411–420 of 438 posts
Earlier quoted context omitted.
Free fall? None of this recent Mythos / Fable drama is going to even slow Anthropic down. The HN bubble strikes again.
It's certainly not going to affect their valuation, but their reputation as the "ethical" AI company is gone.
Earlier quoted context omitted.
Today I told Sonnet (!) to use a browser MCP to enter a username and password for the project it is working on, it told me that it can't do that because it violates its security protocol. This worked fine before. I love Claude, I have stuck with it even through people saying Codex is better but this is definitely getting to be the last straw. It's completely absurd I am paying them $200+ per month along with pushing…
Are you sure it refused because it can't use a username and password? I literally have loops running right now where it uses a database of test users and passwords to log into different roles and do computer use and browser automation testing. Sonnet and Opus complain when I provide credentials and password in chats but it is happy to use ones stored in files and stuff, so it might just be guardrails to push good ops…
It's even gone as far as, on other projects, creating its own test accounts and, without prompting, getting into the local dev database to mark its accounts as verified without being told to do that or that it was allowed.
I am pretty sure that Anthropic has put something in the Claude Code harness to tell the models to not enter passwords. Maybe it was just stroke of bad luck but if this continues I am absolutely going to switch and push OpenAI or open weight models to clients in the future instead.
Earlier quoted context omitted.
Today I told Sonnet (!) to use a browser MCP to enter a username and password for the project it is working on, it told me that it can't do that because it violates its security protocol. This worked fine before. I love Claude, I have stuck with it even through people saying Codex is better but this is definitely getting to be the last straw. It's completely absurd I am paying them $200+ per month along with pushing…
This is funny. Did you try using playwright mcp
Earlier quoted context omitted.
Imagine a nuclear power company being extremely cautious with safety and everyone online is like "make more power already you pussies!"
Sorry this is a very bad analogy, nuclear power isn't the same as a LLM, and definitely not as dangerous either. This is ridiculous.
Isn’t that the point of an analogy?
Earlier quoted context omitted.
i think its a combination of these: - OAIs models are much more token efficient. in some cases their new mythos-level model uses 1/4th of the tokens that mythos does. - they reportedly have recently found another way to decrease compute needed "by half" in a breakthrough - early on, they ensured they have much more compute ready, whereas Anthropic had to disable new users signing up because they ran out of compute -…
Link to source for the breakthrough in inference compute? Are you talking about prefill?
Earlier quoted context omitted.
And that's exactly why 3 years ago we saw people one shotting it in 45 minutes :)
"Some idiot on Twitter did a game tech demo" has been a thing since, like GPT 3.5 or 4.0?
As the generations increase, so too has the complexity of the games it can one-shot.
It is surprisingly half baked for something the company seems to be proudly charging my first born child and both of my kidneys for. Especially considering its all for the privilege of getting half way through a standard dev task before it just bricks out of nowhere on the 'safety' guardrails because of a keyword that often appears to be created by the model itself? I've been running Fable on a few different projects…
Imagine a nuclear power company being extremely cautious with safety and everyone online is like "make more power already you pussies!"
“Safety” in the context of LLMs concerns not offending certain individuals.
Earlier quoted context omitted.
Sorry this is a very bad analogy, nuclear power isn't the same as a LLM, and definitely not as dangerous either. This is ridiculous.
Sorry this is a very bad analogy, nuclear power isn't the same as a LLM Isn’t that the point of an analogy?
(Seems like maybe that’s not the point of an analogy).
A frontier lab releasing the most advanced model causes them to lose customers was not on my AI bingo card.
Earlier quoted context omitted.
So... it IS China, not just companies in China. And why would the CCP want to give out open weights, hmm? Because they are all kindness and love? No. It is just their approach to counter dominance in the space by Western frontier labs. No need to make moral attributions..it is everything, and I mean everything, about Great Power Competition.
Their self serving reasons are irrelevant. The results can be moral even if the players aren't. The key is to somehow prevent any one of them from dominating completely so we can play them off one another and keep reaping the rewards. If China's undermining efforts achieves that, so be it. The fact we're getting permanent access to powerful models as a result is nothing short of awesome. In order for us to win, neith…