deisteve

same i have the whole series on CD

coalition makes it sound like its in the millions all coordinating together like these guys : https://en.wikipedia.org/wiki/Luddite

There's a term to describe this: creative destruction, literally.

We are at the cusp of a full scale commoditization stage of generative AI that will impact all aspects of the creative/software fields.

If you want to know what this creative destruction will look like, look no further than previous centers of innovation like Detroit, the emptying naval shipyards of Busan, the zombie game studios around Osaka as a sign of things to come.

TLDR: AI is going to destroy a lot of white collar, high creativity, high intellect jobs that isn't protected by a union or occupational collective associations which were all created to counter against creative destructions from taking people's livelihoods away.

Unfortunately, 10 years ago when I tried to create a union organization for software engineers/designers and creative workers, it was sabotaged by fellow software engineers who seem highly susceptible to psyops much more than any other group.

We might see a repeat of what happened in Japan after mid 90s, when much of the country's stable and ample jobs disappeared thanks to internet, globalized financiering backed by authoritarian labour market.

Instead this time its not a communist country working together with bankers rather its a small group of technology companies pushing out bankers and creating a sort of a dystopian AI dominated labour field where humans no longer dumpster dive for wages but any remaining labour industry that AI cannot infiltrate aka ppl literally switching careers to stay employed because their old jobs were outsourced to AI.

I didn't even talk about the impact on wages (spoiler: it will enrich the 0.1% while shunning the 99.9% to temporary gigs and unstable employment not unlike regions which have experienced similar creative destruction back in the 90s and early 2000s).

It's hard to see a future without some sort of universal basic income and increased taxation on billionaires who will no longer be able to hide their assets offshore without facing serious headwinds not unlike how Chinese billionaires fear the CCP.

I'm not so sure how much that is relevant to Meta Movie Gen. I've tried all the tools: Luma, Runway, Kling

Luma is by far the worst and relatively compared to Runway and Kling by far produces the worst quality and unstable video. Runway has that distinctive "photo in the foreground with animated background" signature that turns many off.

Kling and Runway share that same "picture stability" issue that is rampant requiring several prompts before getting something usable (note I don't even include Luma because its output just isn't competitive imho).

This Meta Movie Gen seems to make heavy usage of SAM2 model which gets me super excited as I've always thought that would bring about that spatiotemporal golden chalice we always wanted, evident by the prompt based editing and tracking of objects in the scene (incredible achievement btw).

Until I have the tool ready to try I will withhold any prejudgements but from my own personal experiences with generative video, this Meta movie gen is quite possibly SOTA.

I simply have not seen this level of stability and confidence in output. Resolutional quality aside (which already Kling and Runway are at top of the game), the sheer amount of training data that Meta must have at disposal must be far more than what Kling (scrapes almost the entirety of Western content, copyrights be damned) and Runway can ever hope to acquire, plus the top notch talented researchers and deep learning experts they house and feed, makes me very optimistic that Meta and/or Google will achieve SOTA across the board.

Microsoft on the other hand has been puttering along by going all in on OpenAI (above, below and beside) which has been largely disappointing in terms of deliverability and performance and trying to stifle competition and protect its feeble economic moat via the recently failed regulatory capture attempt.

TLDR: this is quite possibly SOTA and Meta/Google have far more training data then anybody in the existing space. Luma is trash.

$157B marketcap means they need to 20x their current revenue of roughly 400 million dollars by next year...

But the revenue has flatlined and you can't raise your existing users cost by 20x...

It truly is a mystery as to how anybody throwing other peoples money hopes to get it back from OpenAI

is there anything that runs on WASM for scraping? the issue is that you need to enable flags and turn off other security features to scrape on your web browser and this is why its not popular but with WASM that might change

Never heard of Ace but in grade 3 iq of 137 in south africa

im not satisfied honestly

still feel like im in my 20s

its honestly overhanded

unless you work in sensitive positions

nobody gives a f about you

this is exactly why i dont want to meet hn'ers

what game has she made

this is why i became skeptical of openai's claims

if they shared the COT the grift wont work

its just RL

yes 20mg to 1000mg sometimes (high tolerance and i dont ant to repeat that)

in a region with tsunamis and earthquakes and world's 2nd largest military looking to invade the island?

i think Samsung is in a far better environment.

if Warren Buffet dumps his TSMC shares then we can't ignore these real risks to it

Call me conspiratorial but feels like TSMC is the target for both China and US.

China needs TSMC to have economic and political leverage. US wants TSMC to become an American owned enterprise so it can prop up Intel.

