Google Bard introduces "Human reviewers," sparking privacy concerns

techstartups.com

35 points·_xivi··10 comments

I am far more concerned about AI ingesting data about myself and others than I am about humans reading it. Humans have limited attention, memory, organizational ability, interest and processing power. Supercomputers with AI remember everything and can find patterns to exploit that no human ever could.

Nothing to see here, move on , move on.

Hasn't Bard told you that your conversations may be reviewed and to not enter sensitive information since the dawn of Bard?

I believe so, and also that ChatGPT has similar terms of use. (Though in ChatGPT's case, they specifically note if you run it in "history off" mode there won't be any review.)

Bard has "history off" too.

"If you turn off Bard Activity, future conversations won’t be sent for human review or used to improve our generative machine-learning models."

It's journalistic malpractice to say "sparking privacy concerns" when it is in fact just the author's opinion. There is no one cited as being concerned. Just that the author thinks the "move" is "likely to spark privacy concerns".

You'll note that there's also no interviewing, no requested comments, and no research besides summarizing widely publicized information from one corporate web page. Holding it to any standard of journalism is a stretch.

Nobody calls it out anymore, but this is a blog post.

Even a good blog post would include research with multiple sources. This is something else. A hit piece, maybe, or a smear campaign.

Yeah, it's just spam.

This is pretty common in media, much more noticeable in a lot of political contexts where the author cites something like “concern from the public” when we really it’s just the author and her pals.

Reply on news.ycombinator.com