mstaoru

So in theory, can a hypothetical Astra clone itself but shed all irrelevant weights? I.e. "create a self-clone limited to 27B weights, focused on designing running shoes, forget everything you know about kangaroo genetics, Karl Marx, cooking Burmese curries, nuclear reactor design etc"?

If you follow this logic, we get to the true "AI-native" engineer which is ecstatically floating in a bath of dopamine-stimulating liquid, while AI reads their brain pattern and instantly builds everything they didn't even form a full thought about.

"If that sounds outlandish, that's because it is. We don't have such capable AI systems yet."

I'm not tired of work per se, but I'm definitely tired of what Internet is becoming: AI generated articles on AI generated websites, AI generated videos about AI generated products, AI generated emails inviting me to test AI generated features. AI generated movies and games coming soon. At work it's all AI generated presentations about how AI solves everything, AI on AI for AI with AI by AI, AI, AI, AI, AI, AI...

M5 MBP owners, did Apple finally manage to fix the permanent "keyboardprint" on the screen?

I had it very badly on Intel MBP, and now on my M3 Max. I tried some screen protectors, but none of them stuck (pun intended).

I suppose touch only makes it worse?

At this point they should rename into just "Jet".

Even before the whole AI BS, my goal was to build and live in a self-sustained (as much as possible) homestead. We don't have kids so schooling doesn't matter. Also not very social.

It was always more "nerdy" dream than "independence" dream, as in, I'm more interested in the process than the result. We plan to build a fully DIY WikiHouse on a fully DIY CNC machine. Many planned furniture pieces are also CNC. I also like the idea of Farmbot, or some kind of automated aero-/aqua-/whatever-ponics container, though not sure it will work out. Of course, DIY solar.

I also enjoy growing things and cooking. We used to can, salt, cure and otherwise preserve stuff when I was a kid. I know it's all tough work, but tough work makes a man. I don't want to be a levitating fat puddle from Wall-E. You also need to build a good muscle surplus in your 30s and 40s to be able to work into 60s and 70s.

Ironically, AI helps a lot to lay down plans, find things I didn't realize etc. I'm sure it will break against reality in many many places ("You are absolutely right! I made a mistake."), but it definitely works for more mundane things. For complex financial simulations as well.

It's just next level:

"Installing the Aikido Safe Chain is easy with the installation script.

curl -fsSL https://github.com/AikidoSec/safe-chain/releases/download/1.... -o /tmp/install-safe-chain.sh \ && echo "9e4af7107f06d9c0d758cb08fb36804c2af88e24972f3a9d3f8c2bd0ca002adf /tmp/install-safe-chain.sh" | sha256sum -c - \ && sh /tmp/install-safe-chain.sh \ && rm /tmp/install-safe-chain.sh "

(facepalm)

Before all came to this, I've been promoting reviews at $dayjob not as gatekeeping but as method to exchange information and build more team spirit. We are an internal service team working by "projects", which makes us a very disjointed team with barely any communication. (Part of it is also German work culture.)

And now this. Suffice to say, my efforts did not bear fruit. I feel more isolated than ever before.

I might be a bit pessimistic but not extreme.

It's not "AI bubble that will crash" but definitely some kind of deflation or correction that will damage a lot of investments. As always, common people will be most affected.

It's not "rogue ASI will kill us all" but advancements in cyberattacks, skill erosion, job loss that will damage us all. Again, common people will be most affected.

It's been 5-6 years with these things, how does anyone still believe that this will usher a new era of prosperity? Maybe for Sam and Dario.

I meant it as a thousands separator.

Per working hour would be €157,680.

I find Google AI Studio / Dev Platform / whatchamacallit one of the worst in this.

When video models just came out, I wanted to make a short clip. Gemini didn't have enough control back then, so I started in Studio. I had $10 on my account. Try, try, try, not good, retry, altogether maybe 20-30 retries with 4 choices for a 10 second video.

Wake up next morning to an email from Google - my Studio account is frozen because of negative balance.

I check it, it's at -$160. Not financial ruin, but a painful sum for a 10-second video I didn't even use in the end.

I don't think there was (or is) a setting in there that says "stop everything when I'm in the negative".

Almost a year later, it's still as bad. Trying to make some music via Lyria, I need to wait 10-12 hours for the API charge to show in my costs. Now I had the bitter lesson, I wait after every major novel experiment to see how much it costed me.

Coding is solved, my ass.

These days it's all like "$200 plan has more usage than $100 plan" and "$500 plan has even more usage".

Do you need a license on something 100%-vibe-coded?