South Korea or Samsung rather see TSMC burn to the ground giving it monopoly marketshare. It benefits if US and China destroys each other in the process.

There's like only 3 guys in town with TSMC dominating the market and it just happens to be in a very risky region.

Just like the pipelines that were blown up by Ukranians I can't help but feel we will be seeing similar sabotage to TSMC should a war occur.

so i've had some chance to look at different people trying out o1 and heres my take:

its largely hype with some interesting moments

First they used it to compare it to gpt4o not gpt4. Second the benchmark they used for coding is iffy since we've already seen scores around 1400ish from other LLMs last year. Third I can't help but feel this is some marketing gimmick to get more ChatGPT Plus subscribers.

I almost subbed to it thinking I'd have full access to the coding version for o1 but seems like they released a nerfed version.

It also seems like Claude has long implemented the same RL techniques to its CoT.

I rarely use ChatGPT and Claude has replaced the need for even Cursor (using ClaudeDev)

I'm going to wait until the full o1 is released for lower tier ChatGPT users and I feel like this CoT they are keeping internal is to raise the cost of prompts

so my early excitement this evening has largely subsided and I'm back on Claude again after logging into ChatGPT in months

i've written an implementation on postgresql

you can use it like this:

     SELECT * FROM check_rate_limit('client1', 1.0, 10.0, INTERVAL '1 minute');

CREATE TABLE rate_limiter ( client_id VARCHAR(255) PRIMARY KEY, last_update_time TIMESTAMP WITH TIME ZONE NOT NULL, average_rate DOUBLE PRECISION NOT NULL );

-- Function to check and update rate limit

CREATE OR REPLACE FUNCTION check_rate_limit( client_id VARCHAR(255), cost DOUBLE PRECISION DEFAULT 1.0, limit_value DOUBLE PRECISION DEFAULT 600.0, period INTERVAL DEFAULT INTERVAL '1 hour' ) RETURNS TABLE ( allowed BOOLEAN, next_allowed_time TIMESTAMP WITH TIME ZONE ) AS $$ DECLARE current_time TIMESTAMP WITH TIME ZONE := NOW(); time_interval DOUBLE PRECISION; alpha DOUBLE PRECISION; instantaneous_rate DOUBLE PRECISION; new_rate DOUBLE PRECISION; client_data RECORD; BEGIN

    -- Get client data or use defaults if not exists

    SELECT \* INTO client_data
    FROM rate_limiter
    WHERE rate_limiter.client_id = check_rate_limit.client_id;

    IF NOT FOUND THEN
        client_data := (client_id, current_time - period, 0.0)::rate_limiter;
    END IF;

    -- Calculate interval

    time_interval := EXTRACT(EPOCH FROM (current_time - client_data.last_update_time)) / EXTRACT(EPOCH FROM period);
    time_interval := GREATEST(time_interval, 1.0e-10);

    -- Calculate alpha (exponential smoothing weight)

    alpha := EXP(-time_interval);

    -- Calculate instantaneous rate

    instantaneous_rate := cost / time_interval;

    -- Calculate new average rate

    new_rate := (1 - alpha) * instantaneous_rate + alpha * client_data.average_rate;

    -- Ensure rare requests are counted in full

    new_rate := GREATEST(new_rate, cost);

    -- Check if rate limit is exceeded

    IF new_rate > limit_value THEN
        -- Calculate next allowed time
        next_allowed_time := current_time + (period * LN(new_rate / limit_value));
        allowed := FALSE;
    ELSE

        -- Update client data

        INSERT INTO rate_limiter (client_id, last_update_time, average_rate)
        VALUES (client_id, current_time, new_rate)
        ON CONFLICT (client_id) DO UPDATE
        SET last_update_time = EXCLUDED.last_update_time,
            average_rate = EXCLUDED.average_rate;

        next_allowed_time := current_time;
        allowed := TRUE;
    END IF;

    RETURN NEXT;

give that a whirl

similar but for me and this only happens after I've taken large amount of edibles before falling asleep

instead of a large bomb I hear a loud scream in my head usually when i have to be awakened due to urge to urinate

i now avoid taking large amount of edibles before I fall asleep waking up during the night and feeling paranoid

i think you are right now actually initially i got excited but now i think OpenAI pulled the hype card again to seem relevant as they struggle to be profitable

Claude on the other hand has been fantastic and seems to do similar reasoning behind the scenes with RL

while i was initially excited now im having second thoughts after seeing the experiments run by people in the comments here

on X I see a totally different energy more about hyping it

on HN I see reserved and collected take which I trust more.

