-
EFF just posted this, “California’s A.B. 412: A Bill That Could Crush Startups and Cement A Big Tech AI Monopoly”:
California legislators have begun debating a bill (A.B. 412) that would require AI developers to track and disclose every registered copyrighted work used in AI training. At first glance, this might sound like a reasonable step toward transparency. But it’s an impossible standard that could crush small AI startups and developers while giving big tech firms even more power.
Back in the early 2000s, we wrote SpamAssassin, a machine-learning driven antispam system which was trained on user-submitted data. We tracked the attribution of every item of input used to train that system. We weren’t even a startup, we were an open source project.
If we could do it, why can’t modern AI systems? And don’t say “because the existing large language models didn’t do it” — that’s just accepting past shitty behaviour as a fait accompli.
Extremely disappointed in the current state of the EFF if this is what they think.