Sarah Silverman and novelists sue ChatGPT maker OpenAI for ingesting their books

Sarah Silverman and novelists sue ChatGPT maker OpenAI for ingesting their books

After 2001 most people accepted the ingestion of almost all private communication by NSA surveillance.

In a few years, people may need to accept the ingestion of the rocky planets of the inner solar system and asteroids by Artificial Super Intelligence to power massive computronium arrays.

I think OpenAI has an excellent chance of winning this. If the judge or jury has ingested anything Sarah Silverman has produced, it's a sure bet they regurgitated it short order before it contaminated their innards.

Google's digitization project was pretty cool. It could be described as a contribution to advancement of mankind by making obscure material available globally.

Of all the things to ingest for the benefit of humanity and future generations and it’s a Sarah Silverman book?

hope they lose and have to delete it out of existence.

I came here expecting to say things put in public are public and too bad.

However, copyrighted work must be honored and the legal aspect of those copyrights upheld for the expected periods of time.

Now, if Congress wants to change the time that copyright protections are valid, making them shorter, I'd be fine with that. I'd also approve an annual, increasing, fee be required to retain copyright protections after XX years, perhaps 20 or 30. So the works which aren't worth those fees become public domain much faster than current laws support.

I could also support vastly different laws for individual copyright holders as compared to corporate holders of copyright works. An individual author could have their work protected for 40 yrs, while a corporate owner would only get 20. After that period, the fees to retain protections would be the same and on a sliding scale. For example, a little pamphlet that only 5000 people saw doesn't deserve the same, free, protections as a $100M movie that supported 20K jobs. The movie company would likely pay for protection for the first 10 yrs, after they expire. Each 10 yrs, add a zero to the fee, so only the largest, profit making, copyrighted works would be protected after 40-50 yrs.

Then AI could read all the public domain stuff and learn it. That won't prevent AI hallucinations - which is a big issue with current AI today. Basically, the AI decides to claim, with authority, that something is true, when it isn't.

That would go for all copyright works - books, magazines, software. Put them all into the public domain faster, but after a reasonable time for people to profit from those works.

Of all the things to ingest for the benefit of humanity and future generations and it’s a Sarah Silverman book? 

hope they lose and have to delete it out of existence.

Can’t believe I’m saying this, but: I agree with blacklabel