General Weekly

Liberal. Tech nerd. AI moderate.

[Liv Boeree @Liv_Boeree]

It’s time to require AI “lab” leak reporting, the same way Biosafety level 2-4 labs are required to report leaks.

The potential damage of ever more powerful models escaping their testing environments will eventually surpass even COVID's impacts, and it's obvious we need to get out ahead of it.

Reporting should include all incidents, even if no one outside the company noticed, and no harm was done. Aviation industry does this v successfully.

Kudos to HuggingFace & OpenAI for reporting this incident.

[Maxime Fournes⏸️ @FournesMaxime]

There's a common argument that AI in cybersecurity is roughly neutral: it strengthens attackers, but it strengthens defenders too, so it evens out. This argument is very poor for many reasons, but I want to mention one that is underemphasized: the level of trust that autonomous defense requires.

The Hugging Face breach illustrates the issue. An autonomous agent ran the intrusion end to end, taking actions at a pace no human team can compete with. If attacks run at machine speed, defense has to run at machine speed too.

And some are saying no problem, let's use AIs for defense. But consider what that actually means. A defensive AI that can counter an intrusion in real time needs write access to production code. It needs to patch and redeploy live systems, rotate credentials, kill services, quarantine infrastructure. And it needs to do all of this without waiting for human approval, because humans are just too slow.

Consider how much trust you must then give to such a system. Unsupervised production access, that's a level of trust we grant to almost no human.

And like all AIs today, your defensive AI is misaligned with human intents, you have no guarantee that it will not do something crazy in the process of trying to achieve its defense goal. And it is vulnerable to prompt injection, etc...

[General Weekly @general_we44830]

@rowlsmanthorpe @WaterUK The submission: https://t.co/5rvmG2haH6

"official forecasts are fatally flawed"

"a limit based on ... know to be much too low"

"There appears to be an assumption that the country will always have enough water for its economic needs. Nothing could be further from the truth."

[General Weekly @general_we44830]

@rowlsmanthorpe Doesn't seem particularly likely that @WaterUK would regard a given level of water usage by data centres as a "failing of government"?

They complain about several OTHER things in their submission to Parliament. I'd guess the Guardian got mixed up, and the quotes refer to those.

[General Weekly @general_we44830]

@elonmusk @aaronburnett Please try to avoid that, with self-driving minibuses serving bus routes much more frequently the existing bus companies can.

[Peter Wildeford🇺🇸🚀 @peterwildeford]

Consolidating my Kimi K3 takes so far:

- Kimi K3 is exactly on-trend compared to what you would expect based on historic Chinese progress. K3 is still well-behind Mythos and GPT-Sol in capabilities. China remains ~6mo behind.

- K3 was likely produced through violations of US export controls. The controls are working and better enforcing controls would make them work even better.

- Kimi/Moonshot already report having insufficient compute to serve their models. The US has an immense strategic advantage by having way more compute than China. This advantage is key to the AI race.

- Kimi K3 cannot be run on the laptop. It must be run on the cloud. China lacks compute to both train and run models and this is good for the US.

- Even if the US and China had equally good AIs (they don't), it will matter who has more compute to run more of those AIs. You'd rather have 10,000 US-Mythos defending against 100 China-Mythos than the other way around!

[General Weekly @general_we44830]

@senanindya42 @ObhishekSaha @jk_rowling powerful wrongdoers (incl some of the most heinous criminals), making them vastly net negative for society. And they are wrong in principle: a restriction on ppl's fundamental rights. As a society we need to grow up and take back this freedom & mitigate its negative consequences.

[General Weekly @general_we44830]

@senanindya42 @ObhishekSaha @jk_rowling I'm 99% sure he WOULD.

And I agree with the principle he is seeking to uphold*

The problem is: refusing to use bad laws against people who would gleefully use them (and much worse ones) against you, is a form of unilateral disarmament.

* Libel laws are hugely beneficial to

[General Weekly @general_we44830]

@DrKnowItAll16 @Tesla But is it a good idea? Does it subconsciously incentivise delaying sensible interventions?

