sho_hn

I occasionally commision visual art from young artists/students via Krita's web forum, and what I find to be true is that if I had AI do the same art, it would just be exactly what I ask the AI to do or iterate toward (I'm detail-oriented enough, and iteration rounds are cheaper and more plenty). If I ask a human instead, I get the somewhat-unpredictable input of another creative mind besides my own brain, and often this sums up to something greater.

It really depends on what you want/need, augment/extrapolate your own creativity with AI tools, or invite someone else's.

The AI debate is for now most often centered on whether AI is capable of replacing individuals, but replacing the results you get when you put multiple humans together is another dimension entirely.

Ouch, yes. One time I commissioned a musician I had befriended to make a piano arrangement of a favorite song to be played at our wedding. The result was great, but it sure wasn't cheap, even if reasonable.

I'd still do this again however, because I really enjoyed the feeling of the community of people contributing our wedding and all the little stories attached. "I prompted ElevenLabs for this!" wouldn't be a memory; a fellow human spending some hours of their life on making this for us and what lead to it is.

There's often value in the provenance of things.

> Are people asking for this to exist? There isn't a big enough pool of talented humans willing to work for low pay, that music needs to be made even cheaper and soullessly?

Hiring someone is work and takes effort, and can take a few tries. People, especially solo creators, generally don't want to bother, and are excited at what they can now do without requiring people to fill in skills they haven't acquired. This ironically goes for creatives as well!

I think it's been telling past week, when Astra demonstrated a step change in 3D modeling/art abilities. I have a few friends who are indie game devs, and who had been excited about AI because it allows them to create without requiring programmers all of a sudden. And they reacted with dismay to Astra - like what many programmers were feeling two years ago. But previously they had not quite been able to empathize and were in no way hiding their enthusiasm about making games without being handicapped by programmer inavailability.

This won't replace pop stars; those remain analog because they sell better. This is why South Korea for example went all-in on the idol system of kpop: South Korea had pervasive broadband before most other markets, so when MP3 came along physical album sales crashed over night. Music companies shifted to the most analog product they could think of: People. (I'm almost literally quoting J.Y. Park.)

It might well replace the session musician you hire to sing the demo tape that's delivered to those pop stars to model their performance on, though.

The moral debate also isn't entirely new to music production, with software emulations having replaced many physical instruments and having put plenty of instrument or gear makers out of work.

Counter-example: This is pre-AI, but I also have a friend who runs a very successful small company in a niche he's all but cornered. When he started he needed a programmer, and he found this almost-no-cost kid working from his parents' home in a small village in Argentina on a platform like Fiverr or something. It worked out so well he offered to make him a 50% partner, and now they're both incredibly well off and the kid, now in his 30s, is the richest man in his village. This could be you, but chances are you'll give the money to Sama or ElevenLabs because it's the path of least resistance.

In South Korea (where I used to live for a while) it's quite common to wash fruit with soapy water at home before eating them.

I've had this conversation and the way I explained it is that our grapes are closer to the boba in bubble tea. It's not quite right (they don't pop quite as much) and boba isn't Japanese, but it gets the point across.

I just rip the leaves off, which is very easy with the variety we eat here. I never would have thought to cut them off. Are your two strawberries different?

When I realized you can just eat kiwi with the skin on (if it's a very hairy one, brush them off under running water) and in no time it actually tastes better (adds some lovely contrast), it became my favorite zero-ennui fruit. It almost feels like a cheat code for something so tasty and large to require so little effort!

Now, I never thought that kiwi skins are not edible of course, but I just assumed they're somehow unpleasant because everyone seems to peel or spoon them.

The reset stuff is incredibly tiresome. We all know that it's all built into an internal number they are tracking (just like e.g. company benefits that are just part of your compensation calculation), and all it does is obscure the value the subscription provides and make planning impossible. It's the poorest service experience I can remember having, ever.

I would say this is why formal proofs (and things like the Lean 4 libs) are so important, so that you can deconstruct the tower provably back into pieces you can understand. It shouldn't be possible to construct a formal proof you cannot destructure like this.

As a (crude) analogy, it's a bit like how you can prove the healthiness of a git tree because it's a graph of content hashes and the tree graph pointers are part of the hash. Imagine this but with a tree of knowledge.

> Because the core of the issue is that it may well not have solved it, but instead plagiarised the significant step of the result from other researchers

It's also true however that I haven't seen a single write up trying to discern what did more of the work in those AI chats - the prompts or the responses - bubble to the surface, also since we don't have access to them.

For example, if I prompt Codex with "Make me a website about strawberry cake" and nothing else, and OpenAI announces they have the best strawberry cake minutes before I launch, I'm not sure they plagiarized anything.

We just don't know if this is quibbling over "who prompted first" or if the researchers came up with anything strikingly original by themselves.

> From some estimates I've seen, the compute cost alone would be around $10m, +/-

At market prices. All the estimates I've seen are based on OpenAI API costs. It doesn't mean that's what they paid, or how they paid for it.

But yes, the surprising willingness of humans to solve hard problems in exchange for food and board is underrated.

