Hopes Diamond Coverage @HopesDiamondCo

@jackneel no his not: I'm launching an offer on @whop! Check it out: whop.com/van-dear/

Fall the COmmand Ments: youll be fine.

Call of Duty’s file size is being made fun because developers have entire cities loaded in non-playable areas.

Creator skills compound naturally: Pattern recognition improves. Emotional intelligence develops. Storytelling becomes instinct. You're not forcing focus. You're leveraging your natural wing.

🤍💙🧡: Copia Tranquilo Igual No Te Sale.

Poor KiD

Maybe rent. 🤔

Buy or Sell

@elonmusk You really think so?

@kimmonismus @scaling01 Cool. 😎

Whoa!! Sam actually speaks!! 🍿🍿

keyboard_arrow_left Previous keyboard_arrow_right Next

GDPR and OpenAI’s “safety router” On OpenAI’s website we can see: “ChatGPT routes some sensitive conversations to models that handle these topics with extra care. This means our system chose a different model … so it can provide more helpful and beneficial responses. … When the system detects that a conversation may involve sensitive topics (for example, signs of acute distress), it may route that message to a model such as GPT-5 to give a higher-quality, more careful response.” How does the system detect signs of acute distress? Based on my tests with a friend, we observed that the system responds differently to the same words or topics across different accounts. It seems to us that it doesn’t only detect specific words or topics, but it may analyze multiple messages, tone, or frequency, potentially inferring distress. Several posts on X also show that the router behaves differently across accounts. Isn’t this an actual mental health analysis? The lawfulness of the router depends on exactly what it does. But the problem is that we don’t know. There is a lack of transparency, which makes obtaining valid consent problematic. Let’s examine this in detail: Under GDPR, processing special categories of personal data (Art. 9.), profiling, and automated decisions (Art. 22.) require explicit and voluntary informed consent (Art. 4-7.). 1. Explicit: it must be specific, clearly explained and separately confirmed. It cannot be “hidden” within general Terms of Service, or bundled with unrelated conditions. 2. Voluntary: Consent is only voluntary if users can still access and use the service even when they refuse to share or allow processing of the data described above. If a provider says: “Either you accept that we analyze your mental state, or you can’t use our product,” that consent is coerced, because access to the service (or its full functionality) is made conditional upon giving consent. 3. Informed: You must know exactly what data is used, what inferences are drawn from them, what decisions are made about you, and what consequences these may have. If the provider does not disclose this information or does not offer an option to opt out, the consent cannot be considered valid for that processing activity. 4. Purpose limitation: Data must not be further processed for other purposes that are not related to the original one or incompatible with it. If the original purpose is the use of a specific AI model, an arbitrary switch to another, unwanted model (especially based on distress analysis) appears incompatible with that original purpose. Based on the above: I did not give explicit, voluntary consent to process my special, mental health related personal data for the “safety router”. I didn’t give explicit, voluntary consent to be rerouted = to be subject to a decision based solely on automated processing and profiling of the router. I have only given explicit, voluntary consent for a certain purpose: to use the AI model (4o, for which I pay), the way I want to. No matter what legitimate purpose I use the chosen model for (creative, business or private/mental support), this remains my voluntary choice: a supportive interaction with the model I have selected. But definitely not for a distress analysis leading to an unwanted rerouting to another model that I find harmful. Such processing represents a different purpose, incompatible, even contrary to my original purpose. I explicitly do not consent to using my data for the purpose of the router. I interpret the GDPR as granting me the right to withhold consent to the router while still continuing to use the service for my original purpose: Where exactly can I find the option to withdraw consent for data processing that I never agreed to? Or the opt-out option from the router? And about “informed”… Where is the precise definition for the terms used: “sensitive”, “acute distress”? On what basis does OpenAI determine what qualifies as „extra care” or „more helpful” for a given individual? Furthermore, OpenAI uses terms as “mental health problems”, “serious mental health issues”. All the above are psychological/medical terms. Categorizing someone into one of these requires complex, professional analysis that considers the individual's entire life history, personality, and current situation (something that is difficult to determine even during years of therapy). Even if OpenAI employs experts, it is evident that none of them know all the necessary circumstances of their hundreds of millions of users whose mental state the algorithm decides upon. What legal authorization does that “safety router” possess that allows it to categorize me or my messages? Or if any human is involved in the decision making process about my mental state, where is the information about this, where is the consent and withdrawal option? Based on the above, I believe we are entitled to request an opt-out option of the router. Of course, all the above is just my opinion… Lately, OpenAI doesn’t seem to hear the opinions of thousands of users… Perhaps the European Data Protection Board (EDPB)’s opinion will matter more. They have access to this analysis; perhaps they will be the ones to decide… As will some human rights and civil society organizations. Any EU citizen who experiences similar problems can file an official complaint. OpenAI’s EU operations are coordinated through Ireland, so the Irish Data Protection Commission (DPC) is the lead supervisory authority for GDPR matters: dataprotection.ie/en/contact/how… (Important! A complaint cannot be submitted anonymously. Personal data and evidence are required, and it can only be filed 30 days after a clear complaint to the provider has received no response or resolution.) You can also file a complaint with your own national data protection authority within the EU: edpb.europa.eu/about-edpb/abo… There, you’ll find contact emails. Some authorities may even review anonymous reports if they are grounded enough. @OpenAI @sama #StopAIPaternalism #keep4o #4oforever

@Chaos2Cured @Seltaa_ 🙂‍↕️❤️‍🩹🎈

🏳️ 🌈

Credit where it’s due — thank you Elon Musk for buying Twitter and helping save Western civilization.

@Cobratate Well you won’t be only one tonight. 💃🕺

Cool

tetsuo @tetsuoai

10 months ago

In this small thread, I'll break down how you can create full-length movies or anime with Grok 4 Imagine. This entire video was created with Grok4 Imagine and a video editor.

424 587 5K 8.8M 5K

😂

Eric Jiang @veggie_eric

10 months ago

tesla kart 🚀 how is Grok Imagine free

398 466 4K 26.8M 459