You have no clue if you think they dont. The M5 max GPU is roughly as fast as a laptop 5080. If you want either pro apps, or AI ability, the M5 isnâ(TM)t beaten by anything at all.

That doesn't really make any difference to the RAMpocalypse though - the thing that's hard to get is the RAM chips that go into those tightly integrated packages. Apple is having just as much trouble as everyone else getting RAM.

I'm not sure what the distinction you're making is.

The memory manufacturers stopped making your desktop memory in order to make RAM for AI data centres.
AI data centres are taking all the RAM.

These are the same statement.

Theyâ(TM)ve been hypothesised for a long time. Actually finding that little red dots actually are them, ⦠no, thatâ(TM)s not been a thing since 1994.

I mean, we don't know at all if they're making profit. I very much doubt it, but you're asserting that with extreme confidence, when it's not known.

At least in the UK, when you're operating a boat with a VHF radio, it's a requirement to leave it on channel 16 when not actively using it. Is that true in the US?

Domestic panels will get more expensive, too, as they are made from polysilicon.

I really don't think that's true. Scientists studying the last universal common anscestor have found extremely short pieces of RNA that can self replicate https://www.science.org/doi/10.... The machinery of a cell is comparatively enormously complex compared to this simple self replicatiing chemical.

Yes all the time. Using a Tablo DVR

Yeh, but heâ(TM)s also skipping right past âoecomputers can keep track of far more variables than humans.â Watch a Teslaâ(TM)s screen while itâ(TM)s in self drive and youâ(TM)ll observe it tracking 10s or even hundreds of cars exact position and speed. A tesla is already significantly superhuman in some regards. Where itâ(TM)s not is in decision making.

I mean, itâ(TM)s fine to make the argument that âoewe want as much redundancy as we can because weâ(TM)ve not found the place where more redundancy doesnâ(TM)t make things significantly betterâ, but thereâ(TM)s also a question of whether not perfect, but significantly better than human acceptable if it means that we donâ(TM)t have ugly, non aerodynamic bumps with lidar systems in them.

What theyre really arguing is âoeour system canâ(TM)t be made to perform well enough without lidar, so youâ(TM)d better accept the big bumpsâ.

That's what they're doing - they're storing your encryption key on your devices. That's why they can't access it, and nor can they give it to the UK government.

Why not? Some of them could be the next great astronomer.

I donâ(TM)t understand - what better monument to the sky gods could you build than an absolutely fecking mahoosive telescope?

Why isn't there a second independent AI that examines the main AI's output, and decides if it should be censored or not. If it thinks so, the answer is not delivered.

Since the second one is only looking at the first AI's output it seems like it would be difficult or impossible to fool it, especially if it's window is very short (like only the current message).

I assume there is some reason this does not work. Any explanations?

Those are visible harassment campaigns (and visible on purpose to scare others). He was wondering about ones people don't know about, secret and/or unreported harassment campaigns.