Good. And I think watermarking AI output is also a dead end. Better that we simply assume that all content is fake unless proven otherwise. To the extent that we need trustworthy photos, it seems like a better idea to cryptographically sign images at the hardware level when the photo is taken. Voluntarily watermarking AI content is completely pointless.
OpenAI shuts down its AI Classifier due to poor accuracy
81–90 of 292 posts
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#82Earlier quoted context omitted.
I've also noticed that ChatGPT tends to respond to short prompts, especially questions, in a predictable format. There are a few characteristics. First, it tends to print a five-paragraph essay, with an introduction, three main points, and a conclusion. Second, it signposts really well. Each of the body paragraphs is marked with either a bullet point or a number or something else that says "I'm starting a new point."…
I have to admit I'm struggling to tell if this was done ironically, but your comment is exactly a five paragraph essay with an introduction, three main points, and a conclusion. If so, nice meta-commentary.
Thank you, it was intentional!
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#83I'm glad that they did, although they should obviously done an announcement for it. The amount of people in the ecosystem who thinks it's even possible to detect if something is AI written or not when it's just a couple of sentences is staggering high. And somehow, people in power seems to put their faith in some of these tools that guarantee a certain amount of truthfulness when in reality it's impossible they could…
I see no reason why watermarking can’t be broken by having someone simply rephrase/redraw the output. Yes, it’s still work, but it’s one step removed from having to think up of the original content.
Watermarking was never going to be successful except for the most naive uses.
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#84Earlier quoted context omitted.
Even the idea of it is bad, ChatGPT is supposed to write indistinguishably from a human. The "detector" has extremely little information and the only somewhat reasonable criteria are things like style, where ChatGPT certainly has a particular, but by no means unique writing style. And as it gets better it will (by definition) be better at writing in more varied styles.
Nitpick: ChatGPR is supposed to write in a way that is indistinguishable from a human, to another human. That doesn't mean that it can't be distguishable by some other means.
If a human can't verify whether distinguished text is actually AI or not, detection will be full of false positives and ultimately unreliable.
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#85Good. If it is not reliable, it's a further harm than good if it exists as a false security. An analogous example: my local pizza delivery (where I worked) would shut the box with a safety sticker, to avoid tampering / dipping by the delivery boys. Now, sometimes they would forget to do this for various logistical reasons. Every one of the non-stickered ones started getting returned as customers worried a pepperoni s…
Eh, I'd consider that a failure of employee training and reverse the situation by giving out a weekly bonus to shifts that did not fail to put the security stickers on.
Kinda like if they forgot to put the security seal on your aspirin, I'm not going to take them all off because someone forgot to run production with all the bottles sealed.
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#86Earlier quoted context omitted.
From the article, "while incorrectly labeling the human-written text as AI-written 9% of the time." Seems like from what the article we're talkin about says it definitely ain't worse than random by far. Thing you most want to avoid is wrongly labeling humans as AI-written so that seems pretty good. Though it only identified 26% of AI text as "likely AI-written" that's still better than nothing, and better than random…
You're right, I should have been less specific. If the harm of false positives is significant you may not need to have random or worse than random results to feel obligated to stop the project.
alright. thanks for your thoughts
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#87How could we have both the AI that is indistinguishable from humans and the AI that can detect it with good accuracy? That would imply the race on both sides for an AI that is more intelligent than an high IQ human.
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#88Good. And I think watermarking AI output is also a dead end. Better that we simply assume that all content is fake unless proven otherwise. To the extent that we need trustworthy photos, it seems like a better idea to cryptographically sign images at the hardware level when the photo is taken. Voluntarily watermarking AI content is completely pointless.
I can see that working for specialized equipment like police body cameras, but if every camera manufacturer in the world needs to manage keys and install them securely into their sensors then there will be leaked keys within weeks.
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#89Good. If it is not reliable, it's a further harm than good if it exists as a false security. An analogous example: my local pizza delivery (where I worked) would shut the box with a safety sticker, to avoid tampering / dipping by the delivery boys. Now, sometimes they would forget to do this for various logistical reasons. Every one of the non-stickered ones started getting returned as customers worried a pepperoni s…
It’s a law of nature that pepperoni thieves cannot take a job at a pizza place. They are forever doomed to be delivery guys.
Re: OpenAI shuts down its AI Classifier due to poor accuracy
#90Earlier quoted context omitted.
Even the idea of it is bad, ChatGPT is supposed to write indistinguishably from a human. The "detector" has extremely little information and the only somewhat reasonable criteria are things like style, where ChatGPT certainly has a particular, but by no means unique writing style. And as it gets better it will (by definition) be better at writing in more varied styles.
Nitpick: ChatGPR is supposed to write in a way that is indistinguishable from a human, to another human. That doesn't mean that it can't be distguishable by some other means.
Precisely -- watermarks are an obvious example of this. To me, this is THE path forward for AI content detection.