Three kinds of AI products work(seangoedecke.com)

99 points by emschwartz 11 months ago | 96 comments

Human Routers of Machine Words(borretti.me)

55 points by zx321 4 months ago | 26 comments

Human Bottlenecks(borretti.me)

106 points by zdw 5 months ago | 34 comments

2023: The Year of AI(journal.everypixel.com)

96 points by talboren 2 years ago | 54 comments

Three AI agents, two countries, and one uneven world wide web(royapakzad.substack.com)

35 points by effects 24 days ago | 1 comments

Substack post by researcher Roya Pakzad comparing how GPT, Claude, and Muse AI agents handle multilingual research, source access, human-in-the-loop, and observability.

1ML for non-specialists: introduction(pithlessly.github.io)

30 points by birdculture 9 months ago | 10 comments

The force-feeding of AI features on an unwilling public(honest-broker.com)

330 points by imartin2k 1 year ago | 279 comments

Ted Gioia essay on The Honest Broker arguing tech companies are forcing unwanted AI features on users, calling it tyranny rather than innovation.

Implications of AI to schools(twitter.com)

322 points by bilsbie 11 months ago | 386 comments

An X post by Andrej Karpathy sharing points from a school board talk on AI in schools, arguing AI use in homework is undetectable and grading should shift to in-class work.

I'm Tired of Talking to AI(orchidfiles.com)

1753 points by theorchid 5 months ago | 850 comments

A personal essay recounting frustration with people replying using copy-pasted AI answers, including on GitHub discussions and from an employer sharing ChatGPT screenshots.

AI outperforms law professors in Stanford Law study(law.stanford.edu)

379 points by berlianta 4 months ago | 327 comments

Stanford Law School press release on a blind study where law professors preferred AI-generated answers to student legal questions over answers written by fellow professors.

There is minimal downside to switching to open models(marble.onl)

370 points by amarble 4 months ago | 298 comments

Blog essay by Andrew Marble arguing there is little downside to switching from proprietary AI models like Claude to open models, comparing it to adopting Linux.

Coding is not solved(blog.alexewerlof.com)

458 points by firstSpeaker 28 days ago | 459 comments

Alex Ewerlöf's opinion essay arguing that AI and LLM coding tools have not solved software coding, drawing on his experience building AI systems.