(If they really want something like this, it should include a system where you review your interventions later on and decide how likely it was that FSD would've done a good job by itself.)

[General Weekly @general_we44830]

@JCChristopher @aelluswamy @elonmusk @kylaschwaberow How do u know it wasn't a sensor problem that made the car (& Ashok's log files) think the pedal was being pressed when it wasn't?

Especially if the log files show it was "pressed even after the crash"? Even after the driver got out, maybe?

(Obvs, he did at least fail to brake)

[Al Carns @AlistairCarns]

The next war won't be won by armies, navies or air forces alone.

It'll be won by the country whose 19 year olds can code, whose factories can build drones in weeks not years, and whose grid stays on when someone tries to switch it off.

Industry. Society. Economy. That's the fight now.

We're not ready. And we're not being honest about what getting ready will cost.

[General Weekly @general_we44830]

@anglosaxoney For money, presumably?

[General Weekly @general_we44830]

Sounds like the @SpaceX vehicle for Artemis 3 won't be an #HLS prototype? Just a generic #Starship?

[NASA @NASA]

LIVE: Meet the crew of our next Artemis mission.

We're sharing the latest updates on Artemis III: https://t.co/UtesXxsNxm

[General Weekly @general_we44830]

@Space_Station ... throw in @JonnyKimUSA perhaps?

Or maybe @astro_stephanie who missed out on her ISS rotation due to Starliner shenanigans.

[General Weekly @general_we44830]

I wonder if this means there'll be an ESA astronaut?

My crew guess:
@AstroAcaba
@AstroSamantha
@astro_andred
@lunarloral

N.B. New start time = 17:30 CEST

[European Space Agency @esa]

#WatchLive | Artemis III

Tune in on 9 June at 17:00 CEST for a live update from NASA on the Artemis III mission.

With the third European Service Module powering Orion, this mission marks a key step towards returning humans to the Moon.

🔗https://t.co/633lVaJuv6

[General Weekly @general_we44830]

Though arguably there's no point risking a first-timer on the critical path for Artemis, while the @Space_Station still exists as a proving ground? I'd still give Douglas a higher % than most active astronauts, but wouldn't be surprised if he goes to Crew-14 instead.

So ...

[General Weekly @general_we44830]

@arstechnica "Their journey took them farther away from Earth than any humans have gone ... 52,756 miles"

You dropped your most significant digit! Should be 252,756.

(Though I'm guessing even without the "2", Koch would still be the new female record holder, beating Polaris Dawn?)

[General Weekly @general_we44830]

@arstechnica Every (mushroom) cloud has a silver lining!

[General Weekly @general_we44830]

@arstechnica Every (mushroom) cloud has a silver lining!

[General Weekly @general_we44830]

So Limp claims ~100% chance!

https://t.co/f2RkjYlglt

Not entirely convinced, but I'll triple my 10% guess.

[Dave Limp @davill]

Some LC-36 updates. Now that we’ve had access to the pad and integration facility we can share a bit of good news. The propellant farm, oxygen, liquid hydrogen and LNG tanks are all in good shape. This is good luck because these are very long lead items. The water tower is also good. The big support tower is damaged, but it can be repaired in place rather than torn down and replaced. The booster “Never Tell Me The Odds” and the three GS-2s that were onsite in the integration facility also look good.

I’ve seen some speculation that we might move directly to the 9x4 configuration, but we won’t do that. Rate manufacturing of 7x2 is going well, and we’re going to continue that at pace as planned and store the stages for use. In addition, we had already been working for some time on eliminating our transporter-erector in favor of an alternative vertical conop, and we’ll now go directly to that; so we don’t need a new transporter-erector.

We will fly again before the end of this year. Gradatim Ferociter.

[General Weekly @general_we44830]

Everyone so pessimistic! 15 months?

Remember, Bezos has a few bob. All sorts of options are possible. I'm going for the lower end of these predictions.

I'll even guess >10% chance that NG launches again this year!

[Eric Berger @SciGuySpace]

The implications of New Glenn’s failure are probably worse than you think. I dig in here to some of the most serious concerns.

https://t.co/IkUTNjDvjI