News for hackers. Stuff that matters.
Re:AI text generator not released for concerns about implica... by ancarda on Friday February 15, 2019 @02:46AM (#19169204)
>i have realized that from now on i cannot trust 90% of the comments i read on reddit. this is insane.

I hate to be cynical here but I'm glad this has made you realize something that's been true since the Internet started; you shouldn't trust what's written on any forum! Be skeptical.

Re:AI text generator not released for concerns about implica... by chrisco255 on Friday February 15, 2019 @02:53AM (#19169227)
I agree with being skeptical but there is a point where rational skepticism just turns into pure cynicism and that's not productive.

Also, with the advances in deep fakes and synthetic video...how long before you can't trust video evidence either?

Re:AI text generator not released for concerns about implica... by stef25 on Friday February 15, 2019 @05:39AM (#19169861)
Got news for you! It's already possible to produce fairly accurate fake video of people saying things, in their own voice

https://www.youtube.com/watch?v=cQ54GDm1eL0

Re:AI text generator not released for concerns about implica... by ancarda on Friday February 15, 2019 @03:18AM (#19169322)
Hmm, video evidence may be trustworthy - eg. video from a CCTV system. Perhaps it could be written onto some write-once, tamper resistant format? Not sure how that would look.

I suppose one place to start thinking about this would be photos. Are photos admissible evidence or do courts only allow negatives? Photos have been modified for a very long time. This is probably the most famous example: https://amp.businessinsider.com/images/52af668569bedd3b2643d...

But yes, I am far more worried about how much more effective fake news will be once they start coming with actual videos.

Re:AI text generator not released for concerns about implica... by ohtwenty on Friday February 15, 2019 @04:04AM (#19169508)
Cctv on blockchain?
Re:AI text generator not released for concerns about implica... by njharman on Friday February 15, 2019 @03:30AM (#19169374)
How long? Several years ago. Videos have been faked forever. There’s been all sorts of optical illusions, forced perspective, and special effects for 100 years.

You shouldn’t trust any single source. Only a preponderance. Even then be open to skeptics

Re:AI text generator not released for concerns about implica... by bronz on Friday February 15, 2019 @03:01AM (#19169253)
i meant that i cant trust whether or not it was written by a human, not whether or not i can trust the correctness of the comment. edited for clarity.
Re:AI text generator not released for concerns about implica... by dwaltrip on Friday February 15, 2019 @03:13AM (#19169305)
Trust and skepticism aren't binary.
Re:AI text generator not released for concerns about implica... by ancarda on Friday February 15, 2019 @03:19AM (#19169328)
Interesting, would you elaborate? Do you perhaps mean that even if you do trust someone, you should still be skeptical?
Re:AI text generator not released for concerns about implica... by dwaltrip on Friday February 15, 2019 @03:31AM (#19169381)
Someone's skepticism, knowledge, and careful assessment might lead them to think that a forum post has a X% of being machine generated (as one example scenario). There are big differences between values of 0.1%, 1%, 10%, 50%, 90%, etc, and the resulting impact of the who are involved in that system.

Because of this, it isn't helpful to say, "Oh you should always be skeptical! It doesn't matter if things have changed significantly such that we have more reason to be skeptical now."

Re:AI text generator not released for concerns about implica... by ancarda on Friday February 15, 2019 @03:35AM (#19169404)
That makes sense, thank you!
Re:AI text generator not released for concerns about implica... by mamcx on Friday February 15, 2019 @08:09AM (#19170572)
"Trust, but verify"