Data Theft: Community Slams AI's Foundation Built on Uncompensated Scraping
The core conflict centers on generative AI's reliance on massive datasets allegedly gathered without creators' consent. This perceived data harvesting is framed by many as systematic intellectual theft.
The debate over platform moderation is equally heated. Some users, like 'allo', advocate for 'SpeedCleansing'—a detailed campaign of posting, banning, and re-posting to purge unwanted content. Conversely, 'MigratingApe' argues this proactive banning is an overreach, violating principles of cooperation. 'PhilipTheBucket' drilled down on the moderation spiral, claiming banning people for perceived future infractions is just a self-sustaining harassment tactic.
The weight of opinion shows two distinct camps: one condemning the technology's ethically tainted origins via unconsented data, and the other battling moderation overreach. The primary fault line remains the nature of power: who controls the data, and who controls the platform's enforcement mechanisms.
Key Points
AI training data collection constitutes intellectual theft.
Multiple users, including 'WonderfulWanderer', assert the technology is fundamentally built upon an immoral act of theft because data was hoarded without consent.
Proactive moderation bans are an abuse of power.
'MigratingApe' contends that banning users suspected of being 'Anti-AI Trolls' without specific misconduct violates cooperative principles.
Moderators should actively 'cleanse' negative content.
'allo' detailed 'SpeedCleansing': repeatedly posting, banning downvoters, and deleting content to reach a pure state.
Enforcing bans based on potential future misconduct is logically flawed.
'PhilipTheBucket' compared this system to a 'stop and frisk,' calling it a disproportionate harassment campaign.
Resistance to AI is futile and energy-wasting.
'GardenVarietyAnxiety' suggests accepting AI as a permanent economic feature, while others propose technical countermeasures like 'poisoned datasets' ('glowing_hans').
Source Discussions (3)
This report was synthesized from the following Lemmy discussions, ranked by community score.