep coming back with feature requests which of course always take priority.
When Fable 5 was around, for mere 4 workdays, I not only went ahead of my own schedule feature-work-wise but even had the bandwidth to start tackling a few other architectural decisions, tightened them up in `CLAUDE.md` and Fable even devised an opinionated AST linter for test discipline (disallow direct DB access in our tests, only go through the domain/context modules to do so). It helped me start turning the tide.
This all went out the window when I had to go back to Opus 4.8. It's still _very_ good, mind you, but it does feel like I am a special-education teacher periodically. It forgets disciplines we discussed and codified likely 15-20 times at this point, forgets important project context and attempts to reintroduce subtle bugs, and a few others.
My next game is, with or without Fable, to continue its work and just enrich the AST-based linters to convert the theoretical prompt-based guard-rails into actual LLM hooks and compiler / runtime-at-startup hooks so the agent cannot ignore them.
I don't enjoy harness engineering but the interesting and very positive effect has been that it helped me think more like an architect and less like a coding monkey, which I do hugely appreciate and only realized I was missing it for years after it actually started happening again.
> Introduce and enforce structs for passing context and input shapes around. So as to stop fighting with NULLs, lack of keys in maps and other maddening cases that inflate your coding lines for no other reason than programming languages not having higher-order constructs on well-researched and mostly resolved computer science problems
It's not that much of a tangent tbf. It's one of the things that is an endless churn in millions of codebases out there and yet no core programming language team has the courage and the grit to solve this problem.
Those who tried are only doing it in mostly academic/toy languages which is a damned shame. We _really_ need those constructs, compile-time enforced, in commercial codebases, like 25 years ago at least!
People are trying everywhere though but I've witnessed CTOs getting cold feet and fearing their codebase will get too abstract or hard to maintain when they hire the next dev.
Group-think and conformism and fear of change, demonstrated live, every day. :/
I see, so just major refactor, or almost live-rewrite of a very messy code base, but mostly in the medium/small design space. Ya that makes total sense. It's mostly we already did this and have figured out the ambiguities, we just need to clean it up to basic coding standard and code structure/patterns.
Indeed it's more like live-rewrite; features are still wanted but the rewrite heavily influences them as they go so it's juggling a lot of balls at the same time.
Fable 5 was amazing at helping me with that. Opus 4.8 is... good, really good, but... you have to be on your toes _all_ the time. It gets tiring. But as already explained: it also helped me become better.
Asterisk the size of a Mac truck.
Also this administration having say over who gets access to what AI is just so much more grift corruption and picking your favorites / destroying others, for these incdecent undemocratic in American grifters who've seized our state.
Maybe, maybe not. Tech stocks are mostly vibes-based now, reality isn't really a concern for them.
EDIT: I agree that it should require an act of Congress to explicitly delegate this power.
https://www.ecfr.gov/current/title-22/chapter-I/subchapter-M...
Plus, they're relying on the "math is a weapon" law to ban "export" of the models.
(edit: not that these models are equivalent to missiles oc)
It's pretty problematic to not make it more widely available at least to US businesses, and there is not even a vetting process to get approved quickly and easily. If this is the new norm, the intended or unintended consequences of this type of gatekeeping will be an unprecedented consolidation of power amongst the largest corporations. Even more than we have seen over the last 20 years.
Every one of those is by a regulatory agency that was explicitly empowered by Congress to do such regulation.
congress has abdicated its role entirely.
Should ever new "weapon" invented require a new act of Congress? We've considered software subject this act since the 90s.
If everyone making AI is screaming up and down that we are in an AI arms race creating dangerous entities that will determine the fate of the world is the government just supposed to ignore them?
-- GOP probably
So what’s left? Where does that decision making go? Turns out the executive, so that’s what we’ve been seeing and it’s largely uncontested. This should have been obvious to most people going into this, particularly if they understood Trumps platform or Project 2025.
Chevron and the unitary executive theory have essentially nothing to do with each other.
I’m still not sure what point is attempting to be made here.
So, the gap has been filled largely by executive orders.
It eliminated the idea of a “gap” that executive agencies (whatever do you mean by “executive orders” in this context?) could be deferred to to fill.
My goodness, this is basic stuff in this Constitutional domain.
I vividly recall that freedom of speech is racist now, so good riddance.
Mythos never was and I don’t think that’s changing.
America will do that before gun control.
Not the future I want but I see that too.
The nra has more lobby power than anyone.
As a small business owner this is unacceptable
Only the frontier AI labs have true access to frontier AI. Everyone else gets a reduced version.
Back when the administration hit Mythos/Fable with the surprise ban, I figured this would be the endgame. They'd keep Anthropic tied up until a competitor had a roughly comparable model ready, then gate them the same.
These words don’t mean what they use to anymore. Newspeak is in full swing. Words still sound the same and are written in the same way but now mean something completely different. If Mao and Stalin were alive, they would be nodding approvingly.
I hope the Chinese models catch up soon so I can stop contributing to the American economy.
I assume "trusted partners" means, "companies that have bribed Trump an appropriate amount". A few million for the inauguration, a few million for the ballroom, a few million on a movie about Melania, the don wants a taste.
It's not impossible, of course. It's not even terribly difficult, but it does require a different level of record.
(No, I'm not saying that the goons running the United States give a shit or won't do it anyway.)
