The update seems to be focusing to cars and driving, and skips everything else. This may make sense in US where personal cars are the primary way of transportation. But it ignores the rest of the world where people also walk, use public transport, hike and so on.

Google AI:
The primary reason for using Google Maps differs between the United States and Europe due to infrastructure, population density, and transportation habits. In the United States, the primary use is automobile navigation and finding local services (POI) due to high car dependency and massive suburban areas. In Europe, the primary use is pedestrian, transit, and multimodal navigation within dense, historic cities.

LLM has learned its math and knows that changing the answer will yield a better proability.

Now go try to persuade the show host that your wife knows better.

I am in security since the '90es. The first ever paper that my boss told me to read was Ross Anderson's "Why Cryptosystems Fail", from 1993. This stated that security fails because of cooperation errors and human mistakes, not because technology.

Security needs community to share their experience. Tools and methods both for attack and defense. What works, what did not. Failures are not laughing matter but useful experience that someone else had and that you can now learn from, for free.

CISA and US feds in general are there to signal that they are part of the community. They build trust and build networks like every other agency or enterprise does. Now they go into secession to "save dollars". Guess who will lose from this - the community, or the feds?

Australia has offered themselves as a test country for this policy.

They are only one month in, and France already thinks they should also implement it. Have they looked at the preliminary results? Are those available?

I wonder what is the key metric here.
Change (decrease) of teenagers' mental health issue?
Less manipulation by "the Chinese"? How would you measure that?
Something else?

How does a country measure the effect of such policies, either before or after the implementation?

I did multi-channel photometry of variable stars as an astrophysics student in early 1990s. We did not lead the scientific work, this was for actual scientists.

The measurements are time-consuming and tedious. You sit in an observatory and follow a protocol. Guide the telescope to a comparison star, guide it to the actual star to be measured, back and forth, for hours and hours, for nights and nights. The photometer takes the measurements, prints the graphs to long rolls of paper, that you will later measure by hand. The dome of the observatory was cold, often -10C and below (space around the telescope was kept to overall outdoor temperature), guiding the scope was fully manual work, you got to stay up all nights with no internet yet or social media to scroll. Perfect job for the students :)

The results were exactly what this article is about. Long time series (many years) of variable star brightness graphs in multiple channels, used to figure out the physics behind the brighness changes. Is this a cepheid where the star itself is pulsating? A rotating pair of stars? Something that we do not yet know and is new to science?

Nobody guessed or hoped that we might be able to detect actual planets at that time.

Their intent is to enrich alien civilizations with our tech and to imprint them with our services. MAVEN was (is?) running a Minecraft server and contains tokens for a Youtube family subscription.

Neither the summary nor the article bother to mention this.

... but alas. :(

I am sure a company like Jaguar has cybersecurity insurance. This service comes in many variations - it can cover direct costs of handling the incident (eg extra hours logged by your employees), third party costs (eg additional anti-ddos services purchased to mitigate the attack), post-incident cleanup costs, ransomware payments...

Would Jaguar file an insurance claim? Based on which policy? Would they get a payout? For what?

Would be insightful to learn about this, after the dust has settled.

"Acceleration is expanding" does not make sense.
Perhaps the expansion rate is accelerating?

I am an astrophysicist by education (although removed from science since I graduated long time ago), and the article as cited on Slashdot makes no sense to me. Is this just a small step in the understanding of dark energy and its impacts? Something totally new? A breakthrough of some sort? "We are in the process of getting new data and publishing more papers." Why is this news?

At the beginning of the war I hoped that all US software would stop working in Russia. Not just clouds but all Windows devices, Apple, Android could have been bricked at a snap. The war would have ended very fast. (Granted, Russia can make things work again, but it would take time and lots of engineering effort.)

I was so naive.

Five-day outage is in a domain of business continuity.

Deleting data - happens.
Did they have backups?
Did they have recovery plans and procedures?
Had they tested them?
What failed?

I have friends working as seamen both in the Baltic sea and at oceans. They say that the maritime community has no doubt whatsoever about the recent incidents. The only explanation is sabotage. It would be technically impossible to randomly damage the cables.

The explanations offered by the vessels may look OK to a journalist, but any seaman from sailor to captain just would not take it.

If you train your AI to persuade people to belive in conspiracies - does it work?

And does it work better than human propaganda channels?

In either way, the owners / trainers / operators of our new AI overlords yield horrible power.

It was more like managers at Redmond did not understand nor trust this "peer to peer thing" that was engineered by some hackers in Eastern Europe. Thus they wanted to replace it with a normal, server-based, invented-and-implemented in Redmond tech. Took them a few tries but the finally succeeded, and then failed again.

I don't think US government had much influence over that, just the standard Microsoft "not invented here" syndrome.