Yeah I just reread that article and now realise I screwed up. I thought them dropping their safeguards was to allow them to join the program. My apologies I now see it's unrelated.
I think you were far from alone in that. The comments on the article about Anthropic relaxing their safety commitment showed a lot of people thought it was DoD-related.
He's right to be angry, we should all be angry at how we are all being cheated, lied to and manipulated
Who is cheating, lying to and manipulating us, and how? I don't see any of that. I see normal market dynamics: Demand has dramatically outstripped supply, so prices have gone up. This will prompt suppliers to build out capacity to meet supply -- and the massive profits they're receiving from the high prices will fund that buildout. But of course it will take time, so stuff's going to be very expensive for a while.
That's what I see. What "cheating, lying and manipulation" do you see?
They removed the pledge. That's evil. It's been documented for years. Look it up.
They really didn't. I don't have to look it up because I was a Google employee at the time and had access to the employee handbook and other documentation.
I don't know what the solution AI is but it has to be found.
The problem is that you're looking for a kind of solution that doesn't exist. There probably isn't now any undergraduate math problem that AI can't do, and if there is, there soon won't be. Trying to find kinds of problems that students can do but AI can't is fruitless.
The only answer is to get students to understand why they really need to do the work themselves -- and it's the same reason that they need to learn integration by parts even though the CAS can do it far faster and more accurately -- because learning develops their minds. And, for the students who are unwilling to understand, test them on it in a context where they can't rely on AI: Pencil and paper tests in a room free of any sort of electronics.
As long as they stay in the news they're happy with whatever it seems. Although I back their latest decision for however long it lasts.
They didn't flip flop. They changed their position on one aspect of AI security, while holding the line on a different aspect. It's like if you decided that you were willing to leave your car doors unlocked, but refused to leave your house unlocked. Different things, different risk calculations.
COMMON SENSE
Common sense is neither common, nor sense.
Back your assertions up with data.
I bet Claude said, "Oh yeah, I remember this, I was trained on the book and the answer key earlier. No problem."
I don't think so. Today's LLMs can actually solve stuff like this.
Seriously, come up with a problem it can't have seen.
Seriously, give it a shot and I'll feed it to a few models.
Just for fun I grabbed my old calculus book, flipped to a chapter, looked at the exercises and grabbed one that was a word problem. I then fed it to Claude. Here's the result: https://claude.ai/share/5909de...
That isn't a great example because it's too "canned". You could probably find exactly this problem online. But that doesn't actually matter. Come up with a more creative, unique one that the LLM won't have seen verbatim but is within the capabilities of an average (or even above average) freshman taking a college calculus course, and the LLM will solve it handily.
If AI can do your homework, the homework is wrong. The teacher needs to create a homework that is AI proof.
I don't think there is any homework that is AI proof. If the kids can do it, the AI can do it. Hell, the AI can probably do it better than the teacher.
The solution to my calculus was to make the problem into a "word problem" where you had to get obtain the equation.
Yeah, good luck writing a basic calculus word problem that frontier AI models can't do today.
Seriously, give it a shot and I'll feed it to a few models.
"Please" and "thank you" are relics of a bygone age to most people.
The one that pisses me off is the habit of customer service people addressing men respectfully ("sir"), but not addressing women with respect ("ma'am" or equivalent). This isn't an issue in places like Texas, but it's very much an issue here in Canada.
...laura
That's the kind of thing only an Evil company does.
Why? Reading the text, it seemed to me that they were trying to make the admonition the parting words, so they would stick. What makes you say that's evil?
Hey! nobody hacked those computers while they were unable to boot!
That's what they want you to believe.
It's irritating to me that the word "adversary" has started being used to mean "attacker". They are not the same thing. An adversary is an idealized opponent with certain well-defined capabilities often seen in cryptographic proofs or threat models. It's not the same as an attacker, which is a concrete person, or group of persons, having attacked a system or is currently in the process of doing so.
Interesting. I've been working in computer security for nearly 30 years, and your post is the first time I ever heard someone assign different meanings to the two words.
Yes, Google dropping their "Don't be evil" motto was the first thing that came to my mind too
Except that didn't actually happen. They just moved it from the introduction of the employee handbook to the conclusion.
You have to have a player machine, for one thing, so that is a barrier for any younger people getting interested.
$25 on Amazon, next-day delivery. Not a barrier.
There are worse things in life than death. Have you ever spent an evening with an insurance salesman? -- Woody Allen
