What would make AI on iNaturalist tolerable to you?

[nathantaylor]

Here are some of the suggestions so far. If yours didn’t make it, you might consider rereading the original post. :-)

1. Attribution

@nathantaylor

The AI says that the information was aggregated (paraphrased) from comments by [users]. I would rather this than a misquote, though adding quotation marks around direct quotes would be ideal.

2. Disclaimer

@nathantaylor

A statement provided with the AI summary that distinctly says that information may be misquoted or paraphrased incorrect.

@ShotShot

And roughly similar to your ‘2.’, a big, clear statement/disclaimer that the machine-generated content must not be considered authoritative nor sound.

3. Source accessibility

@nathantaylor

Links to the original posts that had the original wording and the context of that wording. This may need to be accompanied by duplication detection so that only unique comments are linked to. Even better is if each word or phrase was either hyperlinked to the original comment or hovering over the text showed a popup of the full original comment.

4. Correction/user input

@nathantaylor

A flag option to indicate when the AI is misquoting a user. This would allow further improvements to the AI to help guide it to quoting the source correctly. Alternatively, this could also be used to stop the AI from quoting the particular comment entirely if all instances of the use of the comment are misquoted in a severe enough way.

@natev

Curators or even users should have the option to hide LLM responses until further review, not just flag them.

@natev

Nice to have: some way to “promote” comments that the AI can then draw from (though at that point, a non-AI solution may be simpler).

@jrcagle

• The algorithm should include feedback from human identifiers. It needs to notice when it’s getting voted down.
• In connection to this, I would suggest an “upvote/downvote” button that is specific to the text. It is entirely possible that the AI model could get the right ID but for the wrong reasons - in which case the text would mislead.

@deboas

ability to immediately correct or disable erroneous information that might be generated.

5. User choice

@ShotShot

An option to opt-out, so that a username and their contributions can in no way be fed, processed, or linked to the above-mentioned AI. (As a bonus, a way to hide users according to one’s own taste – including those users who opt-in this AI).

@natev

Some way for users to opt out of their comments being used this way, or an opt-in system (the latter would probably cripple the whole project though TBH)

6. AI quality

@natev

Multiple advanced anti-hallucination techniques should be implemented.

7. User notifications

@natev

Nice to have: if the system is opt-out, iNaturalist could send a notification to any users referenced by the LLM with a link to the relevant page.

8. Overall system function

@ekmes

AI that looks for comments on relevant observations

@ekmes

it compiles a list of links to these observations, presenting them as “for discussions on how to differentiate

9. Transparency, communication, and community involvement

@jrcagle

If I’m understanding correctly, your proposal is not changing the way that IDs are made, but the way that they are explained. THAT NEEDS TO BE MADE CLEAR to the iNat community.

@deboas

a robust discussion and testing of any new tool before full release. The community must be largely happy to see it implemented, which could be assessed through a poll, the tenor of forum discussions, or other means.

@deboas

transparency about the energy use involved in training and providing the tool.