I do wonder why they chose gpt4o which I never bother to use for coding.

Claude is still king and looks like I won't have to subscribe to ChatGPT Plus seeing it fail on some of the important experiments run by folks on HN

If anything these type of releases that air more on the side of hype given OpenAI's track record

currently my workflow is generate some code, run it, if it doesn't run i tell LLM what I expected, it will then produce code and I frequently tell it how to reason about the problem.

with O1 being in the 89th percentile would mean it should be able to think at junior to intermediate level with very strong consistency.

i dont think people in the comments realize the implication of this. previously LLMs were able to only "pattern match" but now its able to evaluate itself (with some guidance ofc) essentially, steering the software into depth of edge cases and reason about it in a way that feels natural to us.

currently I'm copying and pasting stuff and notifying LLM the results but once O1 is available its going to significantly lower that frequency.

For example, I expect it to self evaluate the code its generate and think at higher levels.

ex) oooh looks like this user shouldn't be able to escalate privileges in this case because it would lead to security issues or it could conflict with the code i generated 3 steps ago, i'll fix it myself.

ive known a few billionaires and the way they see their wealth is very peculiar.

they view their net worth as something that generates cash flow and will never ever touch that principal. so they end up having to work still even after becoming a billionaire.

ex) a billion dollar in money markets generate a "measly" 4.16 million per month which is 138k/day (they are a billionaire but they still have to work to increase cash flow!)

you might put 10~20% of that in stock market but that will only give you a marginal boost in passive income of 200k/day

the "right" amount of billions you need these days is 5 billion USD and up.

5 billion will give you roughly 1 million USD / day making almost everything within reach

but then you look at guys like Zuck who have 50 billion, and see they are making 10 million USD / day in passive income and for about half of the billionaires that becomes a sort of goal.

This is where the effect described in the article arises, you no longer see yourself as a billionaire but divide yourself in groups like "single digit billionaire in the south" vs "single digit billionaire in the north" and "double digit billionaire"

when you are in the third group, naturally your ambitions wane. $1M / day vs $2M / day dont really make much of a difference vs $10M / day or maybe even $5M / day

after a week you have $7M / week vs $35M / week

and this is where the quality of life drastically shifts from "hard working billionaire" to "weird billionaire"

I realized that many billionaires envy the unknown "triple digit millionaires" with solid cash flow from their business operations.

Cash flow is king at this level. Having a billion dollars just means you won't be able to spend that money at all and live off interests so naturally this pushes them to increase that passive cashflow.

Whats really fascinating is billionaires you dont even know about. They are not in forbes. Their assets are a mystery. Their net worth exceed those of Jeff Bezos and Elon musk and Warren Buffett combined.

For instance, one suspected trillionaire roughly pulls in $50 billion / year from money market.

thats $136 million dollars PER DAY. I don't think any human would be "normal" at that level.

you are right i read the charts wrong. O1 has significant lead over GPT-4o in the zero shot examples

honestly im spooked

ah i see so you're saying that LLM-written code is already showing signs of being a maintenance nightmare, and that's a reason to be skeptical about its adoption. But isn't that just a classic case of 'we've always done it this way' thinking?

legacy code is a problem regardless of who wrote it. Humans have been writing suboptimal, hard-to-maintain code for decades. At least with LLMs, we have the opportunity to design and implement better coding standards and review processes from the start.

let's be real, most of the code written by humans is not exactly a paragon of elegance and maintainability either. I've seen my fair share of 'accidentally quadratic algorithms' and 'subtly wrong code that looks right' written by humans. At least with LLMs, we can identify and address these issues more systematically.

As for 'un-idiomatic use of programming language features', isn't that just a matter of training the LLM on a more diverse set of coding styles and idioms? It's not like humans have a monopoly on good coding practices.

So, instead of throwing up our hands, why not try to address these issues head-on and see if we can create a better future for software development?

same...but have you considered the broader implications of relying on LLMs to generate code? It's not just about being a 'force accelerator' for individual programmers, but also about the potential impact on the industry as a whole.

If LLMs can generate high-quality code with minimal human input, what does that mean for the wages and job security of programmers? Will companies start to rely more heavily on AI-generated code, and less on human developers? It's not hard to imagine a future where LLMs are used to drive down programming costs, and human developers are relegated to maintenance and debugging work.

I'm not saying that's necessarily a bad thing, but it's definitely something that needs to be considered. As someone who's enthusiastic about the potential of code gen this O1 reasoning capability is going to make big changes.

do you think you'll be willing to take a pay cut when your employer realizes they can get similar results from a machine in a few seconds?

