Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Should I Record (And Publish) my Voice & Face? #91

Open
nelsonic opened this issue Apr 27, 2024 · 0 comments
Open

Should I Record (And Publish) my Voice & Face? #91

nelsonic opened this issue Apr 27, 2024 · 0 comments

Comments

@nelsonic
Copy link
Member

nelsonic commented Apr 27, 2024

I've thought about this a lot over the years and want to capture some of those thoughts for posterity. 💭 📝
In many ways I wish I had just gotten over my fear/procrastination and started recording screencasts years ago.
Then I wouldn't have to be thinking about this ... But here we are and I think it's super important to capture this.

Should You Record Your Voice & Face?

Before we had children, I thought screencasts with voiceover and face were a no-brainer. ✅
A great way to capture what you learn and share it with others. 📺

But with the rapid development of AI Deepfakes, I'm increasingly thinking it's a terrible idea to publish your likeness online.
It's only a matter of time before your face and voice can be flawlessly duplicated in realtime. 👤
At that point your whole family is vulnerable to being scammed. 😬

Note: almost everyone has already voluntarily shared enough on various social media platforms (Facebook, Instagram and TikTok etc.) photos/videos and audio recordings for their likeness to be synthesised by AI. The cat is very much out of the bag! 🐱 👜 🙃
Between Meta, Google, Microsoft and Apple which hoover up people's personal data, your voice & face are very much captured. Never mind the insane amount of data captured by the CCP via WeChat ...
The only question is who at the Mega Tech companies has access to the data and how likely is it that a malicious actor can hack it?

Snapshot vs. Current Face?

If I record and publish my face today that will be a snapshot of what I look like at a given time.
If a malicious person uses the snapshot as their training data for creating a Deepfake they will create a
fake version that looks like "old" you which is less convincing.
However if the algorithm integrates "aging" or "distress" it could still be convincing to a loved one. 💭

Voice Changes Less Over Time

Your voice changes over time, but less than your face/appearance.
The attack vector for impersonating you to your loved ones is far more likely to be voice than face.
e.g: the malicious person makes a phone call to your Mother using a script with a decision tree for her responses.
So simply recording a voiceover on your screencast is giving malicious people everything they need to "clone" you.

Note: I can't stand voice messages in a text conversation! 😡
Besides forcing people to waste time listening to audio they can't easily scan, ⏳
they use more bandwidth and consume space on device. 🤦‍♂️
But they are also a high fidelity training data of your voice that will be used against you and your loved ones!
When people send me a voice message I always reply with text.
Voice messages disrespectful because they waste the time of the recipient.
If you send me a voice message you're basically telling me you think your time is worth more than mine.
Spoiler alert: it definitely isn't!! The value of my time is in the top 0.1% and no longer "for sale"; just don't waste it!

Thoughts?

I'm still leaning towards recording my face & voice for the simple reason that I think the benefits outweigh the potential cost.

Benefits

Captured learning in a format that can be easily digested (consumed) by others.
Improve your own knowledge/skills by focussing on teaching it.
Exposure as a subject matter expert; useful if you want to gain influence or be hired to do the job.

Costs

The main cost - besides the money to pay for recording/editing equipment and time - is the potential to be digitally cloned by a malicious person and have your likeness used for criminal activity. That cost is real for celebrities and wealthy people. 😕
But as the AI deepfake technology proliferates (becomes ubiquitous and free) the changes of your likeness/voice being used against you or your family becomes increasingly likely. ⏳

Deepfakes are Already A Problem! It will Only Get Much Worse ... 😢

https://www.bbc.com/search?q=deepfake
image

By far the most insidious deepfakes is of children.
Horrible people making deepfakes of children to blackmail them ... 😭
As a parent this totally terrifies me and is the biggest reason to protect our children from any social media exposure
and extra careful of sharing photos/videos with family ...

I intend to train our children to not trust any digital communication/media. 🙅
Yes, this will be difficult when they interact with family via FaceTime ... 👪 📱
Will have to figure this out in due course. 💭

But as for recording myself - the topic of this issue - I'm still inclined toward doing it. 🧑‍💻 ⏺️
Simply because, in the scheme of things, I'm a nobody. 👻
Criminals are going to target higher profile people (celebs, sports stars & "influencers" 🤳) and their families
long before they target me/mine. 💭

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant