I'm an arms race the only winner is the arms-dealer.
2,912 karma · joined December 17, 2017
Yes it's the same with Claude. However, OpenAI allows you to use any harness you like. Which makes sense and that's the primary reason I have their plan now rather than Googles.
I also had that weird Youtube problem. I had to go without it for several days because signing up for Ultra hijacks your YouTube account for no reason.
1) Try to integrate agy into a workflow. It can't do standard I/O like: tail -200 app.log | claude -p "Find the problem"
2) Hard iteration limits. Preventing runaways is good. Preventing me from looping on purpose is anti-user. See also number 7.
3) Not open source so I can't fix any of these problems.
4) No skills. In 2026. Yikes.
5) No persistent memory (see Claudes auto memory)
6) No sub-agents or orchestration of any type really.
7) Weird hard coded limits and constant API errors on everything (scaling problems?)
8) No /loop command
9) /btw is weird and ephemeral. No way to merge it back to the conversation.
10) Unstable in general.
11) No way to control it via API.
I could keep going on. I would suggest taking a class on Claude Code or Codex then using it for a few months. Swapping is always painful, but it's so worth it. Then if you want try to go back to agy. Don't worry, agy won't have changed much. It improves at a snails pace.
I think many/most of the players will crash and burn, and the ones that are left will divide the world.
I cancelled Ultra because they forced me into their harness like I should adapt to them, rather than the other way around.
Apparently they don't want to compete now that they're doubling the costs.
Now I can finally build that half a thing I've had my eye on.
1) Worrying about your health is mostly useless and probably won't move the needle much unless you have obvious problems.
2) The United States spends 20% of it's GDP to accomplish nothing. That 1 year of life expectancy gain is probably within the margin of error. Certainly modern healthcare can help at the individual level, but the society is so sick that any gains are washed away by the noise and we spend a truly enormous amount of money to get nowhere.
Our worry costs us a fortune, and generally just doesn't accomplish anything worthwhile.
I would agree with at least this part, if I didn't live here. I'm reminded of the MRI post from a few days back. I think the stat was that Boston had more MRI's than all of Canada combined.
How much healthier are people the United States than Canada - for all that extra spend? I'm fairly certain the number is negative.
EDIT - I googled it. The Canadian life span is four years higher than the average American and they're healthier than the average American when they get there. So, yes, the number is negative.
Are you sure that people should try? Will they live longer, or happier lives if they do?
From 1900 to 1950 the average US life expectancy went from about 47 to roughly 68. Penicillin and sanitation did most of that.
From 1950 through present day we added another 10 years. What most folks don't realize is that:
- 70% of those gains, or 7 out of the 10 extra years, came from cardiovascular health. Mostly because we got better at surviving heart attacks, and people smoked less.
- 19% came from infant mortaility gains (2 of the remaining 3 years of gain).
- 11% came from EVERYTHING ELSE that people did to become "healthier." So 1 year.
The US spends $5.3 trillion per year on healthcare, give or take. That's about $15.5k per capita every year. It's roughly 20% of the total GDP of the United States. That's 2 out of every 10 productive dollars going towards healthcare.
In exchange for 20% of every hour you work for your entire adult life you get 1 extra year at the end of your life.
In exchange for constantly worrying, exercising, staying up to date, eating right, etc you buy yourself one year more of life, and it's not even going to be a GOOD year. It's going to be one more year of being 80+ years old with everything hurting and constant medical needs.
Is it really worth 20% of the money you make every year for you entire life to live one extra, shitty, year?
I just don't think people have done the math on this one.
Suicide is at once the worst and most effective cure for cancer.
The things you mention here make it happen sooner, but no amount of healthy living lowers your risk to zero and I think that's important to note when you say things like "cancer is caused by".
Cancer is caused by math. This stuff just steepens the slope.
That said, this is also inaccurate at a technical level.LLM's are very capable of doing math and they ARE calculating internally. Most of what they do is calculation, not storage. It's just not done in a way that it's trivial to explain here.
It's described in some detail below, though it's a bit dense.
https://www.lesswrong.com/posts/E7z89FKLsHk5DkmDL/language-m...
I imagine that living in a place with that sort of history was the kind of experience one can't really convey in a few words online, but if you have written about it (or intend to) I would certainly love to read more.
Any chance it was THE Vallejo where Alan Watts lived, and the grateful dead once played? Or maybe she was just named in homage?
Either way, it's very cool! Thanks for the great read.
A jury just found that there were more than 43 million crimes (total) committed on the ad side. I think you're saying that the illegal data collection wasn't related to the advertising end, but it was. That's why they did it. Whether it was actually useful or not is irrelevant.
More importantly to me, the billionaires using it to launder their bribes to the US president probably won't ever be convicted, but that does not mean they aren't just low class crooks who can afford good lawyers.
You're right that the tools were hype, and that they didn't work. So why, despite knowing that Cambridge Analytica tools were useless, did Trumo keep paying it?
Well, Robert and Rebekah Mercer were primary backers of the Trump campaign, and they also owned a major stake in Cambridge Analytica. Giving something like $49 million dollars to Trump and Co.
What are the odds they made a profit on the deal?
You're not concerned about the impact on people's freedoms or the economy? Not concerned about the chilling effect on scientific progress?
I thought this was like the last one where a private third party company misused it.
If it was OpenAI at the wheel then they're 100% responsible for how it was used. Mea Culpa.
Do you sincerely think that the company producing tools people use to break the law should be held responsible?
Like, do you genuinely think Ford execs should be rounded up if someone does a DUI using their product?
Or do you just not like AI, because you fear it's replacing you?