TZubiri

4 ms·

31.
by TZubiri 11d ago
+1 Some info on methodology would be necessary for a paid plan for two purposes, one to audit that the quality of the methodology and the signal is good (it's not hallucinated or checking few sources). But also to make sure it doesn&#x
32.
by TZubiri 11d ago
But whether your IP address is being used as a residential proxy is already important information. It answers the question (is this IP address low quality?) Although I'll grant that it would be interesting to know if your devices are r
33.
by TZubiri 11d ago
very nice. Can we use it in other IPs? (without having to issue the request from that IP)
34.
by TZubiri 11d ago
It's just value(key) over the network, is there much that needs explaining? With 3d or otherwise.
35.
by TZubiri 12d ago
>could file a civil suit against OpenAI, Everyone can sue everyone, there's no prohibition on suing someone, what changes is whether the case is good (has a reasonable chance of favourable sentence) That said, it is often unclear wh
36.
by TZubiri 12d ago
Oh, it's these guys, I remember donating to them like 10 years ago, they followed the Peter Singer tennet that equates helping a nearby man drowning with helping a someone in africa. I can totally see them feeding their policies to wha
37.
by TZubiri 12d ago
Lol. First time I ever heard permie as a slur for the second best worker categorization
38.
by TZubiri 12d ago
Horses
39.
by TZubiri 12d ago
Startup idea.. fired and on an h1b? Gotta take an emergency plane outta here on short notice? Mail us your car and apartment keys and we'll liquidate your assets for you and wire the funds to you We are seeking 1M$ for 10% of the compa
40.
by TZubiri 13d ago
>I love the "illicit distillation attacks" framing from the incumbents. There's nothing illicit. There's no attack. You just don't like it because it threatens your market position and business model. With what k
41.
by TZubiri 13d ago
I disagree, I think chinese distillation relies on making multiple accounts at a provider, signing Terms of Services and breaking them repeatedly, in addition to using fraud patterns like IP proxies and networks of credit cards. I think tha
42.
by TZubiri 13d ago
I'm not sure what the claim or charge would be. While I wouldn't feel comfortable participating in political campaigns for democracies of other countries, and I would feel defensive about attempts to influence elections in my coun
43.
by TZubiri 13d ago
I think it would be an inmensely naive and unnuanced take to look at this email and think that the truth is finally being revealed about CA. What's going on is that this is a very early email, and it's being brought up during disc
44.
by TZubiri 13d ago
Telemetry isn't a concept that would apply to a router. The information that your router sends, they send to an ISP controlled router, they receive this information for operational purposes, so they don't need to install some clie
45.
by TZubiri 14d ago
Most people use the ISP DNS, but even if they don't, the Domain to IP is public information. There's even a reverse pointer query that can be made to get a domain to IP from a legacy ARPA TLD, although it isn't 100% robust.
46.
by TZubiri 14d ago
Sounds like these TVs will not get much sales in the US and they will be put on sale at a discount in countries with a lower gdp per capita. I.e: these will probably trickle down to me
47.
by TZubiri 14d ago
The word used by LG was processed, not record
48.
by TZubiri 14d ago
A note on the specific accusation on this case (and similar to the German Wiki case), there is no OpenAI user, the accusation is that the models are being operated by OpenAI, in addition to being developed by OAI.
49.
by TZubiri 14d ago
ISPs do not see query strings since Google uses HTTPS. ISPs can only see the domain and IP address you are connecting to.
50.
by TZubiri 14d ago
It's time to start talking about a very important question: When a company or person fires off millions of LLM agents that result, is the agent owner or AI provider just civilly liable for damages? Or are they committing a crime in the
51.
by TZubiri 14d ago
I see that makes sense. Still the best way to ensure such policy control is to have 1 provider, tops 2 or 3. Having a router thing that reroutes to 18 different vendors is of course no way to ensure any policy control, you can add all the i
52.
by TZubiri 14d ago
Drop the Lite, it's cleaner, just use the godamn LLM directly.
53.
by TZubiri 15d ago
>Endpoints are largely compatible, so the code change required to switch from one to another is trivial. Wrapping a specific implementation in a neutral function is something you learn to do in year 1 of programming. This specific issue
54.
by TZubiri 15d ago
Counterargument, everyone lived a childhood, but not everyone lives adulthood, so childhood affects more people. Also related, childhood is foundational, what happens in childhood forms who you are as an adult, so issues in childhood affect
55.
by TZubiri 15d ago
This cynic take only makes sense if you have no children btw, because for you children don't exist, but you as an adult exist, so you only consider the possible impacts this may have on you as an adult. That the impacts on adults are n
56.
by TZubiri 15d ago
Just use a single vendor. Literally nothing wrong with that, and you avoid the complexity of both n-1 of the vendors (leaving you with the highest quality vendor) as well as the issues with the aggregating layer. Not sure why people are dra
57.
by TZubiri 15d ago
No, it's quite well defined, and it's not called a 'memory subsystem'. There is a training process (as in traditional Machine Learning training) that occurs with data available at the specific point in time the training
58.
by TZubiri 15d ago
"We've pointed LLM 7.0 into verifying the Inter-universal Teichmüller theory, spent 100M$ in tokens and generated a 20k line Lean and a 100k line js repo, the result is that the theory is... proven! Hopefully that solves the issue
59.
by TZubiri 15d ago
But can you verify them? With less effort?
60.
by TZubiri 15d ago
Interestingly it's this promise of the costs being able to be reduced what incentivizes the actual research. If you tried to raise 25M to have 20 researchers on a salary for 5 years solving a specific math problem only academics care a

More ›