wouldbecouldbe

I don’t know, I wouldnt be suprised if he finds a way. All the tools around, he just have to make a jump in the quality. With GLM as example they should be able to het to opus level and cut the costs

His longevity protocol trial with group of people had negative overall results according to ny times. His supplements tested to not contain what it claimed

https://www.nytimes.com/2025/03/21/technology/bryan-johnson-...

I think modern domestic houses its the opposite. At least in Netherlands insulation is such a strong focus, due to climate change I think, that modern appartments have terrible ventilation

The vps performance is not comparable to a dedicated server. Same cpu’s still means twice as slow. Also hetzner alllows for much larger servers in terms of memory at still a fraction of do pricing

Bunny.net is awesome!

That's the implied suggestion from the verification that they can prove to only share it with US citizens

HN is so negative, they are genuinely making a interesting move instead of just generating BS images, they are actually trying to make something that could benefit humanity. Whether they succeed or not, who knows, but great that they are trying.

I just refund autmoatically when there is a early fraud warning, not worth the fight. We had lot of issues from Brazil, who all just falsely claimed their creditcard was stolen.

If we get our dispute rate down, might consider trying fighting it again, they are always our lowest tiers of 19 or 29 usd. So not really worth it. Just frustrating it's such a terrible process on our end. Also twice users are open to withdraw it when I contacted them, yet their bank / card doesnt make that easy or possible, so they still count to our dispute rate.

Yeah if the refunds are automated Ive had it prevent a dispute, if I had pre-emptively refunded myself it didnt prevent the dispute. I think only mastercard and visa use the early fraud warning system, I have rules that auto refund on a warning.

Chargebacks are a pain and as a seller you almost always lose them, Stripe advices to refund based on signal's they get that a disput is requested.

As a company there is just no point in fighting them. I even had emails of clients they made a mistake and didnt realise it was our payment, but even that wasn;t enough.

I think its more that AI is generally really badly implemented. It means we get a less qualitative experience, mainly on support, but also writing etc.

I just don't see it changing, people don't buy knowledge, they pay for trust and authority. It's already been for decades that you could find every workout & workout program for free online or youtube. Yet people decided to buy anyway, because they trust that that person is right, and perhaps they tried the free stuff before and it didnt work (not because it wasn't correct, but because they didn't stick to it)

Don't know if you did it on purpose or not, but this is hilarious.

Wow, yeah maybe cool off things for a while. You have some wild experiences. We have legal truffles in NL, come and visit, they are not so expensive here. But maybe trip with a guide for a while.

shame, same with mushrooms?

I think its relatively easy to make it beneficial with an adult person taken in most serene settings & with good intentions. The studies they did do seem to reflect that.

I would advice against a too high of a dose first time. The 5 grams they normally give in studies seems to be on the high side for a first time.

You fortunately don't need medical research in learning how to live your life and experience meaning.

It's great if we have therapies that help people and get a proper scientific stamp. Yet we can also discover the benefits for ourselves before that stamp is given.

When you take it, you understand, that if taken with the right approach it can lead to profound insights in changing your life and the effects described: helping with depression, addiction and accepting death are not far fetched at all. Yet it can also, if not guided or done on someone with anxiety have the opposite effect.

The more biological effects I agreed are not conclusion that can be drawn from that.

Yeah and when they give it in experiments they also give participants an eye mask, which doesn't allow you to quickly snap back to reality. I enjoy Psilocybin, but that would be scary for me.

Ive seen opus also doing it more and more spinning up multiple agents, so maybe its a claude code update?

why not just add something like: "No need to run a full build and test suite, I will manually validate"

I mean you can get that for like 300 p/m at hetzner

I asked it to analyse my architecture and find any security issues and it did it perfectly, first identified the issues & then fixed them. Not sure why my prompt managed to get through the guardrails

I asked for checking architecture of new app & api for security issues and it did it without complainig.

Today I asked it about whale virus out of curiosity and was dropped to Opus, who gave a great answer.

They are for sure not using mythos or opus do the safeguard check.

There is 2 ways to have that discussion:

1. Ultra processed food is a media hype -> totally dismiss it

2. Ultra processed food is often used without proper classification and would be more useful to have well defined sub categories

You're just playing the semantics game for the sake of being contrarian.

If you really think Oreos, Pringles, and Lunchables aren't ultra-processed and extremely unhealthy, there's no point in having a discussion.

yeah thats why I mentioned it works well IF guided by the correct expert.

I agree that you can create a set of domain specific rules, reinforcement layer validation tools, like self driving, that vastly improves the accuracy of au & llm's. Making humans less and less needed. But where LLM's comes from the magic of generic knowledge, this will be the opposite, narrowing it down.

yeah you can just start your own! meetup (and if you have fb groups) have a good chance to bring in lots of people.

I think the opposite, only the most skilled will be required.

But it depends on the skill:

- For landing pages & simple saas solutions: marketeers & founders have more skill, since they understand the user best. The real skill is not the basic coding, but understanding the market.

- For security risks/architecture: senior devs can spot things in seconds

Im not a doctor or lawyer, but im sure there are cases where AI is really good in a similar way and cases where they miss the most crucial aspects.

Yeah it can do things unguided if the tests to confirm its correctness are very solid. Thats where a lot of progress has been made and where agents are good, but this is domain specific, and a chance where startups can shine.