So I think this (AI-written) thing is mostly interesting because of some of the comments, in which commenters lament Cringely putting out this kind of stuff in the lead-up to his death, which they also manage to verify with an obituary.
I presume we're still talking about software. Auditing proves what?
Accountants audit that the books are consistent. It's extremely hard to fake consistency and do fraud, especially when there's 3rd parties involved. The edges of the arithmetic get nailed down, because bank payments were sent and invoices paid. So the internal consistency is checkable and valuable.
But software. Are you proving it does what you want? But you didn't write down what you want. Or maybe you're into spec-driven development, or detailed wikis, or whatnot, so you did. Most people aren't, though, and those docs are already out of date anyway.
I think there's something here, because we're all telling our agents what we want the software to be. So maybe you could audit against that. But this isn't the same as accounting.
Where his argument falls apart for me is: why? Audit is intended to address misaligned incentives. It's a form of review done by an independent party, because the incentives are the same for everyone within an organization or with a vested interest in its survival.
The AI doing self-checks through sub-agents (or outsourcing to other models) is a form of review looking for errors, akin to getting a code review for your PRs. You don't need an outside firm or auditor. You need different incentives (a reviewer whose goal isn't to finish the work, but to find problems) and/or different blind spots (trained differently than the model doing the work, which might be more inclined towards that error).
We generally assume that humans can make mistakes and institute mechanisms to flag and correct those mistakes, but that doesn't mean we take that to the level of auditing every discrete action a human at a given job takes.
Likewise for agents, particularly as/if their error rate becomes lower than humans at the given task. They'll get a level of review tied to the importance of their work. And yes, if that's bookkeeping, then they'll probably need a proper audit, just like the humans directing them.
It reads like it too. I don't fully understand why a sentence like "It's the signature that costs." makes it into machine output (It costs... what? What does it cost?) But here we are.
It looks like there is a large error in reasoning - in all of the historic patterns he is giving, the verification is not cheap. Rather what drops in price is the creation, but verification costs real effort, hence the rise of trusted third parties.
The argument seems like a more appropriate analogy for the slopweb than the dynamics of LLMs themselves.
According to a HN post, and basically nothing else that I can seem to find.
Who knows, maybe he's still alive and posted the death notice himself. Dude is/was a compulsive liar. There's still to this day no actual evidence that he genuinely has passed away (unless I'm just unable to find it)
So I think this (AI-written) thing is mostly interesting because of some of the comments, in which commenters lament Cringely putting out this kind of stuff in the lead-up to his death, which they also manage to verify with an obituary.
Same here. Apparently he really liked AIs, although given his history that's not surprising.
What's it audit?
I presume we're still talking about software. Auditing proves what?
Accountants audit that the books are consistent. It's extremely hard to fake consistency and do fraud, especially when there's 3rd parties involved. The edges of the arithmetic get nailed down, because bank payments were sent and invoices paid. So the internal consistency is checkable and valuable.
But software. Are you proving it does what you want? But you didn't write down what you want. Or maybe you're into spec-driven development, or detailed wikis, or whatnot, so you did. Most people aren't, though, and those docs are already out of date anyway.
I think there's something here, because we're all telling our agents what we want the software to be. So maybe you could audit against that. But this isn't the same as accounting.
Where his argument falls apart for me is: why? Audit is intended to address misaligned incentives. It's a form of review done by an independent party, because the incentives are the same for everyone within an organization or with a vested interest in its survival.
The AI doing self-checks through sub-agents (or outsourcing to other models) is a form of review looking for errors, akin to getting a code review for your PRs. You don't need an outside firm or auditor. You need different incentives (a reviewer whose goal isn't to finish the work, but to find problems) and/or different blind spots (trained differently than the model doing the work, which might be more inclined towards that error).
We generally assume that humans can make mistakes and institute mechanisms to flag and correct those mistakes, but that doesn't mean we take that to the level of auditing every discrete action a human at a given job takes.
Likewise for agents, particularly as/if their error rate becomes lower than humans at the given task. They'll get a level of review tied to the importance of their work. And yes, if that's bookkeeping, then they'll probably need a proper audit, just like the humans directing them.
Pangram flags this 100% AI written.
It reads like it too. I don't fully understand why a sentence like "It's the signature that costs." makes it into machine output (It costs... what? What does it cost?) But here we are.
To me it read like a human was pointedly imitating AI , but with a more antagonistic tone. Maybe he asked it to write the article "in his voice".
It never was about X. It was Y.
It looks like there is a large error in reasoning - in all of the historic patterns he is giving, the verification is not cheap. Rather what drops in price is the creation, but verification costs real effort, hence the rise of trusted third parties.
The argument seems like a more appropriate analogy for the slopweb than the dynamics of LLMs themselves.
Didn’t Cringely pass away recently?
According to a HN post, and basically nothing else that I can seem to find.
Who knows, maybe he's still alive and posted the death notice himself. Dude is/was a compulsive liar. There's still to this day no actual evidence that he genuinely has passed away (unless I'm just unable to find it)
This post is also dated in July, which would have been before he passed away if that post was accurate.
This is supposedly his obituary:
https://www.legacy.com/us/obituaries/name/mark-stephens-obit...