Malicious 'Deep fake' videos will be illegal in China

[jandl100]

Clearly intended to constrain anti-Chinese government propoganda, it nonetheless sounds like a good idea to me that everyone should adopt.
Deep Fake tech is deeply dangerous imo.

'The Chinese government just made it illegal to pass off deepfakes or other AI-created media as legitimate.'
'... anyone who uses artificial intelligence or virtual reality tech to synthesize realistic media will have to disclose that their creations aren’t real life.'

futurism.com/the-byte/china-makes-deepfakes-illegal?utm_medium=email&utm_term=0_03cd0a26cd-187a84e8dc-246888077&mc_cid=187a84e8dc&mc_eid=092f9d027b&utm_source=The%20Future%20Is&utm_campaign=187a84e8dc-EMAIL_CAMPAIGN_2019_12_02_07_02

Interestingly, that article also states that California has already passed a similar law.
' China’s new law bears similarity to a California law enacted in October.
California doesn’t specifically mention deepfake tech by name — instead, it makes it illegal to publish misleading media that could damage a candidate’s reputation or deceive voters.'

That seems to take the concept a stage further, and lying politicians like our beloved Bojo and Gove [insert your own examples, driven by your personal political prejudices] would need to beware!
All countries should do the same imo.

Let the bunfight about free speech begin. :D

[Firebottle]

It's not exactly free speech when they are peddling lies and misinformation though is it?

[Alex_UK]

What’s interesting is that a whole new segment of AI has sprung up, trying to spot the fake AI! Deep Fakes or whatever you want to call them are seriously scary though - a bit of googling will soon have you in disbelief if you’ve not experienced them.

To me it’s fraud however way you look at it, so definitely needs to be equally as illegal.

[Landloper]

Another concern is that Deep Fakery has an obvious application in discrediting political dissidents, or by helping fabricate 'evidence' against an individual the security services wish to 'fit up'. Of course the state would seek to ban others using such tech while itself enjoying a covert monopoly.

[Light Dependant Resistor]

Another concern is that Deep Fakery has an obvious application in discrediting political dissidents, or by helping fabricate 'evidence' against an individual the security services wish to 'fit up'. Of course the state would seek to ban others using such tech while itself enjoying a covert monopoly.

And certain political figures just deny having met Prince Andrew, despite overwhelming photographic evidence to the contrary https://www.abc.net.au/news/2019-12-04/donald-trump-says-he-does-not-know-prince-andrew/11764228 Even the Chinese would enjoy the humour of this story.

[Landloper]

The Wikipedia page on Deepfake includes the following quotation on the dangers of the technique:

'The primary pitfall is that humanity could fall into an age in which it can no longer be determined whether a medium's content corresponds to the truth...'

The same entry suggests that an antidote would be to trust journalists and the information they present. I'd say the 'antidote' is simply an earlier iteration of the very condition it is proposed to treat.

[walpurgis]

'The primary pitfall is that humanity could fall into an age in which it can no longer be determined whether a medium's content corresponds to the truth...' .

We are already there and have been for a long time.

[Landloper]

We are already there and have been for a long time.

Agreed. The difference is one of degree rather than one of principle.