Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

"Token anxiety", a slot machine by any other name

jkap.io

Post reply on HN
Print view View original

191–200 of 249 posts

Re: "Token anxiety", a slot machine by any other name

#191

Post by d1sxeyes »

crystal_revenge wrote: post #82
ctoth wrote: post #11

The gambling analogy completely falls apart on inspection. Slot machines have variable reward schedules by design — every element is optimized to maximize time on device. Social media optimizes for engagement, and compulsive behavior is the predictable output. The optimization target produces the addiction. What's Anthropic's optimization target??? Getting you the right answer as fast as possible! The variability in…

> What's Anthropic's optimization target??? Getting you the right answer as fast as possible! What makes you believe this? The current trend in all major providers seem to be: get you to spin up as many agents as possible so that you can get billed more and their number of requests goes up. > Slot machines have variable reward schedules by design LLMs by all major providers are optimized used RLHF where they are opti…

There’s a line to be trod between returning the best result immediately, and forcing multiple attempts. Google got caught red-handed reducing search quality to increase ad impressions, no reason to think the AI companies (of which Google is one) will slowly gravitate to the same.

Re: "Token anxiety", a slot machine by any other name

#192

Post by cbg0 »

HolyLampshade wrote: post #151

I know I'm running a bit late to the party here, but maybe someone can provide some color that I (on the slightly older end of the spectrum when it comes to this) don't fully understand. When people talk about leaving their agents to run overnight, what are those agents actually doing? The limited utility I've had using agent-supported software development requires a significant amount of hand holding, maybe because…

This is very dependent on what kind of work you're asking the agent to do. For software, I've had quite a bit of success providing detailed API specifications and asking an LLM to build a client library for that. You can leave it running unattended as long as it knows what it's supposed to build and it won't need a lot of correction since you're providing the routes, returned statuses and possible error messages.

Do some people just create complete SaaSlop apps with it overnight? Of course, just put together a plan (by asking the LLM to write the plan) with everything you want the app to do and let it run.

Re: "Token anxiety", a slot machine by any other name

#193

Post by dymk »

philipwhiuk wrote: post #155
tokioyoyo wrote: post #92

Earlier quoted context omitted.

> What makes you believe this? Simply, cut-throat competition. Given multiple nations are funding different AI-labs, quality of output and speed are one of the most important things.

Dating apps also have cut-throat competition and none of them are optimised for minimising the time you spend on the app.

They don’t, they’re all owned by Match group

Re: "Token anxiety", a slot machine by any other name

#194

Post by rubenflamshep »

nerdsniper wrote: post #186
jascha_eng wrote: post #159

Earlier quoted context omitted.

My impression so far is that the parallel agent story is a fabrication of "ai influencers" and the labs themselves. I might run 3-4 claude sessions because that's the only way to have "multiple chats" to e.g. ask unrelated things. Occasionally a task takes long enough to keep multiple sessions busy, but that's rather rare and if it happens its because the agent runs a long running task like the whole test suite. The…

This is exactly my experience as well. The feature interaction matrix is growing as models get better, and I tend to build "prompt library components" for each project which saves time on "you prompt very precisely but this also takes time". But so far that doesn't change the reality - I can't find any opportunities to let an agent run for more than 30 minutes at best, and parallel agents just seem to confuse each ot…

Same. The only situation when I've consistently gotten a system to run for 20+ minutes was a data-analysis with tight guardrails and explicit multi-phase operations.

Outside that I'm juggling 2-3 sessions at most with nothing staying unattended for more than 10 minutes.

Re: "Token anxiety", a slot machine by any other name

#195

Post by jascha_eng »

nerdsniper wrote: post #186
jascha_eng wrote: post #159

Earlier quoted context omitted.

My impression so far is that the parallel agent story is a fabrication of "ai influencers" and the labs themselves. I might run 3-4 claude sessions because that's the only way to have "multiple chats" to e.g. ask unrelated things. Occasionally a task takes long enough to keep multiple sessions busy, but that's rather rare and if it happens its because the agent runs a long running task like the whole test suite. The…

This is exactly my experience as well. The feature interaction matrix is growing as models get better, and I tend to build "prompt library components" for each project which saves time on "you prompt very precisely but this also takes time". But so far that doesn't change the reality - I can't find any opportunities to let an agent run for more than 30 minutes at best, and parallel agents just seem to confuse each ot…

idk I haven't really hit the point with any llm that it comes up with useful abstractions on its own unless those abstractions have been in the training data.

E.g. imagine building a google docs clone where you have different formatting options. Claude would happily build bold and italic for you but if afterwards you add headings, tables, colors, font size, etc. It would just produce a huge if/else tree instead of building a somewhat sensible text formatting abstraction.

