Section Five - Fakers
A story about pizzagate, deepfakes, and a dancing machine (17 minutes)
Fake News is a hot topic. There are a lot of people that are making fake news. Sometimes in an organized professional way, sometimes just for fun. In this crash course we look at four reasons that fake news is a large problem.
(1) Everybody fake now!
It is pretty easy to create fake news (anyone can do it!) and it is pretty hard to determine wether something is fake or not. There are different types of fake news.
- Misinformation, which can be for example sloppy journalism. The intent is not to deceive.
- Disinformation, which is false information that is deliberately created to influence public opinion or obscure the truth. Most of the time these are made up stories. These made up stories can have serious consequences. A great example is Pizzagate, a fake news story that led to real gunfire.
Watch this short video on Pizzagate (2 minutes):
Pizzagate is a classic example of a conspiracy theory. Let's see what this conspiracy theory is all about.
Proponents of the Pizzagate conspiracy theory claimed they found emails that contained coded messages that connected several high-ranking Democratic Party officials and U.S. restaurants with an alleged human trafficking and child sex ring. One of the establishments allegedly involved was the Comet Ping Pong pizzeria in Washington, D.C. There was something sinister happening in the basement!
Absurd? Maybe, but there are even more bizarre theories going around. A few examples:
- There is an elite who abuse children and drink their blood (QAnon);
- Small nanobots will be injected during the vaccination against the coronavirus;
- Kim Jong Un is a CIA puppet;
- 5G causes the coronavirus;
- And so on.
You would think that it is pretty easy to debunk these conspiracy theories. Just check this conspiracy theory debunker from Kurzgesagt (2 minutes):
Yet conspiracy theories are read, viewed and shared hundreds of millions of times. Sometimes it seems that the rule applies: the more incredible the story, the better! There are two possible reasons for this:
- News is entertainment nowadays and fake news is the best way of entertainment;
- By 'believing' in a ridiculous theory you show your loyalty to a group or a movement.
(2) Creating chaos
There are a lot of reasons that people want to spread disinformation. First of all it is a great way to attract attention. If you want clicks, likes, followers, retweets and so on, then fake news often is the way to go. Secondly, spreading fake news helps to create chaos. It is an easy and cheap way to influence other people, organisations or countries.
Foreign powers are regularly accused of spreading fake news. After all, it is not complicated. A hundred people in an apartment building somewhere in the suburbs of Moscow can flood an entire country with fake news. The platforms on which that happens will immediately begin to amplify the effect.
How that works is explained below.
(3) Social media platforms
Imagine, 15 years ago, you planned to spread a conspiracy theory. You were sure that Mark Rutte, Prime Minister of the Netherlands, filled his pizzas with children's blood somewhere in a cellar. To get this story out you had to work hard. You had to print a newspaper or distribute a pamphlet or try to get on TV or stand with a banner somewhere.
You probably would not have succeeded.
Now all you have to do is click, tap and swipe and you're done! Chances are that this is reinforced by the algorithms of the large social media platforms. We have already seen how this works in crash course two.
If you enrage them, you engage them!
There has been a lot of research that shows that fake news travels way faster than true stories. This can be explained by the fact that fake news stories are often spectacular and they arouse emotions and interest, so people are more likely to click on them. Clicking leads to more traffic, which leads to more money for the platforms.
Foreign powers did not have to hack Facebook and Twitter. They just weaponized the algorithms of Facebook and Twitter. Of course, some people would argue that this is also fake news.
(4) Deepfakes
Finally, it is hard to spot fake news. Social media platforms try to build algorithms that can spot fake news, but that is very complicated. Is it fake news or is it satirical news ? Is it fake news or is it a parody? There are a lot of stories in which the algorithms got it totally wrong.
As a human being it is also hard to spot fake news. Of course, bizarre conspiracy theories are easy to spot, but often fake news moves in more subtle ways. Look at the headers below, what do you think?
- "California Governor to Relocate Veterans Cemetery to Make Way for Affordable Housing." Real or Fake?
- "New Species of Deadly Spider Kills Five in U.S." Real or Fake?
- "Firefighters Forced to Buy More Expensive Data Package During Wildfire." Real or Fake?
Pretty hard, right?
And it will become harder in the near future, because fake news will get better and easier to create. People creating fake news are a special kind of abuser of technology. They use technology, like social media that was designed for connecting people and selling advertisements, for spreading all kinds of opinions and theories.
