Tivo came after the analog VCR. Affordable VCRs made time-shifting a thing.

But with a sufficient number of fiber-seeking backhoes* censorship will always be possible.

*"fiber-seeking backhoes" being a stand-in for whatever disrupts the physical layer of the network stack. Substitute radio jammers, power disruptions, etc. as appropriate for your target's network.

The technical person would say it's a collection of protocols and the hardware and software that implement them. The information that happens to be hosted there, other than of course the information necessary to keep the internet functioning (routing tables, DNS records, etc.), is not "the internet."

There is no need for the internet as defined above to die, unless something better has displaced it and almost nobody is using it anymore (BITNET anyone?).

Ignore network traffic until security self-defenses are in place.

Yeah, you can have a buffer overflow anytime, and they can be bad, but the situation explicitly pointed out an "invaders not recognized as evil before armed guards are posted" situation.

If this had been a bad buffer overflow that occurred AFTER the defenses were in place, maybe it would've resulted in a compromise, or maybe the defenses would've rendered it ineffective.

... the machine can be affected by outside input before it's security/self-defense system is ready?

Oops.

"Of every 2 million sold, less than 1 million are affected!"

If - and this is a big if - Anthropic becomes and stays the industry-leader in LLMs AND if - and this is the second big if - the LLM/AI industry as a whole grows way faster than the rest of the economy - then Anthropic's stock appreciation will likely well-outpace the LLM/AI sector and the market as a whole.

That's the "big reward." The odds of that happening with Anthropic or any other particular AI/LLM company, or any AI/LLM company at all for that matter? Not big enough for me to want to play.

It's high-risk, high-reward, much like dot-coms of the 1990s.

There's also the "known unknowns" of regulatory risks both on the "what will AI be allowed to do" front and the regulations on power-hungry AI data centers themselves. We know these are categories of risks, but we don't have a good handle on how likely any given outcome will be 5 years from now. What are the odds that in 2031 America will be like "you want a data center here? Oh sure, no problem" vs. "you want a data center here? Not In My Backyard!" Once the regulatory environments stabilize it will be much easier to price this into "how much am I willing to pay for a given AI company's stock."

Right now, betting on an AI company's success 5 years down the road feels too much like betting on next year's World Series when we don't even know what the MLB Players Union contract will look like, or if there will even be a season next year.

You or the person you are replying to may not need a human editor, but here are people whose grammar or spelling is so bad they could really benefit from an editor, human or otherwise.

There are also people whose first language is not English, or who are from a part of the world where English is spoken or written in a different style or dialect than Slashdot and they want their posts to not stick out like a sore thumb.

... my posts on Slashdot.

The work these shadowy agencies and departments do has been going on for decades.

Even without modern tech, shadowy people doing their work in back rooms of government agencies or in royal palaces are nothing new.

If you get credibly accused of.... So don't get credibly accused of....

That's the reality.

Now, if writing in a style that is "bot like" constitutes "f'ing up" then I'll concede your point. But someday soon, bots will be able to write in ways that, in a pre-LLM era, we would call "excellent writing." Even absent LLMs, computers already out-do humans at some intellectual tasks, like playing chess.

of Gus Grissom, Ed White, and Roger B. Chaffee.

Tests like these shouldn't be a "gotcha." They should be fair.

* Had employees already been through some basic anti-phishing training?
* Did the test include something that someone who paid attention during the training course would be reasonably expected to spot?

There is one other use for tests like these: To survey the population of employees: What % of employees clicked on the link, and what do we need to do organizationally (not "just for those who failed") to lower this number to something we can live with?

But it shouldn't be a "gotcha."