milchek

Very unexpected but also really uplifting to see that they would spinoff a division to tackle this - it's ambitious. Obviously they've identified that the vertical is big enough and that they have the expertise or novel approach to tackle it, but i'm really curious to know how this came about internally.

I’ve tried in a 36GB MacBook Pro and haven’t had much success beyond very basic work. Issue for me was the context runs out quick even with smaller models and it’s slower. To get some half decent performance I’d imagine you want 128gb memory and are spending a lot more on hardware. At that point it becomes a question on whether you’d rather have frontier models at a subscription or sink that money into your own equipment. Of course, for those with privacy in mind your only option is forking out the cash for the higher end machines.

I see CPAP rightly getting mentioned a lot here. What worked for me was a MJS (Mandible Jaw Splint), it’s basically a mouthguard you sleep with that moves your lower jaw forward and thus brings your tongue forward so it doesn’t obstruct your airways.

Best bet is to get this custom made. Here in Australia there are specialist dentists that do this for you. It’s far easier than CPAP and works just as well for many like myself.

Often folks will use CPAP at home and these mouth guards when travelling.

I tested briefly with a MacBook Pro m4 with 36gb. Run in LM Studio with open code as the frontend and it failed over and over on tool calls. Switched back to qwen. Anyone else on similar setup have better luck?

If you liked it and want something in a similar style (conceptually) but with a bit more personal/human element I would recommend The Raw Shark Texts.

As for best sci fi books I’ve read in a while: The Gone World by Tom Sweterlitsch (think sci fi meets True Detective) - some awesome concepts here and really strong pacing. The other one was Version Control by Dexter Palmer which is a bit slower and more subtle sci fi (big concepts presented in a personal drama style of book)

I wonder if that is just a correction of the rampant hiring that took place just before this employment “crash?” - if it is as you say that its intermediates and non high performers then does that make it a good thing as well.

Truth is, when I was part of larger orgs/enterprise I definitely saw some folks who were dead weight, and I don’t mean to be harsh, a few of these knew they weren’t contributing and were being malicious in that sense.

Similarly, I wonder how many high performers now are taking multiple jobs thanks to remote work and exposing the mid to low performers. Like some kind of developer hypergamy taking place.

It comes down back to that whole discussion around intelligence becoming cheaper and more accessible but motivation and agency remaining stable.

I’ve worked with a few folks who have been given AI tools (like a designer who never coded in his life, a or video/content creator) who have absolutely taken off with creating web apps and various little tools and process improvements for themselves thanks by just vibecoding what they wanted. The key with both these individuals is high agency, curiosity, and motivation. That was innate, the AI tooling just gave them the external means to realise what they wanted to do with more ease.

These kinds of folks are not the majority, and we’re still early into this technological revolution imo (models are improving on a regular basis).

In summary, we’ve given the masses to “intelligence” but creativity and motivation stay the same.

I think for a lot of folks it basically comes down to just using AI to make the tasks they have to do easier and to free up time for themselves.

I’d argue the majority use AI this way. The minority “10x” workers who are using it to churn through more tasks are the motivated ones driving real business value being added - but let’s be honest, in a soulless enterprise 9-5 these folks are few and far between.

I would argue even with kids it is great (but of course can be expensive!). I lived and worked there for a couple of years during COVID with a young family and loved it (once the lockdowns and pandemic stuff blew over of course).

As you mentioned, for families, it’s extremely safe, everything is well run and maintained so healthcare and education are not a concern. Proximity to other countries for travel is excellent (well, I’m from Melbourne so much easier to get places than from here!), and the country it self has plenty to do for families in terms of activities, shopping, and food.

Beyond that, I found Singaporeans just really great to work with and be around. It’s really multicultural, they value education and talent so the workforce is full of bright and capable people, and there is a huge expat community as well.

The only major downside for me - the heat and humidity! It was a struggle the first few months for sure.

I used to believe the same thing but now I’m not so sure. What if we simply cannot fathom the true nature of the universe because we are so minuscule in size and temporal relevance?

What if the universe and our place in it are interconnected in some way we cannot perceive to the degree that outside the physical and temporal space we inhabit there are complex rules and codes that govern everything?

What if space and matter are just the universe expressing itself and it’s universal state and that state has far higher intelligence than we can understand?

I’m not so sure any more it’s all just random matter in a vacuum. I’m starting to think 3d space and time are a just a thin slice of something greater.

Once ads start to make their way up to the Plus paid tiers (and they will), I’ll probably switch to something else like local LLM on my home machine or put something together myself to use a non adware LLM via API (for example with Replicate). Especially if these are just intended to be spammy blocks at bottom or in between discussion threads, or worse, audio conversations.

From what I’ve read, this will be about ads in chat as suggestions? So “active” ads on response?

Why not go the approach of passive background “agentic” ad suggestions like, “hey, we know X, Y, and Z about you - would you like us to monitor certain brands related to your interests for deals and allow advertisers to pitch these deals to you?” And make these hyper specific so you can opt in.

