50 iterations of CSS / layout? Easy, not even pushing it. A freelancer will cuss you out after 3-4 rounds of re-doing everything, but an LLM is happy to keep generating.
I have been able to iterate with Claude Design in a way that I wouldn’t illiterate with a human coworker.
“Ok, that’s good. Kill options 2,3,4,5 and make entirely new variations of them, be bold and use wildly different design theories”
“Take the submit button from 1, the list from 2, the item spacing from 3, the hamburger from 4 and then make that into variation 5”
“I liked the button design before. Split this design into two and use the old button and while we’re doing this, move the buttons to the top of the page outside the scrollable area”
I’ve found Design has been great for me who can’t blank page a design to save my life.
Take an average 22-year old. Tell them they just won the lottery, and never have to worry about money ever again. Do you think they'll be interested in starting a family?
This routinely happens for professional athletes. Don't have any stats, but American professional athletes seem well known for having many children (with many people).
Yup. People want a bit of hedonism. Who doesn't really? And society paints engaging in hedonism as fine in your 20s and evil for parents. Have kids? Goodbye partying. Goodbye hobbies. Goodbye a sense of agency. It doesn't have to be this way, but this is how it is framed both internally and externally.
This absolutely would have sealed it for me. It would still seal it for me now.
Being disabled, and having AI be a risk to the only work I can perform means financial concerns are at the heart of everything. There is simply too much financial risk even without children.
There are studies on this in both Sweden and the US. Young men who win the lottery have children at higher rates than the control group. Young women who in the lottery have children at equal or lower rates than the control group.
You have to do what I call "Manhattan Project" them. You can almost always evade the controls by carefully prompting them. It just wastes effort and time you should be spending doing other things in an LLM workflow. Essentially, there is almost no single discrete piece of a reverse engineering or CTF process that you can't get Claude to do, you just have to isolate it adequately and avoid letting it use names that attenuate it towards "this is an exploit" or "this is reverse engineering". I have not found a task I could not convince Claude to do. You can also fill the context window up with badgering it and eventually it is likely to simply let you through if you are careful, most of the safe guards are not deterministic.
The circle one would have gotten me. "There's a neat algo for this. The name starts with B, and you just get an oracle that tells you if next y ==current y or y-1. And then you loop x, and you have to mirror that to do all octants of the circle with mirroring and flipping by -1 for some octants"
Writing that oracle after 20+ years would have been left as an exercise to the reader.
Fair, I wouldn't have been able to write Bresenham back then (or now, off the top of my head). I'd have written a simple trig-based one. Maybe I'd have failed the interview :D
I feel your pain. Learning the basics of IP law is one of those "curses of knowledge"[1] because you can never not sound like a pedantic asshole whenever someone who doesn't know IP law tries to talk about IP law. "No, it's not a trademark suit, it doesn't matter if they've tried to defend it before. What, why are you even bringing up patents, it's not a specific, narrow invention, and it would've expired decades ago even if it was. Ahhh!"
reply