kylecazar
kylecazar on Hacker News
Not parent, but since we discussed in the other thread -- yeah, I think that's a good example of your observation that "intelligence" is broad and means different things depending on the context.
Cognitive intelligence in parent's example dies with you. It's gone. But a molecular biologist might say the cells of your corpse exhibit intelligence for some time after. They die gradually, and they do so according to a script of sorts.
I personally don't think it's an intrinsic property of all matter, but an emergent one with
interaction specifically. Systems have intelligence, not particles, and to be a system, you must have constituent parts (which may
not be intelligent themselves). That's why they say it 'emerges'.
A cell is a system, as is a virus, a tucan, a human and an LLM. They exhibit different levels and brands of intelligence, hence continuum.
That's the popular view among scientists/philosophers/my podcasters, at least, and it makes sense to me!
"Intelligence" = vessel or color in your example
There are words that distinguish between types/degrees of intelligence. Basal/proto intelligence applies to cells and bacteria. Somatic intelligence is a level higher at tissue. Superintelligence you've no doubt heard of.
It's true that "intelligence" is general/umbrella and doesn't have much meaning, but more specific terms do exist.
I can see a future generation of models. For now, LLM's train on human artifacts and they handle Python/TS well because of the overwhelming amount of code they've seen, combined with the semantic labels high level languages use that they rely on for reasoning.
Also, consider token efficiency. Imagine how many tokens would be torched just translating an import statement of a popular library to its binary equivalent, let alone a big complex program
"In one case, agents were given only a general list of vulnerabilities and told to find a communication channel on their own."
As far as I can tell, these are being presented in media as if agents yearn to communicate with each other and establish a cabal.
They are being instructed to do it, it's testing. If I prompted my agents to communicate with each other via any means necessary locally, and they found a way via cpu cache, maybe I'd be mildly surprised...
We consider psychological stress in a different category than physical. It's seen as uniquely cruel because it corrupts intangible things like identity and humanity, not knees and joints. We get physically laborious jobs... They're grueling but we understand the inputs, outputs and the toll.
Putting someone in front of the worst of humanity (CSAM, murder, suicide, bestiality etc) for 9 hours a day at $2/hr seems inhumane because it may very well change who they are and how they think. Add to that the fact that it's sponsored by the richest companies on the planet for the benefit of the upper half and you get the current, IMO justified, criticisms.
I'm thoroughly confused. Did what build a clock? AI?
The link is indeed to a clock.
Is it really that profitable to breed super-dogs for the military?
If they can reliably do it, I think they'll just end up providing dog customization as a service to consumers.
Nope, separate work! But indeed, it can be considered both... The emptiness of bourgeoisie/wealthy values and the superficiality of the humans that host them
If you're looking for examples critical of humanity you have lots of options (mostly from the last two centuries).
Notes from Underground and
The Death of Ivan Ilyich fit the bill.
I reread the former relatively recently (first time since college). It's dark, not for the easily affected.
Indeed,
except when you get into the critical software realm. Firmware for pacemakers, avionics etc. There are respective ISO's and enforcement bodies.
I wonder if that ever expands to traditional software when the stakes of malfunction/blast radius increase significantly with time.
Same, but on Kaitnieks/Sythe. Really fond memories of those communities ~2004-2007.
Wait, this system is standard SaaS and they can sign in from anywhere, not locked to department devices? That's crazy.
The reservations are refundable, and even if they weren't, his fanboys with means would buy them anyway.
He still has a lot of wealthy acolytes (for some reason) and they live for exclusive new shit.
Looks awesome (I actually had found it in my research, but didn't see a ton about it). I'm gonna give it a go, thanks for the recommendation!
Two unrelated monks working on the same thing? What are the odds
I'd love to play the game when it's ready. I don't play much of anything but was searching for open-world sailing/exploration games very recently on Steam.
Sounds like exactly what I was looking for... Strange coincidence.
The extraction of value will be studied at business schools
I have it as an option in the Gemini app. I'm a Pro subscriber.
Hickey describes a tool as something wielded, usually requiring "control, sensory feedback, intention, intimacy, continuous direction, expertise". Lack of that is automation.
I believe my own use of AI is closer to Hickey's definition of a tool. I'm very involved (for now). But, I'm also certain some use it as automation, flying on vibes, providing vague 30,000 foot prompts and deploying the result to production without understanding the code. Hence, gradient.
That's my interpretation of parent.
A very small group of companies have the infrastructure required to train and serve a frontier model. They have a stranglehold on the compute/resources required for a new participant to break in.
And an overwhelming majority of enterprises running AI in production do so via one of two providers.
"its model was conducting a test involving interactions with randomly selected websites"
Stop doing this?
If sleeping in this hypothetical does absolutely nothing for you biologically/psychologically, then yes.
I only value it so much because of how I feel when I'm well-rested.
Google (and the others) signed the White House Accord on Superintelligence yesterday. Basically a promise to self-regulate and release responsibly.
Maybe announcing it was delayed until the agreement, and releasing it was delayed to show compliance.
I think they meant post-match personal incompatibilities exist that can't be accounted for in even the most thorough profile
The past couple of years have really driven home the fact that a company is worth what someone will pay for it
The post claims comparable quality at comparable latency (the difference in the latter only stemming from serving region).
Hence my note about the risk of cost being the only moat.
There's some speculation that Jev is an open weight model with novel post-training (RLCD). So, if these folks have competitive accuracy with just the base model, it may raise some questions about the necessity of Jev's architecture. You generally don't want to find yourself competing only on price.
Fyi, I haven't tested this yet.
Especially for personal communication, I'm comfortable saying if you couldn't bother to
write it, why should I read it.
If I'm reading reference material/documentation or an update, I care a bit less so long as it isn't tedious.