andy_ppp

I often think adding minority report style gestures to a laptop would be more useful than a touchscreen, maybe that’s just me…

1. If you get banned from Amazon largely this is very bad for business. They can find a reasonable excuse I’m sure!

2. Fraud is a stretch, these things are labelled even if it’s deliberately designed to be difficult to understand. The discovery on this would be fascinating though because I can guarantee you there’s records somewhere of a ticket to make things more deceptive…

3. I don’t think the fraud thing will get you the results you want (I guess this is a slightly ironic statement given what we’re talking about). According to Claude it’s more likely deception under “FTC Act §5” but you can’t directly do this anymore and apparently “Amazon reinstated binding individual arbitration with a class-action waiver in its Conditions of Use, effective 14 August 2026.” So even taking them to court got much much more difficult.

I think unfortunately until a different administration (maybe one less motivated by billionaire donors) comes in, we are stuck with this type of practice.

You will feel better and be in hospital at the end of your life for a lot less time on average so yes it’s a lot but potentially the chances of being a fit 80 year old could be worth it.

Such good advice. Another probably unpopular use of AI - I was using Claude to figure out all the music I listened to at university and what would have been playing in the bars at the time. I was loving the nostalgia and I would not have remembered 205 of it without the conversation!

Isn't it likely that blended exercise is best for longevity anyway? i.e. per week 1-2h Zone 2 and 20 minutes HIIT (nordic 4x4?) and then as much strength training as you can fit in, say 3 x 1h PPL per week, and maybe some sauna too.

This is likely around what is best for most people and probably cutting the strength training time to 1h before cutting the rest if you "don't have time".

As the algorithm tries to keep the stenography in place this will mean if you specifically ask for a different phrasing of one paragraph, other parts of the document will need to change to keep the supposedly impossible to detect AI watermark in place? Won't people also just quickly do analysis on this to figure it out and remove the watermark - it's hard to do this for individual messages but when you can create an infinite number of messaging to train on, I'd think analysis of how the signal works will be quite trivial.

Doesn’t this mean Anthropic can accuse anyone of using their AI to write for them?

Anyone who knows anything realises banning things is a) impossible and b) your enemies will use them anyway, you are just depriving your own side of the advantages.

I’ve found Grok a bit like the open models from Chinese labs, it seems good in benchmarks and falls apart in real world use. How is this new version?

I think this likely plateaus and we all just get the smartest intelligence humans need running locally…

I have it on good authority they have been doing this for years, maybe this is just an increase to inside the station or maybe just a good time to normalise this…

You make an argument for patents not existing then, certainly they should not exist for more than 5 years, you should have exploited your discovery by then and competition is fine.

I’m not sure how true this is but a Japanese friend said Gochisousama Deshita means something slightly different along the lines of “thank you for your lives” and you’re saying that to the food as well as the chef.

Why do you want this? I honestly think we’ve completely forgotten how to be present and notice the real world without the idea of augmenting it.

If any of us did that clearly we’d be in jail for several years and put on the sex offenders register but when Meta appear to be hugely responsible for increases in teen suicide or crimes like these we get nothing. If corporations want the rights of individuals they equally should shoulder our shared societal responsibilities as well. I honestly don’t get how they can just pretend they aren’t responsible somehow because it’s slightly expensive to not commit crimes or something…

If you are downvoting this please explain why, it’s important that we come to a shared understanding how these things are ever okay for corporations and not individuals…

I don’t think you can steal someone’s work if it’s based entirely off theft tbh, I see distillation as smart - are you denying it’s a huge part of why Chinese models are competitive or do you believe the narrative yourself that they are doing this with several orders of magnitude less compute and the US labs are profligate and fond of burning money rather than optimising?

The truth is probably Anthropic/OpenAI/Google are pretty efficient but less efficient than the Chinese labs, the Chinese labs probably have more compute than they say to undermine US spending and distillation is quite efficient at bridging the gap in compute.

“isn’t worthless” is still to be decided IMO. Until you can sell it it defines worthless!

Do we believe them? It seems there’s no reason with the resources OpenAI and Anthropic have they wouldn’t be incentivised to build equally efficient pipelines? I think it’s mostly the fine tuning with distillation that gives Chinese labs their big advantage. However I’m pretty sure all the main labs are poisoning the output now when they detect Chinese activity so while K3 looks frontier on tests when you use it it’s miles off Sol and Fable.

I think the article means the seahorse emoji, as we all know the unicorn emoji doesn’t exist.

I’m probably going to ask AI to break up my huge PR into a stack then with sensible names? Probably could be a skill?

“It can scarcely be denied that the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience.”

As with all human endeavours, as the tooling gets better the expectation for a launch/product/feature increases dramatically too. I probably do 3-5x the features I used to before AI, yet there's still more to do. I'm not running out of work but making incredible progress on features that save the place I work at weeks and weeks of time every month which frees people up to do more of the other things. It still takes time to build and check and test things though.

I think in theory this is true but in practice I don't see loads more good apps or products. There's a paradox here I think. I just don't see loads of quality competitors popping up I actually think it makes building something harder because the barrier to entry just gets higher somehow.

Shorter is better. This can be applied to human communication too.

How can we measure if things are worse now? Potentially asking/training an LLM on news stories from different eras might be a fairer comparison than human intelligence at this point.

No, you’re totally misunderstanding me, I don’t think technology progress is correlated with progress for humans at all. The point is only the post war period when taxes were 90%+ for the wealthy aligned with a growing middle class and a welfare state I.e. hospitals and housing for the poorest.

Which period of the last 2000 years was good for the vast majority rather than the ultra wealthy? And are these periods aligned with technology progress?

Doesn’t mosh require a server component? I don’t have the ability to control that part…

My god, the printing press was invented in 1440 when most people were living in abject poverty - it took until after WW2 for humans to demand a more shared amount of money between rich and poor. If you think technology delivers improvements in living standards rather than all working people demanding better conditions you’re living in an absolute fantasy and do not understand history at all. I honestly can’t believe you think this.

Looks great! I tend to use autossh as I have a need for these a lot with work and I have unstable internet!