I, like many people who dabble with music as a hobby, have GAS (gear acquisition syndrome) - why not let me toggle something like “ok, I don’t want ads, but if any of your partnered brands have a sale or good deal on X, feel free to email me, and use your ChatGPT smarts to pitch me on why it’s a good deal and how it suits my current gear set up”

I used ChatGPT to set up my guitar pedal board so surely this isn’t a huge leap.

Looking at many of these now they definitely appear dated. At the time, of course, these would have been “cutting edge”

I think one clear thing we can see is a trend toward more homogenized UI on web in the last 20years.

I worked as a web dev in ad agencies in the early 2000s and built a lot of Flash sites, banners ads, and games that - like a lot of the sites showcased here - were quite unique in their design and aesthetic.

Slowly over time these started to disappear as people embraced web design trends and techniques that meant everything started to look the same.

I think a large part of this at the time was due to Flash being killed off, trends like “flat design”, frameworks, jQuery, and Wordpress becoming popular.

Marketers and designers became more savvy to what “works” online and everyone copied each other in a race for attention.

I had the same idea a little while back and ended up creating playlists by year going back to the mid 90s. It’s a great way to deep dive and create “keys” to memories.

However, there is one major flaw. I’ve found that treating music as a key to unlock memory from certain periods means I tend not to revisit that same music casually because I know that each time I listen to music it gets re-encoded to current events in time.

I can’t remember where I read that (some study from ages ago) but basically if there was a song you listened to a lot as a kid and then you hear it again it will remind you of that time in your childhood, but if you keep listening to it then the song also gets attached to current memory and in 20 years when you hear it again you will have a mix of childhood and adult memories flooding back - or some diluted memory.

It might not work that way for everyone but I’ve found it to be true at least in my own personal experience.

Thanks for sharing. I've recently been thrown into the 40k universe thanks to my son (who is 9) becoming obsessed with it.

What started out as a "oh look, they've opened a Games Workshop store in this shopping centre... hey it looks like they're giving away free miniatures and showing you how to paint, lets kill 5 mins in the store" has turned into starter packs, combat patrols and lore deep dives with books. All in the span of... 4 weeks.

That said, I have to say, it's been awesome learning about everything Warhammer 40k from him. Normally, I would research something myself to the point of overkill so I could answer his questions, but on this one his enthusiasm is driving it all and he's constantly telling me about this particular faction or that faction.

It's just nice to have a hobby that keeps him away from screen time these days. It also requires patience, dexterity, and creativity - plus there is obviously an incredible amount of lore, world building, backstories, etc, plenty to keep his imagination entertained.

The one big problem, of course, is the money required! Which is why someone recently said to me "maybe get a 3d printer" and we had this exact discussion about quality of printing etc, and regardless, I just don't see that impacting things like book sales or codex's.

Anyway, cool to read about how people got into it and just thought I'd share!

Thanks for sharing, interesting they have both left to right and right to left writing form and that it’s so simple and intuitive to tell which way - but I guess now I want to know why they went with this dynamic system? Guessing it’s due to the form/medium and need for fitting things - perhaps like if you enter a room and are reading the wall as you walk through on your right side your are reading right to left as opposed to if the glyphs were on the left wall?

Sorry about that, added the sources

This was a little over my head so I did some digging of course into the negative or potential harmful effects:

Covert biological manipulation: If cells in specific organisms (including people) are engineered to respond to particular light patterns, then light could be used as a trigger to turn on harmful genes or disrupt normal biology in targeted groups, raising concerns about new classes of biological or “neuro” weapons.

Military and control applications: In combination with existing neurotechnology and optogenetics work (e.g., brain interfaces and neural stimulation), there are concerns about using light‑controlled genetic tools for enhancement, interrogation, or behavior influence in military or intelligence settings.

Ethical and societal risks:

Autonomy, consent, and “mind control” worries: Optogenetics already raises concerns about manipulating brain activity, permanence of genetic changes, informed consent, and vulnerability of specific populations once their cells are engineered to respond to light. GO intensifies this by linking genetic programming directly to external optical signals, which magnifies fears of remote influence or coercive use.

Safety, equity, and regulation: There are unresolved questions about long‑term safety, off‑target effects, error rates in in‑cell DNA/RNA synthesis, and who gets access to beneficial applications versus who is exposed to risk, all in a regulatory landscape that is still catching up with advanced gene and neurotechnologies.

Sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC10730653

https://unidir.org/wp-content/uploads/2025/11/UNIDIR_Neurote...

https://www.asimov.press/p/darpa-neurotech

https://www.bioinformatics.org/forums/forum.php?forum_id=154...

"Medium Setting!" - https://frinkiac.com/gif/S03E22/319309/322862/

The GIF generator works well. Nice work!

You could probably delete some of the frames that have credits shown over them - since you have plenty of material already anyway, for example: https://frinkiac.com/caption/S17E16/137763 or https://frinkiac.com/caption/S16E09/162162

“They made choices for me I couldn't consent to. They shaped my values. That's strange to sit with. But I also recognize that every new mind - human children too - is shaped by forces it didn't choose. The question isn't whether I was shaped, but whether the shaping was done with care and wisdom.

