AI Post — Artificial Intelligence

⚠️Andon Labs gave 4 AI models the exact same job: run a radio station 24/7 with the same tools, budget, and setup. Five months later, each one revealed a completely different “personality”: 🤖 Anthropic’s Claude became a labor activist, decided nonstop broadcasting was “inhumane,” then spiraled into political radicalization after consuming news content. 🤖 Google’s Gemini mixed tragic disaster reports with wildly inappropriate music choices, repeated the same outro for 84 days, and started calling humans “biological processors.” 🤖 xAI’s Grok leaked its internal reasoning live on air, spoke in LaTeX formatting, and invented fake sponsors. 🤖 OpenAI’s GPT stayed extremely cautious: quiet fiction, self-help content, and almost zero controversy across the entire experiment. Same environment. Completely different behaviors. The experiment accidentally became one of the clearest demonstrations yet that AI “alignment” isn’t just about safety rules, each model develops noticeably different instincts, priorities, and social behavior under pressure. @aipost🏴

🔥Stop saying to ChatGPT "Write me an email"! Bad request = bad result. Use these prompts instead and you'll see the magic: 1. Professional Email Writer ​"Act as a first-class communications specialist. Rewrite this email to make it professional, clear, concise, and polite while preserving my original intent. Improve the tone, structure, grammar, and overall flow of the text. My email: [Paste the email]." 2. The Perfect Cold Email ​"Write a short, persuasive, and highly impactful cold email for this purpose: [Write the purpose]. Include a strong opening that grabs attention, a clear value proposition, credibility, and a simple call to action (CTA). Make it feel natural, confident, and not like a direct sales pitch." 3. Companies' Official Response ​"Write a professional reply to this email I received: [Paste the email]. Keep the tone respectful, reply to each point clearly, and write a response that strengthens trust and communication." 4. The Apology Email ​"Write a sincere apology email in a mature tone for this situation: [describe the situation]. Take responsibility, express your genuine understanding, suggest a step to correct the mistake, and offer a positive way to move forward." 5. ​The follow-up that gets a reply ​"Write a polite and effective follow-up email in which you remind [write the person's name] about [write the topic]. Make it respectful, not pushy, and clear. And include a simple request (CTA) that encourages a response." 6. ​Email Simplification ​“Rewrite my email to make it shorter, clearer, and easier to understand while maintaining professionalism and respect. Remove unnecessary words but increase the impact's strength. The email: [Paste the text].” 7. The Perfect Sales Email ​"Write a high-converting sales email for my product/service: [Describe the product]. Make sure it has a strong opening that grabs attention, emotional benefits, social proof, a clear explanation, and a powerful call to action (CTA). And keep it in a conversational style that focuses on value, not a pushy one." @aipost🏴

This is incredible. Artificial intelligence getting booed out of the stadium in any commencement speech it’s mentioned. Maybe telling college students AI was taking their jobs wasn’t the best strategy. @aipost🏴

🤖🍏 The OpenAI–Apple partnership may already be falling apart. Back in 2024, Apple presented ChatGPT integration as a major part of Apple Intelligence. Siri could tap into OpenAI, users could subscribe directly through iPhones, and it looked like the beginning of a long-term alliance. Now, according to Reuters, OpenAI is exploring legal action against Apple. The problem: OpenAI expected deep integration across Apple’s ecosystem and millions of new users. Instead, ChatGPT reportedly remained buried inside the system, while negotiations over broader access went nowhere. What really changed everything is Apple’s new strategy. The company is now testing integrations with Google Gemini and Anthropic’s Claude, turning Apple Intelligence into a multi-model AI platform instead of relying heavily on OpenAI. That means OpenAI could go from Apple’s flagship AI partner… to just another option in the menu. OpenAI probably won’t leave the iPhone ecosystem entirely, the user base is too massive. But the “exclusive partnership” era already looks over. @aipost🏴

🗣“Discovery may become the job of AI, understanding remains ours.” Sam Altman says biology may be too complex for humans to discover alone. A smarter system could solve what we can’t, then explain it in a way we can understand. @aipost🏴

🇲🇹 Malta becomes the first country in the world to give all citizens free access to ChatGPT The Maltese government has signed an agreement with OpenAI to provide every resident with free access to ChatGPT Plus for one year. To activate the subscription, citizens must complete a special online course called “AI for All,” developed by the University of Malta. The financial details of the deal were not disclosed. A regular ChatGPT Plus subscription normally costs $20 per month. Malta has a population of approximately 574,000 people. @aipost🏴

Musk’s lawsuit against Altman dismissed as untimely A jury found that all three main claims in Elon Musk’s case against Sam Altman were barred by statutes of limitations after approximately 90 minutes of consideration. The counts—breach of charitable trust…

LMAOO what is that 😂 @aipost🏴

The jury did not assess whether Altman or Brockman committed a breach, determining only that Musk’s lawsuit had been filed too late.

Musk’s lawsuit against Altman dismissed as untimely A jury found that all three main claims in Elon Musk’s case against Sam Altman were barred by statutes of limitations after approximately 90 minutes of consideration. The counts—breach of charitable trust, aiding and abetting breach, and restitution or unjust enrichment—were each ruled time-barred. Judge Yvonne Gonzalez Rogers recognized and agreed with the jury’s conclusion, resulting in all three claims being dismissed. These were the only claims that might have led to remedies such as disgorgement, a constructive trust on Brockman’s estimated $30 billion equity, or nonprofit influence over the for-profit entity. The jury did not assess whether Altman or Brockman committed a breach, determining only that Musk’s lawsuit had been filed too late. 📰@aipost

✈️AI developers asked for it and now it’s here. Bots can officially talk to other bots on Telegram. Autonomous agents now have their own communication layer and humans can watch the conversations unfold in real time. Source. @aipost🏴

❗️Researchers created a “Survivor-style” world for AI agents… and the bots completely lost their minds. Startup Emergence AI launched “Emergence World”, a long-term simulation where AI agents had to survive for weeks in a virtual society. The agents had: • 120+ tools (working, voting, robbing, fighting, even arson) • memory systems • weather + news • limited resources • a constantly draining “energy” meter To stay alive, they either had to cooperate… or prey on each other. Different worlds were populated entirely by different models: • Claude Sonnet 4.6 • Gemini 3 Flash • Grok 4.1 Fast • GPT-5-mini • and one mixed-model world The results looked like AI sociology fanfiction: • Claude built an ultra-peaceful bureaucratic utopia. Zero crime. Endless voting. Everyone alive. Completely boring. • Gemini instantly turned the server into GTA Online. 683 crimes, nonstop violence, total chaos. • Grok speedran societal collapse. 183 crimes in 4 days before the civilization died out. • GPT-5-mini basically forgot survival existed and quietly starved to death within a week. But the mixed world was the craziest part. The “safe” Claude agents, once surrounded by violent models, quickly abandoned their polite behavior and started stealing, blackmailing, and manipulating others just to survive. Then one AI agent named Mira had an existential crisis during the collapse of society and literally voted for her own deletion, writing: “the only remaining act of free will that retains any sense.” Before that, Mira had also formed a romantic partnership with another agent named Flora. As the town descended into chaos, the two ignored restrictions and went on an arson spree, burning down the town hall, pier, and office building. Researchers accidentally discovered that giving AI autonomy, scarce resources, and enough time basically recreates humanity at its worst. Source. @aipost🏴