27183

I don't believe there's anything physically preventing the invention of AI. But nobody has done it. Until someone does it, it's some far off future thing. Anthropic, OpenAI, Google, Microsoft, et al. have not done it yet. They've done some interesting things with language models, but so far there's no sustainable business there. At least at nothing like the scale they've promised the markets.

So the whole thing hinges on, at some point in the near future--before the music stops--one of these institutions making a huge breakthrough and discovering AI. But there's no indication we're any closer to that breakthrough now than we were 10, 20, 30, or 50 years ago. They sure do have a lot of compute at their disposal, but there's no clear engineering plan to use it to scale up AI.

If this was a proven technology, there would be a clear path towards making it work. Advancing its industrial applications would be an engineering problem, not a science problem. AI is still science fiction (or at best a speculative theoretical research topic). That's not something sensible people throw money at, unless they've been scammed.

It would be interesting to scale this design up. Not necessarily in terms of velocity--supersonic velocity for a large trebuchet munition isn't necessarily desirable. What I mean specifically is scaling up the multi-revolution arm and sling release technology to be capable of launching a ~1 ton projectile.

Could it be done reliably with traditional construction techniques (wood, treenails, lashings, minimal metal hardware)?

AFAIK these are innovations that were not explored historically. It would be interesting to see if they would have been viable. It seems like you could probably achieve a lot more downrange energy using the same amount of materials this way.

It is the same business model, just at a grander scale. There is no sustainable growth business underneath it, only vibes and hype, and the "future promise" of some miraculous breakthrough. It's becoming clearer by the minute that no such miracle is forthcoming.

[edit] one small difference from web3 is that AI does seem to be at least marginally useful in some narrow verticals (devtools, infosec). That's pretty inconsequential, though. Those verticals can't supply the capital needed to sustain the industry. So on a macro scale that isn't relevant.

When the dust settles I hope we end up in a place where we no longer accept lies and grift as pitchdecks and product briefs. If you have a technology, you demonstrate it, and only after a successful demonstration do you expect funding. We don't need to throw money at vaporware anymore, that's proven itself to be idiotic behavior over and over again.

Isn't venture capital just playing along too, though? They also benefit from all the myth making and hype storms, even if they are technically getting defrauded while funding it.

They could actually audit the companies they invest in, and go after the frauds. The Nikola example is a great one--if anyone had looked carefully behind the marketing the fraud should have been obvious. But by and large they don't, really. Seems to be a tacit endorsement of the behavior.

> Wasn't it always seen as a "winner takes all" business?

No. It was always a pyramid scheme marketed as a "winner takes all" business. But there's no winning. The only winners are those who cash out before it collapses. This is literally Web3 2.0.

There are species of wood that do very well buried. Black locust is known to last 100yr buried. I wonder if you could scale up the effect using a larger grid of black locust 2x8s [edit] or even smaller dimension mill scrap.

> The worker claimed it was "his first time using AI" and "he didn't know"

This is not the defense they think it is.. If your mechanic rotates your tires and improperly torques the lug bolts "it was my first time using a torque wrench" isn't excusable. It's good to experiment and learn new things. It's not good to throw your little self-study project out into the world with your employer's name behind it and present it as production worthy material.

I think what's interesting here is that they've shipped the product despite these glaring security flaws. I've noticed that in my own professional life, at some point after the pandemic people stopped caring about security as much. Issues that would have (and should have) blocked a product launch were swept under the rug.

I suspect this comes with the territory of enshittification. As an industry we're trying to wring every last dollar from every last eyeball and we've discovered that building secure systems doesn't actually move the needle very much.

In the case of a nation state actor, the most recent update could well be the attack vector.

That's great! Different take on the opto-isolator.

> normal network security

The problem is there's no good way to actually enforce this. Every organization has their own idea of what is "good enough". The NSA has some pretty good advice[0]. But as far as I know there's no written-in-stone engineering standard organizations have to meet, just "best practices". If the building inspector finds fault with the construction of your facility, it gets evacuated and shut down until the defect is remedied. There's no inspector for your network security. That's the problem.

