Copy
Share
Tweet
Forward
Share
Fake Regulations

"Deepfakes — video of things that were never done with audio of things that were never said — can be tailor-made to drive Americans apart and pour gasoline on just about any culture war fire."
-Sen. Ben Sasse (R-Neb.)
Altering video footage has been possible for decades, but doing so has historically taken time, professional skills and a lot of money. Developments in areas like AI have now made it possible for anyone to cheaply create a convincing fake video, including those who might use it for malicious purposes. This reality, and some high-profile news stories that we've covered in this newsletter, have brought deepfakes to the forefront of political conversations — and the latest debate surrounds what to do about them. 

The obvious answer: create technology that can automatically detect forgeries. Unfortunately, this is still well out of reach. And until the tech exists, legislation may be the last line of defense. New laws, or lack thereof, will have potential implications, and are something we should all keep an eye on.

Lawmakers including Sen. Ben Sasse (R-Neb.), Sen. Mark Warner (D-Va.) and House Intelligence Chairman Adam Schiff (D-Calif.) have all looked into deepfake legislation, with Sasse even introducing a bill to criminalize the malicious creation and distribution of deepfakes. Their main argument is that deepfakes pose a national security risk and with the right fake threats, could even lead to war. Not to mention they can hurt the reputations of business and government leaders, celebrities and everyday people. The proposed legislation includes fines or jail time for individual deepfake creators, and penalties for distributors like Facebook — but only if they know they're distributing a deepfake. 

Those who oppose legislation including law professors Danielle Citron and Mary Anne Franks think regulation could do more harm than good. They worry legislation could scare platforms into taking down everything that's reported as a deepfake — likely deleting legitimate posts in the process. Others like David Greene, civil liberties director at the Electronic Frontier Foundation, believe civil liberties could be at risk. He worries that making malicious deepfakes a federal crime could endanger protected speech like parody videos. People often use deepfake technology to create comedy like the above video, but the lines between harmless fun and harmful intent are often blurry. 

Until a deepfake detector is proven possible (check out the people working on it today's Deep Take), regulation will be the only viable option. Given the gravity and potential impact on communicators, this is something we should all be watching closely. 
 
Chris Perry 
@cperry248

 
If you find this newsletter valuable, I would be grateful if you encouraged others to sign up. You can do so by directing them here
 
What We're Reading
Platform Dominance
Private Messages are the New (old) Social Network
By Lauren Goode, Wired

Facebook plans to unify Facebook Messenger, WhatsApp and Instagram messaging on the backend of the services. This would make it possible for people relying on different flavors of Facebook apps to all gorge at the same messaging table. This move makes sense as privacy becomes an increasingly important consideration for those on social media. 

Media Forensics 
Who Was Lil Tay?
By Lauren Levy, The Cut

There are two kinds of social media influencers — those with lives almost too perfect to believe, and those who become living, breathing memes. Lil Tay, a 9-year-old girl who's toilet costs "more than your momma’s rent," is the latter. The persona's volatile rise and demise was the result of a 16-year-old brother's obsession with the internet and its star-making power. 

Platform Dominance
Home is Where the Photo Booth is: How Instagram Is Changing our Living Spaces
By Alyssa Bereznak, The Ringer

Dedicated Instagram walls are places where guests can take photos good enough to graduate from the Instagram Stories feed to a post on their permanent grids. They serve as an ice breaker at social events and encourage organic promotion. These spaces are part of a larger trend of Instagram playgrounds that exist to replace experiences with photo ops altogether.

Media Forensics
Alexandria Ocasio-Cortez, Video Games, and the New Online Town Square
By Kyle Orland, Ars Technica 

Savvy politicians have always gone where the people are, which can sometimes be in the most unlikely places. Congresswoman Alexandra Ocasio-Cortez recently took to a Twitch livestream of Donkey Kong 64 to address the kinds of political issues that are important to her. 

Content Experience 
With Interactive TV, Every Viewer is a Showrunner Now
By Peter Rubin, Wired

When it comes to choosing a TV series, choice is everywhere. Agency... not so much. Algorithms forever recommend what to watch. And autoplay functions cue up the next episode without waiting for your input. But Netflix is handing back the reins via choose-your-own-adventure experiences it’s calling “interactive content.”

Media Forensics
Slime Videos Drove 25 Billion Views on YouTube Last Year
By Carla Marshall, Tubular Insights

In 2018, slime content generated 25.4B views on YouTube. Slime influencers generated the overwhelming majority of views, with only 140 sponsored videos last year. This leaves a huge opportunity for brands to partner with publishers and influencers to reach a more targeted audience. 

Artificial Intelligence
Gradually, Then Suddenly
By Tim O'Reilly, LinkedIn

There’s a passage in Ernest Hemingway’s novel The Sun Also Rises in which a character named Mike is asked how he went bankrupt. “Two ways,” he answers. “Gradually, then suddenly.” Technological change happens in much the same way. Small changes accumulate, and suddenly the world is a different place.

Emerging Technology
This AI Generates Images of Food Just by Reading The Recipe
By Jesus Diaz, Fast Company

New AI algorithms can generate fake photos of food from text recipes that list the ingredients and the method of preparation but don’t contain any visual description of how the final plate looks. The AI isn't allowed to read the title of the recipe to generate the image, it exclusively uses the ingredients and the instructions. This ability demonstrates a capacity for abstraction that we’ve assumed computers don’t have.

Deep Take
When Seeing is No Longer Believing
By CNN Business
Advances in artificial intelligence could soon make creating convincing fake audio and video – known as deepfakes – relatively easy. Making a person appear to say or do something they did not has the potential to take the war of disinformation to a whole new level. This CNN guide provides a comprehensive look at deepfakes and what the US government is doing to combat them.
Copyright © 2019 Weber Shandwick



Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.

Email Marketing Powered by Mailchimp