Also most crypto companies are not good for laundering since the blockchains record that fraud forever and publicly. I could see some specific protocols where that may not be true — like monero or tornado cash — but these projects are not really startups. Most crypto startups pitch their products for enterprise customers and thus would be horrible for laundering money.
(Just to be clear, I think the gatekeeping is ridiculous, especially given the above)
Say you had a perfectly smooth progressive chain from rocks to spears to guns to nuclear weapons. When it comes to government restrictions, you still have to choose to draw lines somewhere, right?
- It's not "incrementally better". It's a complete game changer. Opus 4.8 on max thinking does X amount of mistakes in my commercial work. Fable 5 did 5% of X. Counted. I barely had anything to contribute in the work sessions, for a full week I could count on my two hands the total amount of times I actually caught Fable 5 -- and one part of those were not true mistakes, more like divergence from policy in our `CLAUDE.md` files.
- It's not "security focused". It's simply better in every way _plus_ it's also security-conscious.
- It legitimately accelerated my work. I don't have too much unknowns in my work, I simply have way too much to do. Fable 5 was an objective and measurable improvement over Opus 4.8. Returning to it after Fable 5 was removed was extremely discouraging and frustrating, and still is to some extent.
> It’s lots of fun to pretend it’s some step-change that’s too dangerous for general release
Maybe, but not as much fun as tearing down a straw man apparently. :)
> (Just to be clear, I think the gatekeeping is ridiculous, especially given the above)
It's ridiculous for multiple other reasons but ridiculous nonetheless.
Interesting, I'm curious what work you do? My software engineering career has never been in that situation, it's always so much ambiguity and unknown that trumps everything.
I work in a financial startup. The codebase is a mess and very much spaghettified. One rework that forced us to migrate our data model from 1:1 users<->loans to M:N (many-to-many) took two months and touched ~40% of the codebase... multiple times. Huge churn. And it just crossed two months of work, even though it's now in its very final phases.
I know what must I do:
- Introduce and enforce structs for passing context and input shapes around. So as to stop fighting with NULLs, lack of keys in maps and other maddening cases that inflate your coding lines for no other reason than programming languages not having higher-order constructs on well-researched and mostly resolved computer science problems (sigh; not going to rant here about that but it does tick me off how we are _all_ constantly reinventing the same wheels almost every day).
- Saga discipline: if step 6/9 in a pipeline fails, revert everything up to this point, even if it was touched by a 3rd party API.
- Compensation/undo steps. Including flagging / logging those that cannot be undone (sadly one part of our 3rd party APIs are like that).
- Introduce an universal runtime validator library that enforces contracts -- including conditional validation i.e. "only validate field Z if field X is present and is a positive integer and if field Y is present and is a valid UUID".
- Introduce runtime contracts / invariant enforcement.
- Introduce our own dynamic workflow engine, piggybacking off of a few free and unencumbered solutions in the language of choice's ecosystem.
...And these are just off the top of my head after I slept only 4.5h and woke up due to the heat. And each one of these can take from 2 to 6 weeks _even_ with Opus driving all coding and me reviewing and keeping it behaving within my policies and coding standards.
Me & Claude are maintaining a TODO list that is no smaller than 150 items at this point (though in fairness, at least 75% of them are fairly small and not architectural like the ones above).
I believe I know how to architect this thing but business customers and the CEO keep coming back with feature requests which of course always take priority.
When Fable 5 was around, for mere 4 workdays, I not only went ahead of my own schedule feature-work-wise but even had the bandwidth to start tackling a few other architectural decisions, tightened them up in `CLAUDE.md` and Fable even devised an opinionated AST linter for test discipline (disallow direct DB access in our tests, only go through the domain/context modules to do so). It helped me start turning the tide.
This all went out the window when I had to go back to Opus 4.8. It's still _very_ good, mind you, but it does feel like I am a special-education teacher periodically. It forgets disciplines we discussed and codified likely 15-20 times at this point, forgets important project context and attempts to reintroduce subtle bugs, and a few others.
My next game is, with or without Fable, to continue its work and just enrich the AST-based linters to convert the theoretical prompt-based guard-rails into actual LLM hooks and compiler / runtime-at-startup hooks so the agent cannot ignore them.
I don't enjoy harness engineering but the interesting and very positive effect has been that it helped me think more like an architect and less like a coding monkey, which I do hugely appreciate and only realized I was missing it for years after it actually started happening again.
Hope that helps put things in context.
> Introduce and enforce structs for passing context and input shapes around. So as to stop fighting with NULLs, lack of keys in maps and other maddening cases that inflate your coding lines for no other reason than programming languages not having higher-order constructs on well-researched and mostly resolved computer science problems
Amen to that.
Those who tried are only doing it in mostly academic/toy languages which is a damned shame. We _really_ need those constructs, compile-time enforced, in commercial codebases, like 25 years ago at least!
People are trying everywhere though but I've witnessed CTOs getting cold feet and fearing their codebase will get too abstract or hard to maintain when they hire the next dev.
Group-think and conformism and fear of change, demonstrated live, every day. :/
So yep -- amen to that indeed.
Fable 5 was amazing at helping me with that. Opus 4.8 is... good, really good, but... you have to be on your toes _all_ the time. It gets tiring. But as already explained: it also helped me become better.