3 ms·

This part of the story stood out for me: >More emails arrived in my inbox. >“When will you update the article?” one was titled. The email had no text content,

This part of the story stood out for me: >More emails arrived in my inbox. >“When will you update the article?” one was titled. The email had no text content, only an image — a screenshot of my initial interaction with Daniel. >Except it did not show my actual response to Daniel, but a fabricated message that I had not written. >“Hi Daniel, Thank you for noticing, I checked with the IDF Spokesperson and it was indeed intercepted. I sent it now for editing, it will be fixed shortly,” I supposedly wrote. (To be clear, I wrote no such thing.) this seems to be a main issue. Would it help journalists if emails were quotable by default and the first party email providers could verify specific quotations? This way this class of fraud, market manipulation, and fake news would disappear. I don't see why people wouldn't leave their responses as quotable when responding to journalists, for example, and journalists could also set their responses as quotable by default. What do you think, could this help this issue?

tsimionescu 7mo ago
There is nothing that can prevent doctored screenshots of UIs from being disseminated. The only defense is not considering screenshots from random sources even evidence of anything.
- logicallee 7mo ago
  >There is nothing that can prevent doctored screenshots of UIs from being disseminated. Thanks for thinking it through. To combat this, the interface could show the attestation id prominently so that it's obviously wrong or missing something if it doesn't contain it. Something like this mockup[1] and be easily linked so 1) it would be super suspicious for someone to quote someone without linking to the attestation - and 2) people could type the attestation ID from a screenshot into the server themselves. This would require the attesting server to keep a copy of sent email but that doesn't seem too bad. If it's been deleted, you're not willing to be quoted on it anymore. It would be voluntary. What do you think of this idea overall? I realize that the mockup seems fake - the whole point is that it's obviously fake, because it's being hosted elsewhere and not where it says it is. [1] https://claude.ai/public/artifacts/42c38630-2728-4953-b453-a211e8bc794e https://claude.ai/public/artifacts/42c38630-2728-4953-b453-a...
yifanl 7mo ago
The problem here is that people have been provided strong incentive to believe in a falsehood. It is unlikely a purely technological solution is available for this.

tsimionescu 7mo ago

There is nothing that can prevent doctored screenshots of UIs from being disseminated. The only defense is not considering screenshots from random sources even evidence of anything.

logicallee 7mo ago
>There is nothing that can prevent doctored screenshots of UIs from being disseminated. Thanks for thinking it through. To combat this, the interface could show the attestation id prominently so that it's obviously wrong or missing something if it doesn't contain it. Something like this mockup[1] and be easily linked so 1) it would be super suspicious for someone to quote someone without linking to the attestation - and 2) people could type the attestation ID from a screenshot into the server themselves. This would require the attesting server to keep a copy of sent email but that doesn't seem too bad. If it's been deleted, you're not willing to be quoted on it anymore. It would be voluntary. What do you think of this idea overall? I realize that the mockup seems fake - the whole point is that it's obviously fake, because it's being hosted elsewhere and not where it says it is. [1] https://claude.ai/public/artifacts/42c38630-2728-4953-b453-a211e8bc794e https://claude.ai/public/artifacts/42c38630-2728-4953-b453-a...

yifanl 7mo ago

The problem here is that people have been provided strong incentive to believe in a falsehood. It is unlikely a purely technological solution is available for this.