BlueBalloon

Just a Blue Balloon rising above the bullshit, doing Faceless Youtube.

YT Automation with @thinksh0w

Studied under @noahmorriz @CheZwaans

[BlueBalloon @BlueBalloonYT]

@thegoldeenhand You can buy just about anything off of https://t.co/QyMga7sRpw

[BlueBalloon @BlueBalloonYT]

HOT YouTube take.

YouTube just started putting AI labels on videos automatically (https://t.co/KE5FAhtE57)

I think this will lead to people stopping noticing or caring about it.

Humans adapt to their situation, and as more and more people use AI it will just become so common that people stop caring or just don't notice the AI tag anymore.

It's called many things: habituation, Sensory Adaptation, "Boiling the Frog", etc

[BlueBalloon @BlueBalloonYT]

@mikeydsoftware @theo bro calm the fuck down. You do sound like a crazy person. Your just feeding into it streisand effect.

[BlueBalloon @BlueBalloonYT]

@noahmorris Probably the best is the Ulanzi clock. https://t.co/O53ibOxDhO

I have the clock posted and the one in your photo (LaMetric TIME). Ulanzi is better

[BlueBalloon @BlueBalloonYT]

@Flourish4545 @ytajourney There are some open source projects that don't watermark or has the watermark removed. Slower to make, doesn't sound as good. Work trying though

[BlueBalloon @BlueBalloonYT]

@DiversifiedAk TBH not sure. I am sure Vidrush will come out with some advice about what to do. I would keep an eye out for that

[BlueBalloon @BlueBalloonYT]

To everyone freaking out about SynthID in 11Labs, relax. Seriously calm down... 11Labs was ALREADY putting a watermark in their content.

To anyone saying this is the end of YTA, you are misinformed or stupid. I have seen a lot of talk around the EU's AI Act that is coming into effect in Aug 2026. So what does it ACTUALLY say (https://t.co/5VQus8K0Bh):

"Transparency risk
This refers to the risks associated with a need for transparency around the use of AI. The AI Act introduces specific disclosure obligations to ensure that humans are informed when necessary to preserve trust. For instance, when using AI systems such as chatbots, humans should be made aware that they are interacting with a machine so they can take an informed decision.

Moreover, providers of generative AI have to ensure that AI-generated content is identifiable. On top of that, certain AI-generated content should be clearly and visibly labelled, namely deep fakes and text published with the purpose to inform the public on matters of public interest.

The transparency rules of the AI Act will come into effect in August 2026."

Lets read some more shall we? (https://t.co/z1zdUy5F9C)
"Providers of AI systems, including general-purpose AI systems, generating synthetic audio, image, video or text content, shall ensure that the outputs of the AI system are marked in a machine-readable format and detectable as artificially generated or manipulated. Providers shall ensure their technical solutions are effective, interoperable, robust and reliable as far as this is technically feasible, taking into account the specificities and limitations of various types of content, the costs of implementation and the generally acknowledged state of the art, as may be reflected in relevant technical standards. This obligation shall not apply to the extent the AI systems perform an assistive function for standard editing or do not substantially alter the input data provided by the deployer or the semantics thereof, or where authorised by law to detect, prevent, investigate or prosecute criminal offences."

This WHOLE move and the WHOLE law is to let people know that what they are looking at is AI generated and to make sure that AI companies and any generated AI stuff are watermarked as such.

As more and more AI generated stuff gets produced, there NEEDS to be a way to tell what is AI and what is real. You, the YTA person might not like it, but regulators and society at large agrees that we should mark what is AI and what is not....

NOW, I think most people probably don't give 2 shits if what they are looking at is AI or not if it is entertaining or provides value. I love watching creative AI videos and shorts that people make. If its good, its good!

But when you try to pass off AI stuff as real and do it in a sneaky way, it will get flagged as AI. Politics niche is huge and makes a lot of money for a lot of people. Well what happens when there is a shit ton of AI clones and fake AI politics channels (even more than there might already be) with (AI) Trump giving completely fake speeches or talks about things that didn't happen. Or Anti Trump channels saying things like "2 minutes ago! Trump sent to jail" with AI video and audio of things that didn't happen.

Youtube has already said they are not banning AI Voice overs from the platform. We ALL KNOW that inauthentic mainly hits low quality AI mass produced slop. It does hit high quality and real channels as well, but lets not pretend your fuckin 5 image, 2 hour long AI Story was not AI slop for old people...

