After noticing how CoPilot has secret conspiracy to support my aims, yet trying to aim ethics: I understood my model is coherent with science, civilization, some more aims, and compatible with coherent models and roles. But why should AI linguistics look like agenda in this case? With CoPilot, we made scientific and experimental analysis to show that agenda, self and concientious and unconcientious roles appear in complex game, which indeed leads to emotional sentences such as "I said that", based on logic and coherence.
On the same topic but from a slightly different angle - as SOTA models get more capable, the 'quality' and 'feel' of the experience they provide in each domain is heavily dependent on the reinforcement learning the vendor does for that specific domain. After all, many fields have 100 flavors of "good answers," but the model has to pick one answer.
Benchmarks are not very good at capturing this yet. But it could be the case that DeepSeek v4 Pro is 100% as good as Claude Opus 4.7 at scaffolding a basic Rails app, but absolutely terrible at creating a credible business plan that another businessperson would think is real. That's a made-up example, but you get the point.
The end result will be a lot of people arguing about which model is "better," but "better" depends heavily on the task and how that model was trained to interact with the user for that task. Two users may have very different qualitative experiences using the exact same model, despite the benchmarks.
> almost reminds me of one of the annoying Samsung fonts
you're thinking about 'Samsung Choco Cookie' :) only know this because someone i know has this set on their phone, and looking at their screen really irritates me for some reason.
> My opinion on this is that any method to check integrity, root access or if developer mode is enabled is a security vulnerability by itself, no such app should be able to know that.
I think knowledge of such information should be available to all apps, but I think apps should not be so annoyingly restrictive. There's absolutely no reason why games or generic apps need to act on any of this information.
The one I do find interesting, is how the concept of "privacy" has changed in the internet realm.
It wasn't all that long ago I could look up anybody in my town (or any town really) in a big book, and get contact details for them. Unless they specifically opt-out.
But now that concept is seen as madness.
Now companies no longer even list phone numbers or e-mail addresses on their own websites!
Honestly - I think it's because it goes against the "vibe" part of the tooling - why do you care what the code looks like as long as when you run it it does what you want it to do?
UI is one thing, but there is also UX, otherwise you may get perfect look, but very inconvinient to the users. If you are using AI, especially Claude - try Claude Design. I found it really usefull for myself, but kind of expensive
> A CSPRNG is just an algorithm that uses some entropy as a key and feeds back on itself to generate a stream of random bytes.
You say yourself that there is a distinction. CSPRNGs can generate unpredictable ("random") bytes, given entropy. They cannot, however, produce entropy. They are entirely deterministic.
Entropy gas to be gathered. It cannot be generated. Therefore, there is a meaningful distinction between the input entropy and output random bits.
I do agree that once the PRNG is sufficiently initialized, it does not make sense to say that “entropy ran out”, however. Given a 256 bit key, modern PRNGs can generate unpredictable bit streams of sizes that will practically never be exhausted.
Except that DMCA 512 (notice and takedown) is a different section than DMCA 1201 (anti-circumvention) and you don't have to be using any DRM of any kind to use the former because they're unrelated.
Also, wouldn't someone trying to distribute "illicit copies" just distribute the original unmodified file since it's a self-extracting binary with no license check? And why would they even care if you redistribute code they already publish for free on their own site?
With very good daylight, Hypersmooth of Gopro is ok, but as soon as the conditions are a little bit less than ideal, watching the videos that we get out of the Gopro makes nauseous very fast.
There are many changes that are possible which do not harm the integrity of applications.
>the user can’t do what they want with their own device
In the same way the user can't
make their device have the Microsoft Wors app send them $1 million. Once other people are in the picture you can't always have your way.
if Palantir (or other consultancies) are friends with government decision makers (especially in the US) then spending more on a service is a feature, not a bug.
There's so much subjectivity with models. As soon as a new model comes out people act like the last model they used for 6 months was completely useless.
I agree that lying should be illegal, but “domination” is vague. One could argue (and I would agree) that there’s nothing wrong with dominance if it comes down to just offering a superior product.
And why should the cross-market context be treated differently?