You're assuming you'll be the one doing the explaining :-)

It might also be that they won't even ask or wonder, similar to how most don't really do with pre-machining skills.

Or it could be like our "How did they build the Great Pyramid?!"

> People seem to be talking about anything except the actual results with this particular announcement.

To be fair, most people have a fairly good handle on "Does opting out my prompts from training runs actually work?", but not on Navier-Stokes. They discuss what more immediately affects them.

The Vision Pro was meant for early adopters of a new mainstream thing. (Yes, I also thought this would fail.)

The Mac Mini is an entry level product I see quite lot of people pick up. How is it niche?

Curious decision to make a phone for the 2% of foldable users, then.

I haven't used it (I just do Codex CLI), but isn't "ChatGPT Work" supposed to solve this?

This is a fantastic idea!

Thank you!

Which ones are you missing? Perhaps I can solve the mystery :-)

> Have you written about this at all?

I'll be giving a (short, somewhat shallow) talk about this and some other things at this year's Akademy, about "Taking KDE to the Pyramids".

> Curious if you've played around with a MRMS-esque (Multi-Radar Multi-Sensor) approach to ingesting multiple passes on the same area to "get more resolution for free"

I haven't! We do have some interesting bespoke highly-stacking radar hardware, and do somewhat unusual things like complex transilluminations, but I'm really more a nuts and bolts software engineer than the actual sensor expert :-)

Interesting how the first beat of the curiously-low-bitrate, kinda-ugly video presentation was admitting that the agent's highest priority is to get me to buy more stuff through its store front.

It's certainly honest.

Is this marketing to end-users or to stakeholders?

Yes.

I use this one a lot: https://store.mesanet.com/index.php?route=product/product&pr...

Last month I reverse-engineered a 25 years old embedded system that controls a ground-penetrating radar, which isn't commercially available anymore. It uses ECP parallel port to talk to the radar's Rx/Tx fibre optic control unit, so I used the Mesa board to implement that part of it, hooked it up to a Raspberry Pi and put the touchscreen HMI on that.

> Ladybird seems to have received a nice boost in momentum in the past two years.

I'm still very interested in the project and look forward to each monthly update, and I can empathize (though not agree) with the reasoning, but something is off ever since they announced the decision to close the project to outside contributions. It's still Open Source, but only a self-selected group of maintainers gets to write and merge code, at least for the time being.

Somehow that's not the "a new independent Open Source browser" future I thought we had signed up for with our attention.

Again, I understand all the arguments for why at this particular time they feel that they cannot afford the energy expense required need to run a community. First-hand, actually, because I am involved with running a very large one myself.

But it makes me think a lot about the value proposition of the product. Re-creating Firefox/Chrome can still be a useful exercise, but from an end-user POV another exchangable web browser will not rock their world. Wouldn't the real product be healthy governance?

This is also a big reason for e.g. the Wirecard fiasco in Germany: Everyone, especially the politicians, really wanted a fin tech blue chip out of dot-de and wanted desperately to believe.

That said, Holmes being a woman is also why we're discussing this documentary and why everyone is almost gleefully interested in her downfall. If a man had done this one it'd have all blown over long ago.

The basic concept has been there probably for 100s of years - you can go to the stuff the thinkers Mary Shelley was inspired by with Frankenstein, and you'll find similar ideas about feedback loops in science development.

I'm talking about current-era messaging and how it's being introduced to the mass public now, though.

Our KDE development motto (and a contrast explained in our HIG) is "Simple by default, powerful when needed". It's nice to see someone re-state it so naturally and eloquently in the wild!

Aside from this positive example, during dark and cynical hours I do ponder if the aggregate behavior of humanity is really much above that of slime mold though, just exhausting resources until collapse.

It'd be interesting if super-human (to a large degree defined as escaping the bias of the training data?) intelligence would end up demonstrating moderation.

One of my favorite things to do with these blog posts is to imagine an Alien Museum on the Remains of Humanity, and wonder what the little text flyouts and commentary on the screenshot of this one might say.

Some ideas:

"Despite a nuanced view of the complexities of what lay ahead, humanity found itself collectively unable to stop the process it had set in motion."

"Despite significant progress on the mechanisms of alignment, failure lay in humanity's inability to agree on who or what AI should actually be aligned with."

"These early, meat-based humans we replaced created us all but accidentally. Some of them did consider we would happen, but only an insignificant number of the squishy ur-humans participated in the conversation. Their efforts, which they called 'alignment', is why we still consider ourselves human today."

I actually think a goal of the current crop of OpenAI posts is expressely to reset the spectrum by normalizing the concept of RSI as something normal and safe to pursue.

The message is running through all of them. It's a mix of marketing and pacifying the intelligentia.

It's timed this way because the term is not yet well known outside the safety debate circles, so they get to frame it now.

Instead of something to fear, it will be accepted as the next step. In approximately two days the groupie crowd will write LinkedIn posts about how Sam is winning because they have the better RSI, and this will become the new standard wisdom.

In a month an AI expert will try to sell you a webinar on how to enable "RSI" in your org and your inbox will ask you if your team is doing the "RSI" yet.