These people are helped by a development called generative adversarial networks/deepfakes.
The story of the deepfake starts in 2015. In that year Google gave the world a powerful open source artificial intelligence tool called TensorFlow. The idea was that people used this software to build technology that could have an impact as profound as electricity as Google put it. With TensorFlow everyone could build something that would contribute to a better society. A lot of good came from that move, like people building cancer detection algorithms, but the most famous piece of software was built by an anonymous Reddit – user with the username "deepfakes."
It was, of course, porn!
This person created software that automatically stitches any image of a face (nearly) seamlessly into a video. Deepfakes used this to face-swap celebrity faces onto porn performers. Vice Magazine concluded when they noticed this: we are truly fucked! Today, this technology has become better and better and easier to use. You do not need a multimillion-dollar studio to create these videos, you can just download the software or an app.
The results are impressive.
Watch this (2 minutes):
Or this (2 minutes):
Exercise
How to spot fake news? There are all kinds of websites that help you spot fake news? But which websites are real and which are fake? Which tips are useful and which are not. Do some research and list the six best tips using this PowerPoint template (CC8_Bad Actors_Exercise.pptx).
You can download the answer here (CC8_Bad Actors_Exercise_Answered.pptx). Did you find the most important tip?
Generative adversarial networks
The underlying technology of deepfakes is something called Generative Adverserial Networks (GANs). These are artificial intelligence networks that can create new ‘things.' Two neural networks compete with each other, based on a data training set, and the result is something new.
Computers with imagination.
For example, music, faces, images, designs, and so on. A great example is: thispersondoesnotexist.com a website that generates faces of people that do not exist. The software uses the input of millions of pictures to imagine totally new people. Every time you press refresh (F5) a new face of someone that does not exist appears. You can imagine how people can use this to generate fake profiles, for example, in a dating app. You can even use GANs to fake that you can dance.
Watch this amazing video (3 minutes):
This technology is only getting better and better. This will have serious consequences. Some examples:
- Comic artists no longer have to draw backgrounds;
- Low-level models are simply generated;
- Architects can have simple work done by computers;
- Fashion designers too;
- Low-skilled jobs may be under pressure.
But maybe the most important thing is that in the future everything you see on a screen cannot be trusted.
Quick Tip: If you have some money lying around, invest in a theatre. If nothing can be trusted anymore on the screen, then that will lead to an upgrade of the real.
The question: "Is it fake or not?" will be harder to answer. Ironically, this also could mean the end of fake news. After all, if everything can be fake and you can no longer see the difference, and there is no way to check it, maybe people will start looking for news that is true.
The term fake news will be replaced with True News.
Or maybe not.
If you are thinking about, assessing, designing, programming or using a technology then maybe fake news is not always relevant. This applies only to certain technologies. However, when you are researching the web, trying to find out more about a certain technology, it is always useful to be aware of fake news.
And it never hurts for your own development and peace of mind. Especially when reality is an ever-changing concept (the topic of our final section - six).
Take aways from section five:
- Fakers are a special kind of abuser;
- Fake News is easy to make;
- Fake News is amplified by social media;
- Fake News is hard to spot and that only becomes harder;
- Which could be good news.
Some final words on crash course eight
Congratulations! You have completed this crash course, so you got a very small taste of thinking about technology and the potential impact of bad actors. This was just an appetizer. If you are going to think about, assess, design, program, discuss, use or invent a technology we would like you to remember that:
- Most technologies will be abused;
- This can be by breaking and entering (hackers);
- Or using the technology in a different way than it was intended (abusers);
- Because our world is digital, this can have very large negative consequences for the users;
- Also it is important to be aware of fakers.
And remember: when it comes to fake news, always trust your feelings.
Want more?
This was a crash course. It only took you one hour to complete. If you want more, we have some suggestions:
- First, you can check out section six, with all kinds of additional materials. Section six is updated regularly;
- Second, you can do one of the other ten crash courses;
- Third, you can start using the Technology Impact Cycle Tool, especially the questions regarding bad actors;
- Fourth, you can check out our example cases, for example bad actors and the Corona Contact Tracing App.
Or you can do the extra section -> the reality check!
Finally, do you have any suggestions or remarks on this course? Let us know at info@tict.io