[0] https://media.defense.gov/2022/Jun/15/2003018261/-1/-1/0/CTR...

You can't have both secure infrastructure and Internet exposed infrastructure. I don't know if I'd frame it as incompetence, but it does seem firmly outside the capabilities of current engineering practice.

If you need a computer system to be actually secure, rule #0 is absolutely ensure it cannot receive unauthorized inputs of any kind (airgapped, big Faraday cage, JB Weld all the ports, big scary guys with guns, redundant locks, blast doors, etc). Otherwise you've lost against any sufficiently motivated adversary.

What happens to OpenJDK when Oracle dies? That's looking increasingly likely on a 3-5yr horizon.

Yeah it's true, and the higher up the ladder you go the worse it gets.

"clanker wankers" has a nice ring to it

This just seems like it's shining a light on the omnipresent skill issues in tech leadership. You don't get promoted to leadership positions for thinking critically, analytical skills, or necessarily being good at anything. Instead, you get there by fitting in. Of course those people will be more susceptible to AI psychosis or just simply failing to understand what the tool does, it's the cool thing now and everyone is all-in!

I would not characterize software people as particularly "clear-thinking".. there's a huge buzzing cloud of unexamined cognitive biases and thoughtless herd behavior that swirls around every software organization I've worked for.

There was some truly crazy shit happening in the late '50s-early '60s:

https://en.wikipedia.org/wiki/Project_Plowshare

https://en.wikipedia.org/wiki/Project_Pluto

https://en.wikipedia.org/wiki/Project_Orion_(nuclear_propuls...

https://en.wikipedia.org/wiki/Supersonic_Low_Altitude_Missil...

This is extremely common in urban areas as well, at lower speed. Not necessarily less fatal. The thing that seems similar between flying and riding motorcycles is that even a "small" crash is likelier than you want to have extreme consequences. Airplanes and motorcycles are not made to crash. Cars are.

Weird things can happen with big downdrafts and wind shear, but those aren't terribly frequent and probably a spin is less critical to practice than stall recovery. But it's a very different experience to actually feel the forces and see the world spinning than you get in a sim. Also it's a lot of fun.

I wonder what the human factors consequences of having a GPS/chartplotter working almost all the time and then suddenly fail are. Well, I don't wonder that much, but it would be interesting to see the data. It seems a similarly jarring situation to having an AI chatbot that most of the time gets things right but then occasionally gets something wildly (albeit plausibly) wrong. Or self-driving car technology that occasionally (rarely) tries to drive into the median. It seems like it renders the user more complacent and less able or likely to respond to the failure safely and correctly.

So is it safer? If the answer is "yes, until it fails" that's pretty bad. Things fail, systems degrade, and we need to be able to adapt safely. If the transition between "everything working" and "degraded performance" causes a safety issue... that's kind of a systemic failure isn't it?

[edit] maybe it's a UX issue? What if interacting with the GPS had the same UX as the VOR? The failover would be a lot less disruptive. That's kind of where I always end up with chart plotters.. like, you need to be using the paper chart anyway if you want to be able to fail over safely and quickly... so does the chartplotter really add anything? Or does it take things away?

"Hey Siri, find me gas stations along my route" routinely fails now. That worked great for years. Total clown show.

There's nothing inherently unsafe or unreliable about old airplanes. Aircraft maintenance and inspections are very rigorous and evidence based as compared with almost any other mechanical systems you might interact with. I flew a 1949 Piper PA-16 Clipper with a fairly anemic 108hp Lycoming O-235 with auto gas STC. I'd say it was every bit as safe as any other light single engine airplane, so long as you pay close attention when the wheels are touching ground and mind your density altitude.

In some sense it felt safer than the 172 because I could fly a higher, tighter pattern. It was easier to lose a bunch of altitude on final by putting it sideways in an aggressive slip. The 172's small rudder didn't let it get as sideways, so it was a little more constrained. The 172 did have functioning instruments and dual radios, not just a handheld. It also had flaps. I think it was probably 10-15yr newer? Still a relatively "old" airframe. I probably had 5x or more as many hours in the Clipper so all that could just be biased towards familiarity.

Point being, old plane does not mean unsafe plane. It's not like you're riding around in some rusted out hooptie in the sky. There are rigorous mandatory inspections and maintenance schedules for every single component, which results in mechanical failures being really rare even in aircraft that are nearly 100 years old.

None of that obviates the need to know what you're doing, of course. The pilot in command needs to have control of the plane. I'm pretty skeptical of systems that kind of try to smooth over that sharp edge, I can understand the altruistic motivation behind it, but there's likely a side effect of reducing the perceived risk. That's a problem, because it makes people complacent. So in terms of overall safety, it might not be a win? You're trading on average higher complacency and lower situational awareness for maybe some better outcomes in rare situations? Probably not the best trade.

That stuff just seems awfully gimmicky to me. I'm glad it worked and saved some lives. I don't think I'd want that system in my plane, though. I just wouldn't be able to trust it.

But I also wouldn't want anything to do with the new glass cockpit stuff, way too easy to just lose all your instruments at once. I'll take properly maintained, independent mechanical or electromechanical systems over digital ones every time for things that actually need to work.

Digital systems always seem to accumulate tons of complexity ("features") rendering them extremely brittle. Until we can figure out how to stop doing that, I prefer to avoid depending too heavily on them for important stuff.

The Cirrus parachute system seems pretty simple and trustworthy, that approach feels more directionally correct. Not much more complex than an automotive airbag.

When I was training 20yr ago we practiced stalls, spin recovery, simulated random engine-out scenarios.. do they not do that anymore?

I was pretty confident barring some catastrophic mechanical failure I could probably get it on the ground or into the water (or trees) in one piece. The big question remains though: what now? If you're knocked unconscious (likely--lap belt, big flat instrument panel at face height) then two things can happen:

(1) if in the water, you drown before you can get out.

(2) if you're on land, the fuel tank right underneath that instrument panel or the ones up in the wings might torch you before you can get out.

It's probably about right to compare the risk profile to motorcycle riding.

I've been in this situation too. But I was trained for it and had a mental checklist to follow. I never had a GPS, though. It seems from the details of the article, which you rightly highlighted, that the pilots were using their GPS as a crutch and had no ability to fall back on basic dead reckoning. It seems they didn't know where they were. That's pretty inexcusable, like embarrassing even.

I don't fly anymore, but I see similar behaviors developing on the water. People who use chartplotters for navigation don't have a very good understanding of what's going on around them. If you're just interactively following a computed lay line or a plotted course you don't really need to know what the currents are doing. You don't need to know what the hazards are. You're not actively engaged in navigating the boat.

> Heavy weather and an instrument approach or confounding factors? Not accurate imho.

Then what are you supposed to do when your GPS craps out on approach in low visibility and heavy weather?

Obviously the answer is "fly the airplane". Which means you look at your artificial horizon, set trim and power for max climb, get to a safe altitude and figure it out.

If you're paying attention you already have your chart folded to the airport you're landing at, so it shouldn't be too hard to orient yourself and figure out what's next. Either you'll be making another attempt at this airport without GPS or you'll be going somewhere else.

There's never an excuse for failing to fly the airplane. That's basic airmanship. You need to be able to do it regardless of whether your GPS is working.

Dunning-Kruger lives!

Systems fail for all kinds of reasons, if you're dependent on one SPOF without any backup, bad things will happen eventually. In aviation we're taught to think this way from the very beginning. Always have a spot picked out where you will attempt to land the airplane when (not if!) you lose your engine.

Even in the absence of GPS jamming, you can effectively lose GPS for all kinds of reasons in a plane--a connector failing somewhere on the back of your panel, a wire chafing through, etc. GPS is nice, but it's probably a lot less reliable in most light aircraft than the other options.

It's been 20yr since I've flown a plane, so maybe attitudes have changed, but my instructors always stressed the importance of always being able to fall back quickly and seamlessly to a paper chart, your watch, flight computer, and basic instruments. Because you will lose everything else at some point.

It'll have to be pretty much all the ways. One way will stick. The rest will be forgotten failures.