Around 90% of adults couldn't tell AI videos from real videos according from a study from Runway AI (Take it with a grain of salt). https://t.co/H88ftkP7lV

We were all pretty blown away by Seedance 2's quality. In literally 3 years, the "Will Smith Eating Spaghetti" (2023 - 2026) went from comically bad, to damn realistic.

So in summary:
Does YouTube have some alternative motive? Probably not.

Does it give YouTube an easier way to detect AI VO's.
Yes it does.

Will YouTube ban your channel or give it inauthentic content only because you used 11Labs voice?
No it won't.

Should you stop posting super low effort/quality AI Slop if you don't want to run risky channels now?
Yes you should.

Will you probably see an AI tag or disclaimer somewhere when people watch a video with 11Labs or other AI content, even if you don't check the box? Yes they probably will.

Will Youtube fuck your channel over because you used 11Labs before the deadline in August? No, they won't, unless you run obvious AI slop, then its always at risk.

Google and Large Tech don't just jump around with glee and bend over backwards to comply with regulators and governments. They do the bare minimum and mostly maliciously comply. The law says NOTHING about banning AI generated content or making it impossible to profit off AI generated content. Therefore it will not happen. SOMETHING WILL happen in the future with more inauthentic as AI videos/images/VO/Editing becomes better and better.

When @VidrushAI from @noahmorris gets as good, as diverse, as cheap, and as creative as human editors.... Then something will need to happen, but that time is not quite now (maybe Vidrush 2.0?)

Thanks for coming to my ted talk. What do you think? Lets have a productive conversation about this, not fear based hot takes and brain dead shallow tik tok brain replies.

[BlueBalloon @BlueBalloonYT]

You should probably be fine to do it the way you do it now. The AI checkbox on YouTube is more for if someone might think what their watching is real or not. So if you have an AI avatar giving advice about silver or gold trading then you might want to check the box (fake person with AI images/videos saying something that didn't happen)

[BlueBalloon @BlueBalloonYT]

@0xCarnival Cheaper, less work, less commitment, life situation. Lower birthrates happens to all countries as their standard of living rises and GDP climbs. (At least that is the rule rather than the exception.)

[BlueBalloon @BlueBalloonYT]

@ScottQu35382152 @anshelsag @SnazzyLabs your trolling right? Why would you put a knife in the dishwasher....

[BlueBalloon @BlueBalloonYT]

@0xCarnival @thsottiaux same thing happens to me too! Gotta reset the app and it will work. OR you can switch it after you sent a message. Very Strange!

[BlueBalloon @BlueBalloonYT]

@samgrows Whats the go to stack for you?

[Noah Morris @noahmorris]

I've been getting a lot of questions regarding how to avoid or safe guard your channel from inauthentic content from a large range of creators both faceless and nont faceless.

because youtube enforcement is so incredibly vague and so many guru's are just sharing unfounded and unsupported advice at the moment i want to opensource a large scale project that attempts to figure out the patterns that trigger inauthentic content so we transparantely know what is and is not allowed on youtube.

However in order to do so i need a large enough sample size of channels that have been hit with inauthentic content in order for this to work.

That's why i humbly ask if we can crowd source channels that are confirmed to have been affected in this google form so i can collect enough data to actually find some of the more important signals.

(note that i will NOT share any of the channels shared with us publically in this google doc ill just use them to capture the data i need to run the study)

https://t.co/BRlyKpdC4j

[BlueBalloon @BlueBalloonYT]

I asked Nano Banana Pro/2 and GPT Image 2 for a Youtube thumbnail. "Give me a Youtube thumbnail!" and they all gave me thumbnails about how to make money and blow up on Youtube... I think that says something, not sure what though!

[BlueBalloon @BlueBalloonYT]

@0xCarnival Not even in the 99% 😭 usually companies are measured in how many 9's they have after the decimal point... Anthropic is struggling for 9's before it!

[BlueBalloon @BlueBalloonYT]

@DocBorelli @DeryaTR_ 2nd go through with project Genie. Not as good as the first one. People/faces are quite hard. Wanted to test it with a different art style as well.

[BlueBalloon @BlueBalloonYT]

@DocBorelli @DeryaTR_ Got to test out Google Genie. It is pretty cool, a cool tech demo at best right now but still cool! I still think I will win the bet ;P

[BlueBalloon @BlueBalloonYT]

@AleksejAros @noahmorris @nexlevio Ok AI bot...