News for hackers. Stuff that matters.

ceejayoz

Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @12:28PM (#50355069)
The real world seems like it might come as a shock to someone who learned for two decades from an endlessly patient computer.
Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @12:27PM (#50355045)
Isn't it always that time of year at a large corp?
Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @12:24PM (#50354999)
That's more honest, but quite the goalpost move.

The local population is probably allowed to complain that you built the reservoir with toxic waste, too.

Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @12:20PM (#50354921)
I'm not talking about Copilot, I'm talking about IE6 back in the 90s, where Microsoft took the "oh woe is us it'll be so hard to undo the deep unnecessary integration we built for this very purpose" tactic.
> playing at nfl level and kid level is something very different

Correct. NFL payers are better paid, get better medical care, have better training and equipment, and their brains are more developed.

Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:55AM (#50354482)
> The exception to this this where a source contains significant additional information that is not in the original source…

This will virtually always be the case of a reputable news article about a PR announcement written by a corporate marketing department of this nature.

I would also note the same guideline says to use the original title, which was (and remains, as submitted) "Google Will Turn Off Search for Schools That Reject A.I."

Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:49AM (#50354388)
> Like you can't have juice that doesn't contain water.

This particular bit of juice didn't have the "AI summaries" ingredient for 93% of its existence.

Schools invested enormously in Google Classroom and Google for Education, and now the deal's changing. Not for the benefit of students.

Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:46AM (#50354345)
> How could they possibly do worse with it running?

It's almost always possible to do worse.

Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:44AM (#50354326)
That's the same argument Microsoft made with Internet Explorer.

https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor....

Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:41AM (#50354270)
> Experimentation is one of the best ways to learn, breaking stuff is a great way to learn.

Breaking stuff sometimes is a way to learn "don't break that".

Broken children can be tough to fix.

Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:34AM (#50354171)
I'm not a fan of "by ceejayoz" on a link I never submitted, yeah.
Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:34AM (#50354162)
That was originally this link. The mods changed it.
Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:18AM (#50353941)
That's often blocked on school-provided devices.
Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:15AM (#50353886)
I really don't like that my original submission "Google Will Turn Off Search for Schools That Reject A.I." from the New York Times got replaced with "Designing responsible AI for learning" PR slop from Google Corporate that doesn't even disclose this limitation.
Re:Google Will Turn Off Search for Schools That Reject A.I on Tuesday November 03, 2026 @11:08AM (#50353781)
> A.I. features are now a standard part of Google’s commercially available search engine. “Search and A.I. are deeply integrated,” said Chris Phillips, vice president for education at Google. If a school has banned A.I., he added, students “will not have access to search, and they’ll see a message from their school that indicates it’s not available.”

C'mon, it's right in the article.

The school can control AI on/off. Off means no search too.

> likely to Quebec City

But that's... already on the map, and you described it as "a major presence" upthread?

> nothing at the scale of the US midwest

Again, I note the very different ingress/egress requirements of AI compute. The main bottleneck has been power, not bandwidth.

> Your map maps the same point as mine.

But it's the lines between that matter, yeah?

Again, your map would say datacenters in the Australian outback would be insane, but there's huge dark fiber capacity passing through it.

> I also think you're massively underestimating the amount of ingress/egress traffic needed for one of these facilities.

Ingress/egress for AI compute is clearly different than some other use cases. One needs only to watch Claude's network activity for evidence of this.

(Like, it's the least problematic part of the in-space datacenter idea, versus cooling and maintenance.)

https://www.bceglobal.net/our-network

> Compare to the US where the entire middle of the country is covered in PoPs measured in terabits per second

A problem if you're Netflix. Less so if you're Claude, where giant clusters of RAM and compute generate a few kilobytes of text leaving the datacenter at a time.

But there's fiber connecting said dots the parent poster is talking about leveraging, right? Your map appears to only include submarine cables; see Australia, for example, where there's a huge on-land connector between Perth and Melbourne/Sydney (https://infraco.telstra.com.au/solutions/intercity-dark-fibr...) not represented.
That map shows zero on-land connections, as far as I can see. With the possible exception of Brazil, but that might be a cable down the Amazon from the looks of it.
No one's allowed to put down roots in Antarctica; colonization is explicitly forbidden by treaty.

A couple thousand people live in Svalbard, though.

Why not? The big AI labs have provably failed to properly sandbox models.
To a point.

https://en.wikipedia.org/wiki/Feist_Publications,_Inc._v._Ru....

> In regard to collections of facts, O'Connor wrote that copyright can apply only to the creative aspects of collection: the creative choice of what data to include or exclude, the order and style in which the information is presented, etc.—not to the information itself. If Feist were to take the directory and rearrange it, it would avoid the copyright owned in the data. "Notwithstanding a valid copyright, a subsequent compiler remains free to use the facts contained in another's publication to aid in preparing a competing work, so long as the competing work does not feature the same selection and arrangement", she wrote.

A lot of this shit is gonna have to wind its way through the courts before we know for certain.

> Anthropic have decided...

The headline says "allegedly leaked".

> Long-term low-current electroplasticity, induced by the electromagnetic waves from the Arecibo telescope itself, may have sped up the zinc creep, it says.

I presume we aren't using most bridges as interstellar antennas.

Re:Product Managers aren't safe either on Friday October 30, 2026 @05:09PM (#50310170)
There's more than one kind of project manager.

The "I just manage the Gantt chart" folks are screwed, I think.

The ones that are good at translating between teams can be a huge force multiplier.

Older »