Stop uploading photos of yourself to ChatGPT. Seriously, stop it

Believe me, I understand – asking a chatbot ai to transform an image of your pride and your joy into a whimsical cartoon character is really fun. The call is undeniable, but yes, there are risks.

This exact scenario took place during a family barbecue a few weeks ago. When a well -intentioned parent showed me a family photo modified in AI, my stomach fell on my feet. I couldn’t help but think: Ah, shit … This photo is right there, and who knows what could happen to him. I advised him to be more prudent next time, but honestly? I don’t know if my warning has really landed.

The problem here? It is simply a conscience. Pure and not filtered inhonction.

Should you really download this photo on a chatbot?

Please simply stop downloading children’s photos from chatbots. Or really, whoever didn’t say it was going. It may seem innocent, but there are real risks of confidentiality here and not only for you. You may give up much more than you think, and it is easy to forget that when you play.

Questions to ask before downloading a photo

Before downloading photos on your chatbot of choice, it would be wise to have a conference with yourself and ask the following questions:

• Where is this photo going?
• Could it be used to train AI or share without you knowing?
• Is there something that gives too much? (Home number? Street sign?)
• Do you even know what the Privacy Policy says? (Be honest!)
• Did everyone in this photo said it was cool to download?

What could go wrong

Matheus Bertelli, Pexels

I am not trying to scare anyone here – I really think that AI can be useful when used in a responsible manner. But downloading personal photos can turn against him if you don’t pay attention and pay special attention. Your photo shows much more than you think – think of horodatages, location data, maybe even where you live. This kind of information is a straight gold mine for bad people.

In addition, there is the whole risk of data violation. This means that your photo could be disclosed and used for summary stuff. If you have shared a selfie, for example, someone could easily transform it into a Deepke. In case you have no idea what it is, it is when the AI superimposes your face to someone else. It’s crazy to do and it can deceive many people and also have disastrous consequences.

Once you have sent a photo, you have practically no idea where it ends or how it is kept – just because it disappears from the cat does not mean that it has really disappeared. There could be a seated copy somewhere on a server. It could be used for training, perhaps for moderation, which really knows. This is the frightening part. You don’t really know who access your photos behind the scenes.

You have more control than you think

You are not completely helpless here. A good starting point? Take a look at the privacy policy and see what they really do with your belongings. A good (transparent!) We should provide clear answers to questions like …

• What kind of information does he get? (Messages, photos, etc.)
• How do they catch it?
• How long do they keep it?
• Where is it stored?
• Can you delete it?
• Can you choose to be part of the pool of training data?

Openai’s privacy policy covers many of the above questions. One thing you can do is deactivate cat history in Chatgpt – in this way, your conversations are not used to form the system. This is a solid decision, but yes, not a 100%warranty.

If you always want to share photos with a chatbot, you can really eliminate the photo from its metadata. You can either use a third -party application like Exifotool or you can capture the photo in question, a process that automatically deletes this information.

Let’s talk about consent, because it really counts for me.

Children cannot give it. Period. I do not know why it is a concept so difficult to grasp for some people, but we are there. (Looking at you, youtube family channels). Beyond confidentiality problems, the strongly modification of your photos can have a seriously negative impact on how you might see. Self -confidence can really take a dive here, especially if he is an impressionable child.

If you really want to play with AI, try to use photos of photos or faces generated by the AI of this person does not exist. In this way, you do not draw from your personal library.

Do not take AI at its nominal value

Chatbots seem human, but they are not your friend (despite their often joyful disposition!). You can certainly have fun with AI, but don’t take everything he says as a gospel – that can make mistakes. You will want to keep an eye on your privacy and don’t be afraid to ask questions. Remember this: a little skepticism goes very far nowadays.