Yup. It's not trivial now to setup a system where you can get a frontier model to help do the research, draft a spec, humans read and comment on the spec, and then you get an open model to do the grunt work.
If the spec is very detailed, you've solved most of the problems you might encounter with open models.
You can then get a frontier model to then do a review against the spec.
Doing this cuts down frontier usage by a lot, as all the real work is local, tool calls are instant . It just feels nicer.
I think this is why you're seeing frontier models like Claude suddenly ban people using opencode/pi etc with a subscription (API users still good).
Thats the thing, i noticed it almost instantly when trying to install a package that depended on it, as soon as it started, it hard locked my laptop, didn't get to infect it.. but if they had slowed down that fork bomb.. it would have done more damage.
Yeah, and this is a pattern I saw in the Fancy Bear Goes Fishing book, a lot of discovery of malware is either pure luck, or blunders from the malware developers. https://en.wikipedia.org/wiki/Fancy_Bear_Goes_Phishing
I just installed Harbor, and it instantly pegged my cpu.. i was lucky to see my processes before the system hard locked.
Basically it forkbombed `grep -r rpcuser\rpcpassword` processes trying to find cryptowallets or something. I saw that they spawned from harness, and killed it.
Got lucky, no backdoor installed here from what i could make out of the binary
Same experience with browser-use, it installs litellm as a dependency. Rebooted mac as nothing was responding; luckily only github and huggingface tokens were saved in .git-credentials and have invalidated them. This was inside a conda env, should I reinstall my os for any potential backdoors?
After having gone all-in on LLM agents for a while, I'm not so sure anymore. An LLM with lots of context can sometimes generate more accurate code, but it can also hide decision-making from you, the person who actually has to maintain that code. If the LLM pulls in 1000 files to make a decision, that's no longer a decision that you can understand.
Personally kitty is the only one I keep coming back too. Mostly because it's very customisable, fast, lean, ligatures, separate font for italics, great macro support, and supports automatic tiling panes.
No you're not wrong, if you're comparing ARM CPUs on Linux to one specific Intel CPU, the Lunar Lake V ones. Then yeah you're not wrong, it's very much a case of optimisation for CPUs like the Snapdragon X Elite CPUs in comparison.
But if you widened the scope a bit more, then I think there's plenty of more energy hungry x86_64 CPUs compared to ARM.
Watching all these open source, federated versions of social media platforms is like if you found out your favourite drug was actually manufactured by some really bad people and made people around the world suffer. So you make an open source version of the drug. Similar formula, just this time the people can own it.
Sure you cut out the bad people, but is the situation improved now?
There are projects where long term addicts are given pure medical heroin (Diamorphin) in a controlled environment, and they do considerably better than their control group who does not receive their drugs like that.
Their drug is rage, thirst and cute trap videos generated by AI and selected for maximum engagement.
If it's not harmful it's not the same drug. Unlike diamorphine, a medical grade supply does not reduce harm; it's more like sniffing glue, inhaling poison to escape reality.
The harm from social media is at least in part caused by the feed suggestion algorithm being optimized for screen time (aka addiction). Potentially open social media where the suggestion algorithm is not that could be a big improvement.
If the spec is very detailed, you've solved most of the problems you might encounter with open models.
You can then get a frontier model to then do a review against the spec.
Doing this cuts down frontier usage by a lot, as all the real work is local, tool calls are instant . It just feels nicer.
I think this is why you're seeing frontier models like Claude suddenly ban people using opencode/pi etc with a subscription (API users still good).
reply