DeepFake special FX are more and more frightening

It’s been a while since we know how to reproduce somebody’s voice with a speech synthesis software1, in order to make him say things he didn’t say… But now on we can also create or modify videos in that way (in real time), so we can see the guy telling things or doing things… he never told or did! And it’s impressive how the result is realistic for the human eye. In the domain of Deep Video Portraits, Michael Zollhöfer is very implicated in the current innovations, as he is, for example, a member of the HeadOn project.

We found a number of videos on this rather disturbing topic… Thus we share them with you:

Deepfake Videos Are Getting too Good :

‘HeadOn’, An AI That Transfers Torso, Head Motion, Face Expression And Eye Gaze :

Generate a new video of a completely different person’s body performing those actions :

FRENCH / Les « deepfakes », savant mélange de « deep learning » et de « fake news » :

Bonus / Age manipulation by video transformation (high level!) by Rousselos Aravantinos:

Notes (FR)

1■ Article du journal « Le Monde » du 02/04/2017 « L’appli qui imite les voix »

Elon Musk wants to create a website to note the journalists and the media


Some billionaires want to buy media. Elon Musk wants to rate journalists’ credibility.


“Going to create a site where the public can rate the core truth of any article & track the credibility score over time of each journalist, editor & publication,” Musk tweeted.


“The last thing we need is another rich and powerful dude threatening to silence any journalist who doesn’t see things his way,” Timothy Karr, senior director of strategy at advocacy group Free Press, told CNN. “The implication in Musk’s actions are that all news media are untrustworthy. That’s a shameful message to be spreading.”

Could be true. But we’d like this Timothy Karr to say the same about Soros, who is much more Dangerous. Curiously, we think that he wouldn’t.

Musk’s idea is rather a good one. More and more people realize that Press writes less and less Truth, and behaves like a political censorship. We have the same problem in Europe.


To build up false memories

If you want our bot to READ this article for you, click play:


We can make remember somebody of something he never lived.


We saw in another article that we can physically implant a false memory in brains of mice1. But it is also possible to suggest good enough a false memory so the subject believes that he lived what he is told and is even able to remember details that were not even told to him.


You know, with a little of persuasion, you certainly managed already to make believe your colleague Michael that John won’t get back to work anymore because he had decided to manage a dairy with an old acquaintance, 500 miles away of the office. And when John came back to the office one week later at the end of his holidays, he did not understand why Michael asked him what went wrong with the dairy … And all the colleagues had a good laugh.

Well, in the same order of idea, researchers have managed to persuade subjects that they had lived certain things that they not lived in fact!

During an experiment, Stephen Lindsay and his team managed to persuade 50 % of their guinea pigs that they had made a flight in hot-air balloon in their youth, what was false, by presenting them forged photographs2.

So that the transplant has a chance to be a success, three factors must be respected: the false memory must be plausible, the subject has to build itself a mental representation of it and the memory should not seem to be produced by a trance at the time of its construction, but by the recollection of a reliable information3.

Other factors are facilitators, for example the emotional implication of the subject in the false memory. If the false memory doesn’t make taking place particular emotion, it has only few chances to be retained by the subject. On the other hand, if it calls on to feelings to the subject, it has more chances to be accepted by this one, and that’s particularly true if the feelings called by the memory are negative: then the false memory will be rather precise.

In 2008, Stephen Porter and his team made an experiment illustrating this phenomenon4: they incite a group of subjects to remember passed public events, some of them being false ones. The conclusion of the study is that the false positive or negative events are more easily accepted than those having no “emotional particular color”, and that those having a negative connotation generate more precise false memories to the subjects than those having a positive connotation. Stephen Porter’s theory to explain this phenomenon is an evolutionary one: according to him, it is crucial for the survival to remember negative events, no matter what the individual lived them actually or that they are reported to him by a reliable source. Therefore, the brain is more inclined to produce false precise memories for events of this nature5.


The conclusion of all this is that it is necessary to be wary of your own bad memories: they are actually the least reliable ones, because they are the ones that we are the most inclined to forge, if circumstances are gathered! And it is necessary to be aware that, as our memories influence our reading of the present, we need to wonder about the reasons which urge us to perceive as “viscerally negatives” certain events of our everyday life and, in a more general way, the current events reported in newspapers… Hopefully this article leaves you a long-lasting memory, see you!