Tbf I wouldn't actually know how to build this myself but e.g. bold and italic work together but if you add a "code block" thing that should probably not work with font color and putting a table inside that also makes no sense.

Claude might get some of these interactions intuitively correct but at some point you'll have so many NxM interactions between features that it just forgets half of them and then the experience becomes sloppy and crashes on all edge cases.

The point of good software engineering is to simplify the matrix to something that you can keep arguing about e.g. classify formatting options into categories and then you only have to argue and think about how those categories interact.

This is the kind of thing LLMs just aren't really good at if the problem space isn't in the training data already => doing anything remotely novel. And I haven't seen it improve at this either over the releases.

Maybe this kind of engineering will eventually be dead because claude can just brute force the infinitely growing if/else tree and keep it all in context but that does not seem very likely to me. So far we still have to think of these abstraction levels ourselves and then for the sub-problems I can apply agentic coding again.

Just need to make sure that Claude doesn't breach these abstractions, which it also happily does to take short cuts btw.

Re: "Token anxiety", a slot machine by any other name

#196

Post by squeefers »

rexpop wrote: post #179
squeefers wrote: post #165

shame on him for linking his ditzy wife's article where she thinks shes the first to notice lootboxes are gambling

Oh my god! Don't call women "ditzy" you cretin. It's misogynistic as hell.

i could use "airheaded" instead, "vacuous" maybe? dim, dizzy, duncey or daft if we want to stick with the D's

Re: "Token anxiety", a slot machine by any other name

#197

Post by globular-toast »

simonw wrote: post #2

I know it's popular comparing coding agents to slot machines right now, but the comparison doesn't entirely hold for me. It's more like being hooked on a slot machine which pays out 95% of the time because you know how to trick it. (I saw "no actual evidence pointing to these improvements" with a footnote and didn't even need to click that footnote to know it was the METR thing. I wish AI holdouts would find a few mo…

> It's more like being hooked on a slot machine which pays out 95% of the time because you know how to trick it.

I think you are mistaken on what the "payout" is. There's only one reason someone is working all hours and during a party and whatnot: it's to become rich and powerful. The payout is not "more code", it's a big house, fast cars, beautiful women etc. Nobody can trick it into paying out even 1% of the time, let alone 95%.

Re: "Token anxiety", a slot machine by any other name

#198

Post by co_king_5 »

forgetfreeman wrote: post #180
wordpad wrote: post #170

Earlier quoted context omitted.

You're just being pedantic and cynical. Goal of any business in principle is profit, by your terms all of them are misaligned. Matter of fact is that customers are receiving value and the value has been a good proxy for which company will grow to be successful and which will fail.

I'm being neither pedantic nor cynical. Do you need a refresher on value proposition vs actual outcomes on the last few decades of breathlessly hyped tech bubbles? Executive summary: the portions of tech industry that attract the most investment consistently produce the worst outcomes, the more cash the shittier the result. It's also worth noting that "value" is defined as anything you can manipulate someone to pay f…

[dead]

Re: "Token anxiety", a slot machine by any other name

#199

Post by estimator7292 »

It seems like some of us treat tokens as the level of fuel in the code machine. When it runs out, you simply go do something else.

What's wild to me is that there's a whole other segment of people that treat tokens as, I dunno, some kind of malicious gatekeeping to the magical program generator. Some kind of endorphin rush of extracting functional code from a naive and poorly formed idea.

To the former group, the gambling metaphor is flatly ridiculous. The AI is a tool and tokens are your allocation for tool time. To the latter, someone is trying to stifle you and strangle your creativity behind arbitrary limits.

I don't know how to feel about this other than uneasy and worried.

Re: "Token anxiety", a slot machine by any other name

#200

Post by rubenflamshep »

jonahrd wrote: post #167
HolyLampshade wrote: post #151

I know I'm running a bit late to the party here, but maybe someone can provide some color that I (on the slightly older end of the spectrum when it comes to this) don't fully understand. When people talk about leaving their agents to run overnight, what are those agents actually doing? The limited utility I've had using agent-supported software development requires a significant amount of hand holding, maybe because…

I might be able to shine a little light on this. I came from embedded, where I wasn't able to use agents very effectively for anything other than quick round trip iterative stuff. They were still really useful, but I definitely could never envision just letting an agent run unattended. But I recently switched domains into vaguely "fullstack web" using very popular frameworks. If I spend a good portion of my day going…

Interesting. What would you say is your ratio of "sit down and make the implementation" time to "multi-agent system builds the thing" time?
Post reply on HN