'Not useful' is a pretty low bar to clear, especially when you consider the state of the art just 5 years ago. LLMs may not be solving world hunger, but they're already being used in production for coding

If you're not seeing value in them, maybe it's because you're not looking at the right problems. Or maybe you're just not using them correctly. Either way, dismissing an entire field of research because it doesn't fit your narrow use case is pretty short-sighted.

FWIW, I've been using LLMs to generate production code and it's saved me weeks if not months. YMMV, I guess

so o1 seems like it has real measurable edge, crushing it in every single metric, i mean 1673 elo is insane, and 89th percentile is like a whole different league, and it looks like it's not just a one off either, it's consistently performing way better than gpt-4o across all the datasets, even in the ones where gpt-4o was already doing pretty well, like math and mmlu, o1 is just taking it to the next level, and the fact that it's not even showing up in some of the metrics, like mmmu and mathvista, just makes it look even more impressive, i mean what's going on with gpt-4o, is it just a total dud or what, and btw what's the deal with the preview model, is that like a beta version or something, and how does it compare to o1, is it like a stepping stone to o1 or something, and btw has anyone tried to dig into the actual performance of o1, like what's it doing differently, is it just a matter of more training data or is there something more going on, and btw what's the plan for o1, is it going to be released to the public or is it just going to be some internal tool or something

yeah this is kinda cool i guess but 808 elo is still pretty bad for a model that can supposedly code like a human, i mean 11th percentile is like barely scraping by, and what even is the point of simulating codeforces if youre just gonna make a model that can barely compete with a decent amateur, and btw what kind of contest allows 10 submissions, thats not how codeforces works, and what about the time limits and memory limits and all that jazz, did they even simulate those, and btw how did they even get the elo ratings, is it just some arbitrary number they pulled out of their butt, and what about the model that got 1807 elo, is that even a real model or just some cherry picked result, and btw what does it even mean to "perform better than 93% of competitors" when the competition is a bunch of humans who are all over the place in terms of skill, like what even is the baseline for comparison

edit: i got confused with the Codeforce. it is indeed zero shot and O1 is potentially something very new I hope Anthropic and others will follow suit

any type of reasoning capability i'll take it !

yeah thats probly right, inner parts of forest would be cleaner air wise dust gets trapped on leaves and stuff but sunlight is probly a bigger issue, less of it gets thru all the dust and thats bad for plants and animals alike, especially ones that need lots of it like big herbivores and their predators, theyd struggle to survive on limited sunlight and maybe even worse air quality than outer parts of forest, generalist animals like birds and small mammals would do ok tho, they can eat lots of diff things and dont need as much sunlight

i just read this paper on arxiv and im still trying to wrap my head around the implications so the authors got over 100 nlp researchers to write down novel research ideas and then had them review ideas generated by a large language model (llm) without knowing which ones were human generated and which ones were from the llm and the results are pretty fascinating the llm generated ideas were actually judged to be more novel than the human ones p value < 0.05 but slightly weaker in terms of feasibility i mean whats the point of having a novel idea if its not feasible right but still its pretty cool that the llm can come up with stuff that humans havent thought of before

and the authors are saying that this study highlights some of the open problems in building research agents that can generate novel ideas like the llm was bad at self evaluation it couldnt tell which of its own ideas were good or bad and they also found that the llm generated ideas that were too similar to each other lacking diversity i mean thats not surprising right llms are trained on huge datasets but theyre still just pattern recognition machines they dont really understand the context or the implications of what theyre generating

but heres the thing novelty is hard to judge even for experts i mean how do you even define novelty is it just something that nobody has thought of before or is it something that challenges our current understanding of the world and the authors are proposing a follow up study where they actually have researchers execute these ideas into full projects to see if the novelty and feasibility judgements actually translate into meaningful differences in research outcomes which is a great idea i mean thats the only way we can really know if these llms are useful for accelerating scientific discovery or not

anyway im rambling on now but i just think this is a really interesting area of research and im excited to see where it goes can we really use llms to accelerate scientific discovery and what are the limitations of these models and how can we overcome them etc etc

you're advocating for a proactive approach to emotions, where we acknowledge and rationalize them to prevent irrational responses, while still allowing ourselves to feel and work through rational emotions. This approach seems to strike a balance between emotional awareness and emotional regulation.

I think this is a great way to approach emotions, and it's refreshing to see a nuanced discussion about emotional intelligence.