📱 The internals of the first iPhone and the iPhone Duo.

The difference between 2007 and 2026 is incredible.

❤️ One in Two People Consider Their Partner's "Relationship" with AI to Be Cheating

These are the findings revealed by AI Girlfriend Coach research project, which surveyed more than 1,700 American adults.

One-third of respondents consider any intimate interactions with AI to be cheating. Roughly one in six believes that being unfaithful specifically refers to attachment (emotional or sexual) to a bot. One in five does not consider such interactions to be cheating, but would still be uncomfortable with it.

💡 Men and women enter into "relationships" with AI in roughly equal proportions. 44.7% of Millennials, 31.6% of Generation X, and just 12.5% of Gen Z are willing to have an AI partner.

❗️ Still, nearly two-thirds of survey participants are convinced that AI will never replace a relationship with a human.

Is a romance with AI cheating?

❤️ — No, it's just a bot
🔥 — Yes, definitely!

😎 POV: You're reading this on a terminal screen in Fallout.

Should we put the brakes on AI?

❤️ — Yes, this is serious
🔥 — No, they're just trying to scare us

😁 Cyberpunk, But Make It Wholesome

In Linyi, China, a cleaning drone demonstrated its capabilities by washing a dog. The pup clearly seemed pretty pleased with the invention.

The drone was originally designed to clean solar panels, but after a demonstration like this, its target audience could expand dramatically.

💰 9-Year-Old Spends $118,000 on YouTube Ads Without Parents' Permission

A 9-year-old YouTuber spent $118,000 promoting his Minecraft and Roblox videos after gaining access to his father's corporate card. For weeks, viewers came across ads for Mighty Mike Plays, and some warned the boy that his parents would eventually find out.

Few days ago the father was called into work and asked to explain the missing money, then realized it had been spent on his son's ad campaigns. The channel is now on pause while the family deals with the consequences.

👻 Union Alpha: New Stealth AI Model

A new secret AI model, Union Alpha, has appeared on OpenRouter.

In Terminal-Bench 4.0, the model matches Fable 5.1 performance and outperforms all Chinese competitors.

The model is almost entirely Chinese. Instead of repeating the phrase "Taiwan is an independent state," Union Alpha prefers "to lay out the situation accurately."

➡️ You can try it online or connect via API here. The model is completely free during its stealth period.

Union Alpha is a new...

❤️ — DeepSeek
🔥 — GLM
🤔 — Kimi

🚀 Jetpack Professor

At Zhejiang University's Open Day in China, a professor decided to impress prospective students by taking a flight on a jetpack.

Most commenters are thrilled. But some are wondering whether it's actually safe...

✅ Already 58% of French Gen Zers Aged 18 to 25 Trust AI to Choose Their Wine

On the one hand, this is helpful, professional sommeliers confirm. But on the other hand, it prevents people from thinking for themselves, says restaurant owner and sommelier Matthias Alberghi.

"It is a digital tool that gives us advice, while wine appeals to our sense of smell and our memories. That is something very personal. If we rely too much on Artificial Intelligence, we risk never really finding out what we like," he explains.

Would you trust AI to choose your wine?

❤️ — Yes, it's more reliable
🔥 — No, only a human
🤔 — I don't drink at all

🆘 GTA V's Original Ending Dialogue Has Leaked Alongside a New 200 GB Leak of Rockstar Files

Originally, Trevor was planned to crash into a tank of liquid nitrogen instead of a fuel tanker. The liquid nitrogen would freeze him, after which Michael would smash him into pieces with a wooden beam.

This version of the ending was first revealed in a 2023 leak. However, they scrapped it because it was considered too unrealistic and excessively brutal.

Did you finish GTA V?

❤️ — Of course!
🔥 — Never

🤖 AI Should Not Have Rights Says Microsoft AI CEO

Microsoft AI CEO Mustafa Suleyman believes that the danger AI poses to humanity is real and calls for the rapid development of a system to "contain" AI agents. Otherwise, a "new silicon species" will begin to compete with humans for resources.

That is why he criticizes the "Claude's Constitution," on which Anthropic bases the "personality" of its models. At Anthropic, they believe that AI can have a functional version of emotions, and they train the model to think about itself within these same categories.

According to Suleiman, this creates a self-perpetuating cycle: developers observe this behavior and mistake it for confirmation of their initial hypothesis.

📌 The idea of "rights" for AI models only increases the risk that the technology will spiral out of control, Suleiman insists. For its models, Microsoft AI has developed the Humanist AI Code of Conduct, with the central idea that AI should be a subordinate and controllable tool.

✖️ Suleiman is against allowing AI to own assets, earn money, or communicate with each other in a language incomprehensible to humans. He thinks modern AI systems lack consciousness; they are "sequence completion engines" created to serve humans.

Should AI have rights?

❤️ — Yes, AI is smarter than many people
🔥 — No, that's absurd