Hi,
On 17/09/24 11:04, Edward K. Ream wrote:
> I should emphasize that all these AI tools should be valuable as
> starting points.
[...]
>
> The same strategy should work with a blog about Leo or LeoJS. Revise
> the results until they pass expert judgement. Revise the settings to
> eliminate any hint of sexism.
Agreed. The main problem I see with AI is when is unchecked, in several
senses: the information it produces, the expropriation of value from
creators who gave unwillingly training data, the huge ecological cost
and waste behind it, the race, gender and other bias it confirms, among
a long list of problems. The critical approaches from initiatives
feminist[1] and decolonial[2] perspectives can help us to keep AI, its
hype/dangers and uncritical uses at bay.
BTW, the Leo podcast is pretty impressive and hopefully we will go to
other AIs, like Engelbart's Augmented Intelligence, underlining the
human and communities role in the computational labor, data production
and curation, hopefully for the benefit of everyone instead of the
enrichment of few (as usual).
[1]
https://feministai.pubpub.org/
[2]
https://botpopuli.net/artificial-intelligence-and-the-feminist-decolonial-imagination/
Cheers,
Offray