raxxorraxor

Both is completely secure theoretically. You can read the output of the curl before you refetch and pipe it to bash. You can also look at the commands claude is executing.

You always do that, right? ...right?

Reddit more or less distanced itself from him though. Since then the platform thoroughly degraded too in my opinion.

But yes, if the US government would want to appear at least partially "just", it would apply the same pressure here. Somehow that is very doubtful.

And again shame on reddit for being that spineless.

It isn't stupid at all to put out a message like this, correct (unlikely) or not (likely).

For several reasons. Israel will never try to win a popularity contest with moderately educated college kids that will change their opinion in a few years anyway. To do that would be entirely stupid.

For in the middle east such messages do serve a purpose. Main reason here is the election of course.

Yes, many countries have them in some form or another. Their existence is not proof of anything.

In fact there is no proof about any impact. As someone who is online too much already, it is highly doubtful the impact is significant and there are more studies pointing to the opposite. High school teens to more impressive online raids if you say you dislike k-pop. Some claim there is a background effect, but anyway...

Troll farms have an interest to overstate their impact. Are you a troll? Easy accusation, isn't it?

Turns out people that voted incorrectly on issues like Brexit where quite focused on Brexit as a multi-faceted political issue. For example they might have objected to be called victims of Russian propaganda. Which is a statement on the level of tea leaf reading.

That highly depends on the receiver. Currently in any public setting there is more or less no instance where "being sorry" will help you.

Utilitarian view, sure. Outrage sells, some people are addicted to outrage. They don't want an apology, they want to see people burn. Under no circumstances ever apologise to them (or lawyers).

Wisdom and self-reflection has no value in current political climate. It requires the correct audience.

I think building a OpenAI compatible RAG-pipeline that also perhaps includes the anthropic sdk let's you have the best of both worlds. You can build it in a way that a local agent can escalate to a frontier model when needed and also substitute prompts data that you want to hide from being "public". For example. I think that is meant by harness here.

And with some smart feedback loops, you will see that for home assistant type tasks local AI can pretty quickly fulfil all requirements. For quick responses or voice analysis you might want to start with an external provider.

Otherwise, I think for complex home automation around 24-32gb vram are a realistic house number, but more experienced users will probably need less here.

People say that tools are called by the model. That isn't entirely correct. The caller calls the tools, the model just formulates the syntax for it. So the model part can be externalised without you needing to compromise you system fully (just to the degree the tools would allow anyway)

Really nice app. Can partially destroy you faith in humanity if I look at trends in my country, but nice to see how different platforms might diverge or converge on topics.

People can very practically exchange information without any government knowing the content. This was true from the beginning and is still true today.

The network infrastructure makes it possible to exchange privileged information and tunnel it through insecure channels. That is who we do banking on our phones while using the trains wifi (I would still not recommend to do that). Or generally establish VPN connection. Real data privacy is possible without a physical network.

I disagree with the parent, there is an uncensored net. Most people aren't interested in it for convenience but we should be quite happy it exists.

The person who follows the crowd will usually go no further than the crowd. Those who walk alone are likely to find themselves in places no one has ever been before.

No, it is really not. I have no horse in the race and know little about the politics in Brazil. But if anyone frames political competition as a right or left wing debate, it is very, very likely horseshit.

Fact checking systems often lack context and self-correction mechanisms, suffer from selection bias and numerous other problems that doesn't elevate them above other forms of media. Especially media that doesn't claim to be an authoritative source.

You probably could have constructed a fact checker that would have said that 75% is true, the completely other way around. For most truths or facts that is entirely possible without any contradiction if you change perspective. This basic media literacy isn't optional and it cannot be outsourced.

I think this framing is pretty much disinformation itself. Sure, Russia tries to influence elections like every other country does as well, democracies or not.

Currently we see such claims resurfacing on topics like Brexit or other contemporary issues. If you look at the hard data of visibility on the attempts of Russia, they are in a similar spot as their economy. Doing decently poorly.

If any domestic news article or politician points their fingers at Russia, it is probably to distract from something else. Sure, if you are situated in Ukraine, the story is different. The latter does have to swallow the results of the moronic foreign policy of quite a few western nations for that matter and it would be served well if a bit of sensible reflection would be put on the table.

They are impressive but I often ask myself why so much effort is spend here. While people are curious about results, a bit of waiting for the counts shouldn't matter.

The most "sensible" application would be for betting markets I thought.

Agreed. I did do image processing on skin and have to say it is just very very difficult to get 100% correct. In my opinion, even if some systems do outperform some doctors, it can only work as diagnostic support.

Anomalies are very hard to detect on certain skin tones and creating data sets from image material is quite a lot of work, even if you have access to large clinics specialised on dermatology with diagnosed and undiagnosed material.

Not seeing LLM to perform very well here and hallucination would be a major issue (not at all a solved problem, even frontier models do hallucinate/match the most probable path. Perhaps a problem that cannot be solved without arbitrary rules.)

And I guess some SVM still outperform CNN, just that CNN provide a pretty good learning environment that can improve upon ingesting more data. Although there is a risk associated with that as well.

Still, if we had well documented data material, many algorithms and AIs would perform well. The work is getting the data in a dependable state.

Not from the US, but I don't see political pushback on rocket launchers for police cars.