grey-area

7 ms·

91.
by grey-area 2mo ago
Certainly this is the impression true believers have had for at least a year. Nobody seems to have been left in the dust yet though. How odd.
92.
by grey-area 2mo ago
I think you mean provenance, not providence
93.
by grey-area 2mo ago
No I’ve never seen software that broke because someone added a config key.
94.
by grey-area 2mo ago
Never seen this problem in the wild. Seems like it would be a very rare issue limited to very large configs with conflicting names.
95.
by grey-area 2mo ago
Adding a config setting should never be dangerous (if it is your system is deeply broken) and should be distinct from changing an existing config setting.
96.
by grey-area 2mo ago
Users for AI should not be able to access others’ data, or become root.
97.
by grey-area 2mo ago
Wouldn’t it be safer to set up another user entirely for AI agents?
98.
by grey-area 2mo ago
That would be a much more depressing statistic.
99.
by grey-area 2mo ago
In all seriousness, if you have the time the money and the inclination, you should give yourself a challenge like this, and try to complete it. It would be an interesting exploration of the limits of current agents. You don't need some
100.
by grey-area 2mo ago
Go ahead and do it. Would love to see an improved Linux rewritten with agents. My personal belief is it would take a long time, lots of money and much human intervention/guidance but if you think they’re capable alone just do it this a
101.
by grey-area 2mo ago
Yes he could have, if agents were capable of it. Why didn’t he? Why didn’t you?
102.
by grey-area 2mo ago
That is not Modern art but an offshoot of it, like postmodern and pop art.
103.
by grey-area 2mo ago
No, Modern or Modernist art (with a capital m) is a specific period of art history. So yes it doesn’t tend to have museums but there are shows dedicated to Modern art. Contemporary art always refers to art from the specific period the speak
104.
by grey-area 2mo ago
I’m pleased for you. In my experience the use of agents and harnesses and other scaffolding around them doesn’t improve the performance of LLMs much, which is adequate for some tasks under supervision but nothing like general intelligence (
105.
by grey-area 2mo ago
I’ve made my own arcade games with an LLM. It was fun though the code wasn’t great. Mostly I felt they reproduced games from the training data, some with more examples available worked better than others. I use them most days for work, and
106.
by grey-area 2mo ago
That’s not what it means.
107.
by grey-area 2mo ago
Guided by humans, code generators which have ingested the worlds’ code and can recognise and generate patterns can be useful tools. I wouldn’t personally qualify it as a wild success as we are early and there are significant downsides. That
108.
by grey-area 2mo ago
Altman is a salesman selling flimflam to people who should know better. His idea of metering is predicated on the thing he’s selling being AGI, it is not, and all his predictions have turned to dust. Also that isn’t how metaphors work - the
109.
by grey-area 2mo ago
I’m afraid it’s a terrible metaphor, starting with the fact that LLMs are nothing like electricity, and the relation of harnesses to them is nothing like that of electronics to electricity, save perhaps one is a prerequisite of the other. H
110.
by grey-area 2mo ago
Sadly, many people have bought into the cult that LLMs will lead to AGI. I guess if that is your worldview then all this babbling about new frontiers makes more sense. They probably used an LLM to come up with this bizarre metaphor.
111.
by grey-area 2mo ago
It may or may not do that and it may introduce other problems as it does so. Checking the work is not a trivial problem.
112.
by grey-area 2mo ago
Really? Dichotomy is not a particularly complicated word, nor is it ‘performative’ for most readers.
113.
by grey-area 2mo ago
LLMs are not tutors and do not deeply know anything.
114.
by grey-area 2mo ago
They would 1 times out of 10 if they were an LLM. Then say sorry, you’re absolutely right I shouldn’t have jumped off the cliff.
115.
by grey-area 2mo ago
LLMs are not a shortcut to mastery, in fact the opposite most of the time and their errors are qualitatively different from human errors.
116.
by grey-area 2mo ago
Yes sorry for the shorthand, an internal model is a good way to describe mastery. It’s something LLMs don’t have and can’t transmit.
117.
by grey-area 2mo ago
It’s actually a funny oxymoron if you unpack it - agent implies autonomy and intelligence, harnesses are for unintelligent beasts of burden to control them. So you’re right, it makes absolutely no sense and would be hard to decipher, but do
118.
by grey-area 2mo ago
Many people think they are smart and are taken in by LLMs. Also many genuinely smart people are unable to judge their competence in other areas (eg Nobel Laureates making pronouncements in fields they know nothing about).
119.
by grey-area 2mo ago
People who think they are smart often think they can get by in any domain without knowledge and context, but at a certain point this fails and knowledge is required, not just intelligence.
120.
by grey-area 2mo ago
You appear to define words you don’t immediately understand at this point in time as ‘performative’, ie used only to show off. An odd position to take. If you don’t know a word just look it up, you may well learn something.

More ›