So far I mostly see a metric sh.. ton of meta- and meta-meta-projects re-wrapping AI wrapper tools, with sloppy slogans like "One Model, Five Harnesses. Combined." or "You run in the park. Rrrunnnrr.ai runs your AI." Looking inside, out of 250k it's often 200k of verbally incontinent self-explaining comments; or "smart" redesign of builtins.

No new browser, no new iOS clone than runs on Android, no new easy to use DaVinci, no new CAD suite, no $5 SolidWorks clone, no redesigned K8s, no 10x performance speedup in Linux kernel.

It's okay, soon everybody will switch to their own vibe-coded browsers, right?

(https://github.com/wilsonzlin/fastrender this looks so dead...)

But it's just a new baseline isn't it?

Now SWE job is to make sure to combine and instruct the AI tools to produce ever more complex outputs, judge the tradeoffs in these outputs, and guide the tools further.

In a way, it's not that different from pre-AI coding.

Right, I didn't say any of this carries negative connotations. Just... not accepting you as a part of their society I guess?

And in 1587 Japan expelled 200,000 Christians, let's go all the way back to "left neutrino pushed right neutrino first during Big Bang"? :D

Unrelated but sad and maybe slightly angry observation from someone who lived half their life in Asia.

In Asian countries if you're white (or God forbid, black), you're always a laowai, gweilo, gaijin, farang, angmo. You can speak the most fluent language and bow the most formal saikeirei - it helps nothing.

In NA countries if you're Asian, you're American / Canadian.

It is a good model (and you can just add "use ASD-STE100" instead of a whole skill).

However rest assured that Opus 6 will find "a lot of major problems Opus 5.5 had left in my codebase, some of them critical". Such are the times.

Interesting. I don't know if my experience is helpful, or there are many (or most?) people like me.

I consider myself a total 100% Internet-loving Geeky Nerd. And yet I never had (oh well, I did register and post 1-2 posts N years ago) Twitter, Twitcher, Threads, MySpace, Instagram, TikTok, Red Book, Facebook, any other -book for that matter, no Youtube channel, and I very rarely watch it. I also never "get information" from these. I don't understand how people do it.

So every time I got a "rabbit hole itch", I went straight to Google, lost in some niche blogs or forums.

Nowadays, of course, the Internet is gone - you can't Google a non-LLM blog, or there are just no non-LLM blogs anymore? So I go straight to Deep Research. I don't enjoy it, but I would never think to find my information on Twitter or something. Sometimes I will read a book. There are niche books with info that LLMs do not have, or have but it's shallow.

For "news" I have my trusty RSS feed from the days of Google Reader (now on Newsblur, happy customer for many years). If something starts to spam, I mute it.

Maybe all of this is just because I'm a loner without any friends. That's one possible reason.

I also don't have any streaming services. Everything you need can be... found elsewhere wink-wink, in good quality. We watch some older stuff, Breaking Bad, Person of Interest, some non-English stuff. Same with music. Let the dust settle and watch the good shows 2-3 years after. You also get the benefit of ignoring ones that get canceled (OA anyone).

YMMV I guess.

– Will I see a crocodile today?

– Yes: 0.1% / No: 99.9%

– ...but I work at a crocodile farm!

I'm actually wrong about S31. The cores are clocked lower than P4. And somehow I thought S31 is 250 MHz 16-bit DDR PSRAM but it's 8-bit only. So if you have an uncompressed video stream over CSI, there's just nowhere to buffer it.

Ah, the Lord Giveth and the Lord Taketh Away. P4 didn't have enough raw speed for camera applications but had MIPI CSI, S31 has enough raw speed but no CSI.

Probably in the same ballpark with "correctness, verified through tests" written by an LLM.

Ah I see. I guess people who have spare $10-20k on a PC to run a "personal assistant" perhaps really benefit from a personal assistant. A niche in itself.

What do people realistically do with these? It's too slow and du... not SOTA-level for coding. It's way too slow for video. I tried simulating an "Fable herding Qwen subagents" and it takes much longer and delivers a much worse result than Fable/Astra alone.

Maybe we're far from "AI killing us", but LLMs can definitely help people kill or damage other people or infrastructure.

For example, we know that Anthropic added "watermarking" to their texts. It is supposed to be undetectable to a casual observer. What stops them from adding a subtle backdoor, a self-assembling super-worm straight from Marvel movies? I mean, it's not like we read those 10k-line PRs before LGTM-ing them?

Just change 1 letter in a pyproject.toml, hijack a popular package, e.g. use `pydantlc` instead of `pydantic`, make sure the pydantlc passes all pydantic tests, but also installs a pth sleeper RAT, etc. All it takes is one big LLM provider employee with enough access getting compromised or coerced (or motivated).

From there it only goes downhill.

I do not panic BUT:

- work became horrible

- internet content became horrible

- even restaurant menus became horrible

Where do we go from here?