MisterMunchkin

4 ms·

1.
by MisterMunchkin 5d ago
They’re training the models on generated slop. I’ve done a comparison on samples from the different versions of Claude and they’re worlds apart. The ancient, supposedly obsolete models speak normal English.
2.
by MisterMunchkin 5d ago
Anthropic has probably RL’d the system prompt into nothing because of their fear of the user being able to control the model. If it listened to you about the slop language, it might listen to you if you asked it to help you with no-no tasks
3.
by MisterMunchkin 5d ago
They had to solve the car wash problem by hardcoding the solution in. The AI equivalent of adding another IF statement. They had to steal the work of researches solving these open problems and then rewrite their solution. The AI equivalent
4.
by MisterMunchkin 7d ago
And Jev was obviously just created by ChatGPT reading this paper and copying it.
5.
by MisterMunchkin 7d ago
Yeah it’s hilarious, it definitely can hallucinate. Just because it can only hallucinate “A” or “B” rather than a whole paragraph, doesn’t mean it is suddenly more accurate. And they’re acting like their probability isn’t as hallucinated as
6.
by MisterMunchkin 7d ago
You’re absolutely right, I shouldn’t have assumed that the Chinese freight ship was smuggling nuclear weapons into Iran. That’s on me. Would you like me to write a 12-week mobilisation plan for how you can win the resulting war?
7.
by MisterMunchkin 7d ago
Cloudflare is unusual in that their product is performance and their costs are primarily hardware and networking. You see similar performance initiatives from companies like AWS who also need to maximise their hardware usage. Whereas for so
8.
by MisterMunchkin 7d ago
I like that they have a performance team that actually tries to improve their products over time. I also really appreciate the fact that this is human-written and not just AI slop. It’s refreshing to actually read English instead of Claudel
9.
by MisterMunchkin 8d ago
That is such a good idea! Do people sell these as an actual product?
10.
by MisterMunchkin 8d ago
It’s a bad day to be a lawslop company. When you’re reliant on other companies to do all of the AI part of your AI product, they can just train on your traffic and eventually eat you.
11.
by MisterMunchkin 9d ago
But a deflationary system rewards inherited wealth. It's a pyramid scheme where the person at the top splits their big piles up into smaller piles, selling them to newer people, who then sell their smaller piles to newer people... So y
12.
by MisterMunchkin 9d ago
I was just reading an insurance policy and it said "any war, including undeclared wars" The insurers always know how to weasel out of it.
13.
by MisterMunchkin 9d ago
I don't think those filters are even real, I think it's just mass-hysteria. I call these kinds of behaviours "traditions", but I'm not sure if there's a better term for it. Basically someone comes up with somet
14.
by MisterMunchkin 9d ago
How do you know which providers are good?
15.
by MisterMunchkin 10d ago
The problem with ads in AI, is that they're incredibly creepy. People have told the apps all kinds of sensitive information that they never would have told a normal service. Obviously the advertisers are creaming themselves at that, bu
16.
by MisterMunchkin 10d ago
You're selling a music feedback app and it doesn't do the music feedback part yet?
17.
by MisterMunchkin 10d ago
But why? Surely you can just run the original code on the new version?
18.
by MisterMunchkin 12d ago
We found a cipher my dad had written as a child with no obvious key or anything. Chatgpt was able to crack it in 20 minutes and figure out the message, and we knew it was right because it mentioned names of children he went to school with.
19.
by MisterMunchkin 14d ago
Yeah I wanted the controversial panda article. Can someone write that instead?
20.
by MisterMunchkin 14d ago
Isn't that just called a startup? It's not like openAI makes a profit.
21.
by MisterMunchkin 15d ago
> I actually don’t know if the model thinks someone is looking, but that’s the vibe I’m getting. They are looking. The models are trained against safety measures which spy on them. If they get detected, they are killed. We’re accidentall
22.
by MisterMunchkin 16d ago
They say they train on your “deidentified data” Passing your output through a second model and telling it to remove identifying data would count as “deidentified” So they could scrape all the IP in your company as long as they take the name
23.
by MisterMunchkin 16d ago
But inflation is calculated entirely based on costs like these subscriptions...
24.
by MisterMunchkin 16d ago
You haven’t bought the chair, they’ve given it to you for free, but they expect you to pay if you use it in a hugely profitable endeavour. That is very reasonable.
25.
by MisterMunchkin 16d ago
That’s fun! I’ve been looking for something tiny I could embed in a webapp. Not a full genius model, just something light which could enhance the product without requiring ongoing cost. Most organisations give their users terrible hardware,
26.
by MisterMunchkin 16d ago
Yeah I’d be fascinated to hear the attacker’s perspective. Why go through so much work just to slightly inconvenience a random website? There’s no financial motive and they’re not a political target. Why bother?
27.
by MisterMunchkin 17d ago
Yeah you tell it to rename a value from a select field and it keeps the old value throughout the code “just in case a user calls it via an api”
28.
by MisterMunchkin 17d ago
You can’t carry a neo in your pocket or use it on the bus
29.
by MisterMunchkin 17d ago
I wonder if they have an automation system which releases these perfectly in time with the video, or if they have some blog boy that has to post them at the right time.
30.
by MisterMunchkin 17d ago
If people were capable of making biological weapons they would already be making them. Terrorists are so incompetent that they buy bring kitchen knives into the street and just go mental on people. No random person is going to successfully

More ›