Look further / Useful link(s)


 1■ Article « False memories in your brain »

2■ Lindsay, Ha gen, Read, Wayde, Garry, « True photographs and false memories », Psychological science, vol.15, n°3, 2004

3■ Website

4■ Porter, S., Taylor, K., & ten Brinke, « Memory for media : Investigation of false memories for negatively and positively charged public events », Memory, vol.16, n°6, 2008

5■ Website PsychoTémoins, Inist (CNRS)

Back on top of the article

False memories in your brain

Did you know that scientists managed to implant forged memories in brains using electrodes? Unbelievable, isn’t it? But very true in fact…

I said that’s very true, and in fact it is! It’s just that hopefully, scientists only managed to do this to mice brains for the time being. But that’s not really comforting to know that this backdoor on the mind is now open, so that we can manipulate complex life forms memories. We’ll talk about that in two minutes of time.

The experiment consists in associating two different memories in mice brains to create a new one. To make it happen, the Professor Susumu Tonegawa research team, from MIT, placed mice in cages and analyzed the zones of their brain that were bustling as they discovered their new environment.

The next day, they took the mice and placed them in another environment, they reactivate their neurons that were bustling the day before when they discovered their cage, and they administered to them a slight electric shock, not dangerous for their lives but not very pleasant for them too.

Another day later, the researchers replaced the mice in their original cages and they could observe that as the mice were recognizing the place, they showed a frightened attitude because of their memory of the electric shock they experienced the day before and that their brains wrongly associated with this cage. Actually, in the real world, the mice never get electrified in this cage.

All that is fine (if we can say this…), but you have to tell me here that this experiment has to deal somewhere with a practical issue. Actually, how do the researchers to reactivate the memory zone of the mice brains used when they discovered their cage on the first day of the experiment? Well, they dealt with a technic called opto-genetics: the mice used were genetically modified ones, so that their neurons were made light sensitive. Then the researchers could reactivate the mice neurons by exposing them to a light source.

This might look as a little achievement only: the mice were not natural ones, researchers needed to use sophisticated material to lighten specific neurons into the mice bodies… Indeed, genetic manipulation is already kind of forging memories when you think of it…

But you know, this experiment was made in 2013, and it just opened the breach. Many others took place until now, and not only others, but more effective ones!

In 2015, researchers from CNRS2 managed to implant forged memories into mice brains, using electrodes… No more genetic manipulation and no more invasive technologies to achieve the same result! Their article has been published in the review “Nature Neuroscience”, n°18 p.493-495, 20153.

The protocol of the experiment was the following one: while the mice were sleeping, the researchers activated their cerebral zones in charge of the reward system as other neurons in charge of geolocalization stuff.

Doing so, the researchers simulated the presence in a located place and an amount of pleasure in the mice brains. The next day, the researchers could observe that the manipulated mice get specifically on the place where they had synthetized pleasure memories.

The NeuroHack team likes to show you the two faces of the discoveries: the good one… and the ugly one.

Good first.
By applying such a technology on man (what we are unable to do for the time being), we could improve our understanding of the memory device and hope to:
1 – heal mental disorders linked with neurodegenerative diseases
2 – improve the reliability of testimonies in legal cases
3 – managed better the post-traumatic emotional shocks

And then the ugly.

According to sci-fi works talking about this subject, (Inception, Total Recall, Blade Runner…) this technology could have other consequences. And bad news is that in history, when something becomes possible to man, it finally appears in the reality.

Be aware that modifying someone’s memories means modifying his intimate, very true nature and personality. It means playing with his joys, his sorrows, his fears. Doing so, it becomes possible to make him hate or love someone else…

Long story short, unpleasant philosophical question would somehow enter your mind, such as: Can I trust my childhood memories? Am I sure that I really love the people I love? Or hate those I don’t? Am I sure of what I have done the day before? Or one minute ago? Am I living my real life? And if not: what does my real life look like???

But let’s begin with a simpler question: are you really sure to have read this article?





Back on top of the article