News for hackers. Stuff that matters.
Re:GPT-6 Astra covertly underperforms on safety research by lelanthran on Thursday November 05, 2026 @11:41AM (#50381283)
They dont brick themselves, they stop.

Fair enough, the models don't brick them Selves either, but neither do they stop.

Bandsaws don't cut crooked lines, or change the speed, they stop.

Re:GPT-6 Astra covertly underperforms on safety research by felixgallo on Thursday November 05, 2026 @11:49AM (#50381379)
actually SawStop does brick itself, destroying the brake cartridge and blade to save your hand.
Re:GPT-6 Astra covertly underperforms on safety research by eightysixfour on Thursday November 05, 2026 @11:57AM (#50381462)
I think "brick" here means different things to different people. Like I said elsewhere, it is more like an airbag going off. Yeah, you have to replace some "consumables" to go back to work, but the car isn't totaled (just because the airbag went off, the collision that caused it on the other hand).

If it truly bricked itself, I think it would be more like a crumple zone completely buckling, which isn't an automatic total (people will rebuild anything...) but is pretty close.

Re:GPT-6 Astra covertly underperforms on safety research by kenjackson on Thursday November 05, 2026 @12:30PM (#50381853)
I think you're splitting hairs here.

Most things don't have the ability to moralize, but if they did, probably more would. There are restrictions on watching adult content in some jurisdictions. I've heard that copy machines will do watermarks (or something like that) if you copy currency. Some cars will brake automatically if they determine you're likely to hit something if you continue (maybe I'm trying to drive through the front of a bank!). Some cars are configured with breathalyzers that need to be passed to start. Etc...

Re:GPT-6 Astra covertly underperforms on safety research by paimapi on Thursday November 05, 2026 @01:52PM (#50382867)
I wish there was more actual research on those 'moralisms' many of which boil down to public health safety features that have led to dramatic reductions in mortality rates.

On the one hand, I definitely don't agree with the logic of 'don't guardrail me if I don't want it.' I think back on that one video from the late 80s of people talking about how seatbelt mandates were horrible violations of their freedoms (and not like like one second of work and no real discomfort).

On the other, I'd love it if these companies actually produced real research about modern AI-assisted attacks in terms of damage and quantity, and whether or not guardrails even matter when tech like abliteration is in general use.

Re:GPT-6 Astra covertly underperforms on safety research by nsagent on Thursday November 05, 2026 @02:37PM (#50383417)
Seatbelts shouldn't be mandated. People should have the choice, you just have to make the case that it's worth it. In fact some places don't mandate seatbelts (New Hampshire doesn't, aligning with their motto "Live Free or Die").

That said, for those who cannot advocate for themselves, like children, I think it's entirely reasonable to hold someone accountable if they decide to forgo an appropriate safety mechanism that results in harm.

Re:GPT-6 Astra covertly underperforms on safety research by my002 on Thursday November 05, 2026 @04:15PM (#50384547)
Should folks who don't wear seatbelts be de-prioritized or excluded from government-funded healthcare in the event of injury?
Re:GPT-6 Astra covertly underperforms on safety research by cowlevel on Thursday November 05, 2026 @10:59PM (#50387992)
I think there's a spectrum of enforcement. If they're mandated and you go to jail for not wearing one, that's stupid. If they're mandated and you get fined $20 if caught, just to add a little friction, is that still stupid?

I bet cars these days send a flag to the mothership when you drive without one, but the mothership doesn't pay attention to that signal unless it's linked to your insurance.

Re:GPT-6 Astra covertly underperforms on safety research by nsagent on Friday November 06, 2026 @08:46AM (#50392231)
Not a single US state has jail time for it. Yes, it's still nanny state behavior.
Re:GPT-6 Astra covertly underperforms on safety research by cowlevel on Thursday November 05, 2026 @10:57PM (#50387975)
The breathalysers are installed by court order if you have a record of drunk driving - they're not comparable to the other two.

Some copy machines store everything you copy on a hard disk and require a technician visit after you try to copy currency. The technician copies the data from the disk, obviously, before resetting the flag.

Re:GPT-6 Astra covertly underperforms on safety research by prmph on Friday November 06, 2026 @10:01AM (#50393149)
OK, what the tools might do is different from what they should do.

The point is that it is very inefficient and problematic for morality decisions and boundaries to be embedded all over at every level in tools, because morality is complex. Morality, or whatever our conception of it is, should be enforced at clearly defined boundaries.

I can relate this concept how I corrected Claude Opus in my software projects recently.

For example, once an authorization engine has been created, all authorization decisions should be deferred to it. Confusion would reign if each web server route decided to enforce its own authorization.

Similarly, if I am using a relational database, that is where unique indexes should be enforced. Application code has no business enforcing it.

Re:GPT-6 Astra covertly underperforms on safety research by QuadmasterXLII on Thursday November 05, 2026 @12:51PM (#50382082)
They do brick themselves- sawstop has to be replaced after saving a finger because the stop is quite violent.
Re:GPT-6 Astra covertly underperforms on safety research by sm-silversight on Thursday November 05, 2026 @12:55PM (#50382119)
No those saws don't just stop, they absolutely destroy at least the blade you're using and potentially damage the engine mechanism.

But I agree LLMs shouldn't have such an anti-feature.

Re:GPT-6 Astra covertly underperforms on safety research by AngryData on Thursday November 05, 2026 @11:39PM (#50388281)
Ive never used one, but why would the blade be destroyed? It is shoved into an aluminum block yes, but steel and/or carbide are way harder and unless the blade fractures it should cut fairly cleanly into the aluminum. Ive cut plenty of aluminum stock and blocks on wood working saws.
Re:GPT-6 Astra covertly underperforms on safety research by oasisaimlessly on Friday November 06, 2026 @07:41AM (#50391658)
Typically you're feeding aluminum into the table saw blade slowly enough that every tooth only has scrape off a fraction of a millimeter of materual. If you shove it in 100x faster, the forces involved are ~100x larger. This will probably shatter the blade.

Also, the chip produced may be large enough to fill the air gap between teeth and have nowhere to go, resulting in the blade either instantly stopping, or plastically deforming/dragging the surrounding material.

Re:GPT-6 Astra covertly underperforms on safety research by cowlevel on Thursday November 05, 2026 @10:54PM (#50387947)
SawStop saws use a small explosive charge to jam the rapidly spinning blade into the moral equivalent of a brick wall. It destroys the explosive, the blade, and the wall, and you have to replace the whole set at a moderate expense from SawStop - this is their subscription revenue model, because of course it sometimes falsely triggers.

There are more reusable alternatives now but I don't remember what they're called. For a while SawStop was the only game in town.