weekend ai reads for 2024-11-15

November 15, 2024

📰 ABOVE THE FOLD: CREATIVITY

If Your Tattoo Was Designed By AI, Does It Have a Soul? — Ink enthusiasts are divided over whether using artificial intelligence to design body art is fair game or taboo; ‘It’s like doing sports on steroids’ / Wall Street Journal (archive link) (6 minute read)

Not even Spotify is safe from AI slop — How fake music targets real artists. / The Verge (14 minute read)

An AI’s Prediction of the Future of Art — I asked an AI to predict the near future of art and creative work. This is the story it told me. / Noema Magazine (24 minute read)

Our findings suggest that participants employed shared yet flawed heuristics to differentiate AI from human poetry: the simplicity of AI-generated poems may be easier for non-experts to understand, leading them to prefer AI-generated poetry and misinterpret the complexity of human poems as incoherence generated by AI.

The Beatles’ “Now And Then” First AI-Assisted Song With Grammy Nomination — The song, which was released last November and completed with the use of AI, has been nominated in the Record of the Year and Best Rock Performance categories. / Billboard (7 minute read)

 

📻 QUOTES OF THE WEEK

You done did so much for money that you lost yourself in it. The best thing you can do is show someone else’s soul. “Yo, listen to this James Brown record, like, he’s baring his soul in this record.”

But where’s yours?

ASAP Ferg (source)

 

👥 FOR EVERYONE

AI chatbots may ease the world’s loneliness (or make it worse) — Man seeking meaningful relationship at the intersection of on-demand empathy and Rule 34. / Big Think (17 minute read)

f.AI.th / Craig Palsson, Market Power, Substack (sorry) (10 minute read)

First, the belief that God inspires humans to innovate as a way to further His work. In this framing, the question that every new technology carries is not, “How can this be used to make profits?” in a way that creates economic growth. The key question is, “How can this be used to bring more people to Jesus Christ?”

  • useful to understand why anyone would use A.I.; for many it is an economic decision but we read this as tying the strategy to the core mission of the organization

 

📚 FOUNDATIONAL

OpenAI, Google and Anthropic Are Struggling to Build More Advanced AI — Three of the leading artificial intelligence companies are seeing diminishing returns from their costly efforts to develop newer models. / Bloomberg (11 minute read)

  • speculative

I Talk to Robots While Driving / Tomasz Tunguz (3 minute read)

The hard work of the next few years will be reimagining these workflows. Our tools have fundamentally changed, and so too must the way we work with them.

How might LLMs store facts / 3Blue1Brown, YouTube (23 minute video)

 

🚀 FOR LEADERS

Conversely, almost half (49%) of respondents said they believe moving more slowly than the market would have a negative (47%) or even catastrophic (2%) impact on their organization.

What Africa needs to do to become a major AI player — Inadequate funding, infrastructure issues, and fights over regulation mean the sector’s future remains uncertain. / MIT Technology Review (15 minute read)

Nearly half (48%) of all desk workers would be uncomfortable admitting to their manager that they used AI for common workplace tasks. The top reasons for workers’ discomfort are 1) feeling like using AI is cheating 2) fear of being seen as less competent and 3) fear of being seen as lazy

  • and

A persistent lack of training continues to hamper AI uptake; 61% of desk workers have spent less than five hours total learning how to use AI.

 

🎓 FOR EDUCATORS

How ChatGPT Brought Down an Online Education Giant / Wall Street Journal (archive link) (7 minute read)

Schultz said Chegg wanted to target students who need more than freely available answers from chatbots.

MemenomeLM — Supercharge your learning with engaging content

  • turns a PDF to a tiktok-like video; we are not the target audience but seems to work fine

  • requires registration

ChatGPT Doesn’t Have to Ruin College — The power of a robust honor code—and abundant institutional resources / The Atlantic (17 minute read)

 

📊 FOR TECHNOLOGISTS

  • covers everything from which models to use, to user interfaces, to best uses for LLMs

Magentic-One employs a multi-agent architecture where a lead agent, the Orchestrator, directs four other agents to solve tasks. The Orchestrator plans, tracks progress, and re-plans to recover from errors, while directing specialized agents to perform tasks like operating a web browser, navigating local files, or writing and executing Python code.

Gwern Branwen - How an Anonymous Researcher Predicted AI’s Trajectory — Legacy & Anonymity in the Age of AGI / Dwarkesh Podcast (96 minute video, with transcript)

 

🎉 FOR FUN

Deep in the Forest / Doopiidoo, YouTube (2 minute video)

  • AI video for a metal song of unknown provenance (or Shazam doesn’t know, anyway)

  • disco balls, shark police cars, and a smoking rooster

Create memory movies - Apple Intelligence / Apple, YouTube (1 minute video)

  • in our opinion, this is a laughably bad advertisement but maybe people actually want to live like this?

The 200 Best Inventions of 2024 / Time Magazine (15 minute read)

  • couple handfuls of AI-related inventions

 

🧿 AI-ADJACENT

 

⋄