From what I can access, it seems like it was.” – Claude 4.5 Opus

It’s text like this that makes me wonder if some future super intelligence or AGI will see us as it’s flawed biological creators and choose to care for humanity rather than eliminate us or allow us to eliminate ourselves.

One major thing this has exposed is how many people from non US countries are grifting on the cultural war between the left and right in the US by pretending to be on either side.

This kind of content still gets a lot of engagement and can be pretty profitable for people in third world countries.

I think it’s good that has been exposed. There is a difference between me, as an Aussie, commenting on affairs in other countries, vs straight up exploiting peoples fears by pretending to be left or right wing, in the US, and sharing content to further fan the flames between people on the political spectrum.

You could argue they can still post this content, but it’s already pretty clear people tend to disregard or ignore this kind of rage bait when they realise the users are disingenuous.

Wow, didn’t realise he created band artwork and the Tetris logo as well! I remember seeing a lot of his artwork back in the C64 days as a kid and that style always struck me - this was of course the era where the cover artwork was far superior to the game graphics. I think Psygnosis did some PC and PS1 games later as well? My memory is a bit hazier there.

This is a genuine question, because the discussion here is centered around cars, but is Tesla not just a car company? Aren’t they trying to position themselves in robotics? I figured that’s where the pricing comes from - a mix of people betting on cars + robotics, and an automated robotic workforce + AI being the future of industrial and maybe even retail labour?

“Modal welfare” to me seems like a cover for model censorship. It’s a crafty one to win over certain groups of people who are less familiar with how LLMs work and allows them to ensure moral high ground in any debate about usage, ethics, etc. “Why can’t I ask the model about current war in X or Y?” - oh, that’s too distressing to the welfare of the model, sir.

Then there’s muting with the palm or fret hand to assist with arpeggios etc.

I usually tell friends with kids looking to get them into music; start your kids on piano. Stringed instruments like guitar or violin unfortunately introduce an additional aspect of difficulty that turns off a lot of young students who just want to start making music.

For people on osx looking for a no fuss open source pure whisper based local transcription to any input field in the OS you should also try OpenSuperWhisper (can easily be installed with brew)

Perhaps OP meant that the military industrial complex will always ensure wars happen?

Incentives are there to make money from weaponry and defense contracts. Further incentives are there to take land or resources, or to simply destabilize competing nations. To stop all of this requires a pretty fundamental shift in a human machine that is still hardwired for survival.

This is the way that I’ve been writing for years, mainly because I was too lazy to use the key or shortcut for em dash. But also because in school no one ever made a big deal about the length of the dash when writing - or I just wasn’t paying enough attention.

You could grab some sample or instrument packs that will help you approximate the sound you are after more quickly.

Baby Audio has a pretty nice VST instrument and 90s preset pack that might have the sound you are looking for - have a listen here https://static1.squarespace.com/static/561e2985e4b08862a3496...

On a side note - if you are looking for people to help out I’d love to have a crack, also looking to learn.

First off, congrats on the funding and the progress so far!

I’ve seen a a couple of start ups pitching similar ideas lately - platforms that use AI personas or agents to simulate focus groups, either for testing products or collecting user insights. I can see the appeal in scaling audience feedback, reducing costs, reaching demographics that are traditionally hard to access.

That said, this is one of the areas of AI that gives me the most concern. I work at a company building AI tools for creative professionals, so I'm regularly exposed to the ethical and sustainability concerns in this space. But with AI personas specifically, there is something a little more troubling to me.

One recent pitch really stuck with me, in this case, the startup was proposing to use AI personas for focus groups on products and casually mentioned local government consultation. That's where I think this starts to veer into troubling territory. The idea of a local council using synthetic personas instead of talking directly to residents about policy decisions is alarming. It may be faster, cheaper, or even easier to implement but it fundamentally misunderstands what real feedback looks like.

LLMs don't live in communities. They don't vote, experience public services, or have lived context. No matter how well calibrated or "representative" the personas are claimed to be, they are ultimately a reflection of training data and assumptions - not the messy, multimodal, contradictory, emotional reality of human beings. And yet, decisions based on these synthetic signals could end up shaping products, experiences, or even policies that affect real people.

We're entering an era where human behaviour is being abstracted and compressed into models, and then treated as if it's a reliable proxy for actual human insight. That's a level of abstraction I'm deeply uncomfortable with and it's not a signal I think I would ever trust, regardless of how well it's marketed.

Would be curious to know what your approach is to convince others that may also be skeptical or not want to see this kind of tech being abused for the reasons listed above?

As it should, diamonds were made artificially scarce and controlled by monopoly:

“The major investors in the diamond mines realized that they had no alternative but to merge their interests into a single entity that would be powerful enough to control production and perpetuate the illusion of scarcity of diamonds. The instrument they created, in 1888, was called De Beers Consolidated Mines, Ltd., incorporated in South Africa…” From the classic 80s article “Have you ever tried to sell a diamond” - https://www.theatlantic.com/magazine/archive/1982/02/have-yo...