Any experience? Are there any with updatable records through some API that does not just give access to the whole account?

What is the average damage caused by OpenAI/Anthropic hacks? I feel like we should aim at lowering that number, not increasing it.

What's the rule for fully autonomous cars?

Should we allow this as unsupervised at all if we are really afraid of deaths, while there are practically no states that allow 100% autonomous cars (where we are actually afraid of deaths)?

If "an independent third party registered with the Department of Commerce would periodically evaluate the lab", will this turn into an expensive certification/insurance that cuts off smaller competition (totally by mistake, I'm sure)?

I am also very very sure this is most definitely not at all going to become like product certification if EU, where there are literally 2 companies certifying various standards: they police each other and a stupid electric fan can cost you tens of thousands of euros to certify.

Are you sure you are not thinking about autonomous taxis, that I only ride in, and don't actually drive?

If OpenAI does it, it's gone rogue and it is nobody's fault.

They need to be sued, the laws are already there.

Your prompt, your responsibility.

Been a good time for a while. Too many scandals, too much to look after, too many economical problems.

When people don't even have the mental strength to give blowback, google blocks and constraints android, new mandatory subscriptions come up.

What are you going to do, boycott and switch brands and buy another pc? with these prices?

What I see from that report is that a machine was doing random stuff completely unsupervised and unchecked for more than a month. And the manufacturer did not notice even after the work was finished.

1200 agents and not one dedicated to check the progress over a month?

The problem is not sandboxing, it is quality control that comes from an accountability sink.

The point is that the accountability sink can't be allowed to exist. It can't be a hot potato game, but it can be the fault of multiple people in the chain.

The normal laws, since you are still liable for everything you own.

"Rogue AI" is solved by removing the accountability sink "it was AI, not me!"

It's your chat, your prompts, and all output that you should read and understand.

Don't know how to operate a power hammer? don't care, any problems it causes are still your fault.

...like it already happens in europe, or at least in Italy.

This is basic law and risk management in society. Let's not pretend we have just invented corporations and don't know what to do with them yet.

Quantity has a quality all of it own, and all that.

Ukraine is doing wonders managing tons of different producers and models, but Russia production capabilities have always been a problem.

Ukraine was in the lead with their interceptor drone. Unfortunately now russia has refined their drones to not be multiple pieces sticked together, and has added small jet engines.

The new drones are both too fast for interceptor drones and might even survive a hit thanks to the body made from a single sheet of metal.

Now the new russia drones cost about double, but intercepted rate went down from 90+% to about 55%. Math says the new drones are more than paying for themselves.

GLM-5.3: 753B, 40 Active

I was hoping for something that hinted at smaller models too, but I guess not.

Any competition is still good, especially now that the USA AI labs are starting to do regulatory capture.

Still means that currently the phone has no support for MTE.

I have bought too many things on vague promises that did not happen.

It's not there now -> it's not supported. EOL.

200-400 LOC, 10-100x = 2.000-40.000 LOC/hour for human review?

reviewer: LGTM

Just merge in main, what are you even pretending to review?

AI review should happen before human review, not instead of it.

I see frontier AI giving up and finding only nitpicking things on huge PRs, then finding logic bugs that were always there after cleanup.

Split your PR in smaller ones, both humans and AI will work better.

Q1 was producing some garbage at times, generating wrong urls on webfetch, then convinced itself there was some url rewrite in the middle. With IQ2 it happened much less but still happened, and once it would all webfetches became like that. IQ3_XXS is the maximum I can run: I don't have problems anymore, though I have less available context window.

Surprisingly useful as long as you can leave it running a couple of hours at the very least.

While huge models will still be better I think the general availability of RAM might be the downfall of AI companies.

...for human code.

In the small startup I work for boss (ex-programmer) discovered fable, and ai-coded 15K lines . So much productivity! So great! He even asked multiple reviews and it was fine!

I ask it a couple of reviews and it finds only minor things. The code is a mess of duplication and different coding styles, so I start cleaning it up. After a couple of months the reviews (same ai model) start actually finding big logic bugs that were always there.

We might already be at the point where the Ai-Coder is generating stuff that ai-reviewer can't find and will automatically pass.

--

1M context window is what? 70-80k LOC, tops? Without comments or documentation?

That is a smallish project of a couple of components. AI will remain inherently myopic until it can keep in context whole codebases.

Exposing current problems is fine to me, but I am worried of how brittle AI code will be.

It it completely pointless. you can't even make a "read-only" agent. allow "cat *" for every file? congratulation, that allows "cat file > output" and now you have read write.

Allow python? more free reign that allowing all bash. The models (qwen or claude) will still try to use the disallowed things multiple times.

read/edit permission are bad enough that the model themselves don't understand why they don't have permissions: they double check the conf, and think they should have access.

I am switching to using one firejail per project to containerize as much as possible, and leave all permissions to allow.

I have no idea how to limit network access, and I have no idea how to prompt and steer subagents when they are going off the rails.

The whole thing is built to be completely impossible to limit and steer.

Will they even get their money back in 50 years at this point?

Genuine question, who has done the math on:

* how much investment they have gotten

* how much they need to make each year to give it back in X, Y years?

* how much they are currently making?

Is the end game one gigant megaAIcorp that you can't say no to? The chinese models seem to preclude that.

Have we collectively decided this a big game of investment-chicken, but of course we are smart enough to get out of it just before it crashes down?