rekrsiv

6 ms·

1.
by rekrsiv 5mo ago
Out of curiosity, was that before or after LLM agents?
2.
by rekrsiv 5mo ago
Oh right, I keep forgetting about those useful-but-historically-mentally-draining-to-update sources of truth. This might be extremely useful when paired with an agent.
3.
by rekrsiv 5mo ago
My theory is that most problem solvers are bad at solving problems, and most managers are bad at managing, and it doesn't matter how evolution created them: They'll make mistakes, they'll have finite time and energy, a finite
4.
by rekrsiv 5mo ago
No, AI wasn't supposed to solve all that drudgery. The hypothesized AI singularity would, but an ordinary AI agent running an LLM is just a problem solving automaton with no will of its own, just like a fleshy brain solving computer pr
5.
by rekrsiv 5mo ago
Yet another statistically misleading headline.
6.
by rekrsiv 5mo ago
Time is finite and regression testing always gets punted to the back of the line when humans are excited. This simply reveals a staggering level of humanity.
7.
by rekrsiv 6mo ago
The endgame in programming is reducing complexity before the codebase becomes impossible to reason about. This is not a solved problem, and most codebases the LLMs were trained on are either just before that phase transition or well past it
8.
by rekrsiv 6mo ago
I agree with everything you said except that the US is able to stop it. Think about it: Wouldn't Israel simply use these same tools in the US to install a puppet president they can easily manipulate? Can we even prove it hasn't al
9.
by rekrsiv 6mo ago
A bribe in a low trust society is not the same as a bribe in a high trust society. In this context, a bribe is irrelevant compared to the act of election interference by a foreign actor.
10.
by rekrsiv 6mo ago
Quick question: Could they also be manipulating this message board's voting?
11.
by rekrsiv 6mo ago
You are correct. Robustness requires a system that is working within it's tolerance margin, and stressing that inevitably leads to failure. A fault-tolerant system in this case would require a large amount of redundant humans. Unfortun
12.
by rekrsiv 6mo ago
I am alarmed at the high number of supposed engineers on this thread that are seemingly unaware of how safety-critical systems work. Literally every other piece of this system has redundancy built into it. Robustness is never optional in a
13.
by rekrsiv 6mo ago
Can a single human being reliably and robustly maintain a safety-critical system alone under any circumstances, ever? Ever?
14.
by rekrsiv 6mo ago
A terminal still offers a more composable interface than a GUI. Analog feedback is still a concern for high level pilots. You are confusing power tools with entry-level instruments.
15.
by rekrsiv 7mo ago
It is still called the Department of Defense.
16.
by rekrsiv 10mo ago
On the other hand, as long as the entire internet goes down when Cloudflare goes down, I'll be able to host everything there without ever getting flack from anyone.
17.
by rekrsiv 10mo ago
You forgot to provide a counter-argument to the author's position while you attacked them personally.
18.
by rekrsiv 10mo ago
You're ignoring the fact that going outside to protest wasn't something you could just decide not to do, just like buying groceries. BLM protests and gathering indoors for fun are not equal.
19.
by rekrsiv 10mo ago
If it quacks like a Markov chain...
20.
by rekrsiv 10mo ago
Not all artificial neurons are LLMs. Machine learning can be applied to any kind of large data set, not just human prose, and will start finding useful patterns before a human brain has time to learn how many fingers it has.
21.
by rekrsiv 10mo ago
We use logic to design our technology, but evolution does it by literally shaking all the atoms into place, no design involved. Our brains were created randomly.
22.
by rekrsiv 10mo ago
But we don't even need a human brain. We already have those, they take months to grow, take forever to train, and are forever distracted. Our logic-based processes will keep getting smaller and less power hungry as we figure out how to
23.
by rekrsiv 10mo ago
On the other hand, a large part of the complexity of human hardware randomly evolved for survival and only recently started playing around in the higher-order intellect game. It could be that we don't need so many neurons just for play
24.
by rekrsiv 1y ago
Development is often divided into 80% known unknowns and 20% unknown unknowns. AI can only help with one of those, and it's the one that takes the least amount of time to complete. Research and thinking is always going to be the bottle
25.
by rekrsiv 1y ago
Feel free to point to said data.
26.
by rekrsiv 1y ago
I believe this is true for literally anything that replaces practice. We're meant to build muscle memory for things through repetition, but if we sidestep the repetition by farming it out to another process, we never build muscle memor
27.
by rekrsiv 1y ago
This a supply and demand problem, supply will continue until Americans stop needing drugs. Maybe stop creating situations where people need the drugs in the first place?
28.
by rekrsiv 1y ago
While I agree that it's very easy to create too much code, I think one of the bigger sources of debt we face as an industry is poorly fitting our tools to the tasks at hand. Sure, a SQL database is nice if you need to query the data in
29.
by rekrsiv 1y ago
At the end of the day, rationales like this are moot when the final deciding factor will be an envelope full of cash to some politicians.
30.
by rekrsiv 1y ago
Both of those have trillions of dollars more in counter-initiatives, and that was before the global democratic backsliding started snowballing. I don't think it's going to happen this century.

More ›