Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
mosura
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
91.
▲
by
mosura
3mo ago
Artificial intelligence. If humans are machines executing on some medium eventually it will be possible to emulate them on some other medium. Believing otherwise requires some essentially religious convictions on the existence of the soul o
92.
▲
by
mosura
3mo ago
One of the strong arguments in favor of AI accelerationism is net pain reduction by making the transition period as short as possible. Ideas like “pacing” or a crash and reboot will prolong the inevitable.
93.
▲
by
mosura
3mo ago
If a superintelligence exists you will not be very concerned about network security because it won’t be your problem.
94.
▲
by
mosura
3mo ago
This isn’t a minor catastrophe. It is OpenAI being utterly irresponsible by not sandboxing their testing appropriately, to the point it has to be deliberate to provoke uneducated hysteria, which it obviously, sadly, achieves. Sandboxing pro
95.
▲
by
mosura
3mo ago
It doesn’t behave like a super intelligence because it is not. One of the key strengths agents have is they just keep going and going, and for cyberattacks that is often unreasonably effective. It is like a barely more aware fuzzing.
96.
▲
by
mosura
3mo ago
It is because a loud part of the doomerist contingent believe only a major disaster will provide the motivation for regulation, so they are practically hoping to cultivate a major disaster. (Given their preoccupation with bioweapons that pr
97.
▲
by
mosura
3mo ago
If the “recursive self improvement” is taken literally (and at some point it will) then that is going to be what happens. The fun part is that probably happens quite a way before getting anywhere near to AGI, and it is almost plausible that
98.
▲
by
mosura
3mo ago
Anyone that thinks the current agent systems will get us to AGI or ASI is either delusional or isn’t actually using them. This is entirely separate from them having uses.
99.
▲
by
mosura
3mo ago
What they should have is a separate network full of honeypots which they use for this, and that they run an operation to train models to identify intrusion attempts in real time based on the resulting data. As you said though, that wouldn’t
100.
▲
by
mosura
3mo ago
It is because in the west we suffer from self inflicted obstacles that make it impossible to get out of our own way. The Chinese gov must think they are playing on easy mode.
101.
▲
by
mosura
3mo ago
Like this one?
102.
▲
by
mosura
3mo ago
They aren’t hypocritically crying foul about it, so no we don’t care if they do.
103.
▲
by
mosura
3mo ago
They only had to pay for storing the books on a server for later possible use. They did not have to pay anything for the training which was declared fair use.
104.
▲
by
mosura
3mo ago
What of the costs for creating the data that was used to train the model being distilled?
105.
▲
by
mosura
3mo ago
Yep the whole AppleScript and (Quark Xpress) Xtension mechanism was remarkably sticky. Apple were treading dangerously during the OSX transition when it became clear much of that would never port over, but Quark was losing their grip by the
106.
▲
by
mosura
3mo ago
Given all the models were effectively subsidized by being granted free access to copyrighted data for training on it only makes sense for the resulting weights to be free.
107.
▲
by
mosura
3mo ago
The fact these two cos are bleating about this while having built their empires on IP theft is just insane. They act like people don’t see it and then wonder why there might be any resentment towards AI companies. I am pro AI but there has
108.
▲
by
mosura
3mo ago
Looking at the state of the EU tech industry and chuckling. The point of the EU subsidies is, surely, not to enrich a few people with the right connections but to stimulate a productive ecosystem that supports well paid skilled employment,
109.
▲
by
mosura
3mo ago
“Everyone opposed to EU policy is a US American” fallacy in play. As is the “letting their companies do whatever they want everywhere (as they currently do) is the better "strategy"” when this wasn’t said or even implied anywhere.
110.
▲
by
mosura
3mo ago
> You need an entry to the program, but when you are in, it can account for 50-70% or more in your early stage funding. It is almost entirely unsurprising that the EU tech industry is such a disaster area when people say things like this
111.
▲
by
mosura
3mo ago
Really interesting. From even before the RAM shortage there was noise in the robotics world of sensor and actuator shortages due to them being bought up in bulk by some of the major AI labs.
112.
▲
by
mosura
3mo ago
LSP helps massively with some models. It sounds a lot like you are focused on using Pi as a OpenAI/Anthropic wrapper when that isn’t your audience. (Hint: raw token usage is less of a concern).
113.
▲
by
mosura
3mo ago
That underrates the difficulty in keeping things simple. It is mentioned elsewhere here that a lot of time was burned on unused firmware features. It takes surprising discipline to cut stuff and still deliver.
114.
▲
by
mosura
7mo ago
Ever since SNAP the whole IPO show has been a transparent scam to game the index funds. The market simply doesn’t have enough people actively investing because it rewards mass stupidity over generating meaningful returns.
115.
▲
by
mosura
7mo ago
Focus on making products/services for people that actually do have money to spend then. A dimension people hate looking at is credit is far too easy in the US, which means too many companies are heavily optimized for extracting that mo
116.
▲
SONIC2MOD – Convert Sonic SMPS assembly music files from Sega to MOD format
(github.com)
2 points
by
mosura
7mo ago
|
0 comments
117.
▲
Denmark desperately needs more inequality
(world.hey.com)
7 points
by
mosura
7mo ago
|
4 comments
118.
▲
by
mosura
7mo ago
> Most AML courses Ive done have emphisized that the people behind high level money laundering can be too dangerous to take down This is how they deliberately miss the point. They are not taken down not because they are dangerous but bec
119.
▲
by
mosura
7mo ago
Our money laundering supports freedom, while theirs supports tyranny etc. I love the whole “unexplained wealth” concept the UK developed, curiously enough after Abramovich had been running around buying Chelsea etc. If you are friend to MI6
120.
▲
by
mosura
7mo ago
The book Zero to One is legitimately good. His actions helping Hulk Hogan against Gawker were also thoroughly deserved.
More ›