I have some friends adjacent to gov who are actually looking into a token tax on the premise that we tax income / labor. I think it's a great idea. That said, I think if you fast forward 20 years, it is fairly likely that cost per token is going to be on the floor
> It is not better than me, and will never be better than me, at my profession.
from the article:
> The only remaining fundamental disagreement between people who think AI is a normal technology and those that don’t is the belief that the AI can do everything a human can do, but better.
Yes, I read that part, thanks. The article then goes on to characterize the first group as desperate, in denial, and reliant on the argument that "AI can't love or laugh", essentially asserting, baselessly, that it is a fact that LLMs will always or soon always outperform humans.
If A -> B and B -> C are within human knowledge, then I don't really consider that A -> C to be outside of human knowledge.
Sure, LLMs are very good at discovering hidden links. But that is about it. And don't get me wrong, it is a very powerful thing. But it is not god reveling the secrets of the universe...
First, what an incredible article. Just an extremely concise and clear explanation of all of the problems with AI right now.
Second, wow, the list of signatories is like a whos-who of mathematicians.
Third, I love the clearly intentional use of ‘alignment/misalignment’ language, applied to targeting the entire industry instead of AI in particular. I’ve said in the past that optimizers are substrate agnostic. Companies and governments can be misaligned, just in the same way AI can.
Fourth, I'm not sure that we can stop the optimization machines. Not the LLMs, I mean the incentives that lead to companies implementing dark patterns, lying about addiction, securing effective monopolies through downright shady behavior, and generally trying to jailbreak the system instead of improve it
i get that there is a vocal contingent yelling about kids using AI tools, and frankly I also think things like character.ai are potentially really harmful, but this is like saying you have to be age verified to use google
do you mean like a cloud agent provider? What we're working on at noriagentic.com may be relevant -- you can fire events from slack/web/cli to kick off an agent in a box and talk to it as if it was running local
First, a lot of people in thread are saying you don't need skills. This is pretty wrong. There is a lot of alpha in using any set of skills that implements SPACE (search, plan, assert, code, evaluate). See: https://open.substack.com/pub/theahura/p/agentics-using-meta...
Second, we share all of our sets of skills in a purpose built registry: https://noriskillsets.dev/ you can use any of our public skillsets from there. If you're on a team you can also sign up to get your own private registry. Makes organization much easier.
Finally, for local development, we use this CLI to manage skills (https://github.com/tilework-tech/nori-skillsets). This is a tool that lets you bundle skills into groups, and then switch between those groups. So for eg if I'm making a slide deck I'll use an admin skillset, and for coding I'll use a swe skillset, and for debugging I'll use a debugging skillset.
We do keep tinkering with our skillsets, but not very much. I don't get the need to adjust things for every model release, doesn't seem necessary for us in practice
I'm building something similar with `Agents`[0] (think of it as a package manager for AGENTS.md snippets). I'm getting good results by including detailed instructions for certain tasks in the repo, and providing hints in AGENTS.md about when to use them. It's basically a simpler version of Skills—but it feels like AGENTS.md adherence is higher than Skill adherence.
for the most part skills are so portable that fiddling with moving them around is not that hard. The really hard thing is the 'is this skill worth the bits that its written with'. I'd say 99.99% of skills out there are so bad, you're actually worse off for installing them
standards are good though, we should make plugins for our skillsets so theyre easier for some folks
Definitely. Especially since those standards are fairly aligned across lots of tools. The other cool thing is a lot of desktop clients like Claude Cowork and ChatGPT Desktop also speak plugins, so hooking up marketplaces help with further distribution outside of technical teams.
FWIW I work on AI enablement in my day job for both our build org and the broader company and we’ve found success with using marketplaces for distribution.
I also built a free tool that diffs a lot of agent configs to help me keep current on changes.
Here’s the main page, but you can checkout the marketplace and plugin entries if you’re interested!
the original version of our high autonomy skillset was a fork of obra's superpowers back in...October? November? of last year. Whenever it first came out.
I think basically all of the best skillsets implement SPACE, including superpowers, mattpocock skills, get shit done, etc. There are aesthetic reasons why you might prefer one over the other, but candidly I think any of them will get the job done and are strictly better than not using one of the SPACE skillsets.
(I haven't played with the latest version of superpowers. My hunch is that on average our skillsets have better compliance rates with the models because of some of the ways in which we write the skills themselves, but its a hunch)
i'm sorry to say that besides like the first paragraph, this article is mostly ai slop. i thought that we mostly weren't allowing ai written content on hn? (@dang whats the policy here?)
aside from that, i think that this is overcomplicating a more basic idea. You should have agents that leave docs behind in your codebase, and those docs should be localized to the structure of your code, but you do not need any fancy format or manifest.
We've had a lot of success with having the agents write and maintain a docs.md file in each folder, and requiring that they both read and update that file whenever they make a change in that folder
reply