🤖 OpenAI's first device is a "humanlike" AI speaker

📖 Anthropic launches free Claude for Teachers

🔒 Microsoft's CEO says you pay for AI twice

⏸️ New York is first state to pause data center permits

📱 China approves Apple Intelligence for iPhones

+ 📊 Daily poll and results

+ 📈 Trending tools and resources

Bloomberg reports OpenAI's first hardware product will be a portable, screenless smart speaker that syncs with ChatGPT, controls smart-home gadgets, and functions as what the company internally pitches as a "humanlike AI companion."

Sources say the device carries a camera, moves on its own through mechanical parts, and draws on personal data like emails to learn its owner over time. OpenAI plans to unveil it this year and ship in 2027.

Apple sued OpenAI on July 10 over alleged trade-secret theft, calling it "the tip of the iceberg." OpenAI insists the device differs from anything Apple sells, though Sonos shares still dropped over 10% on the news.

Anthropic launched Claude for Teachers on Tuesday, giving verified US K-12 educators free premium access for a year. Lesson plans map to academic standards in all 50 states through a Learning Commons connector.

Claude connects to nine classroom tools, including Canva, MagicSchool, and TeachFX. Its agentic Cowork and Code features can grade exit tickets and prep next-day lessons overnight, though teachers must first load class data.

Anthropic says student data stays out of model training under a FERPA agreement co-written with the American Federation of Teachers. It arrives late though, since OpenAI's free ChatGPT for Teachers launched last November.

In a Sunday post on X, Microsoft CEO Satya Nadella coined the "Reverse Information Paradox," arguing companies pay for AI twice: once in cash and again in the proprietary knowledge they feed the models.

Nadella says models absorb "exhaust," the prompts, tools, and corrections that reveal how a business runs. He urges firms to own their data, build private learning environments, and add layers to switch between providers.

The pitch carries some irony. Microsoft invests in both OpenAI and Anthropic and profits from the setup Nadella critiques, and every fix he proposes runs on cloud infrastructure that conveniently points back to Azure.

Governor Kathy Hochul signed an executive order Tuesday pausing new state permits for hyperscale data centers that draw 50 megawatts or more, making New York the first state to freeze the AI buildout.

The pause runs up to a year while regulators write new standards for water, energy, and grid impact. Existing approved projects keep moving, but the freeze stalls more than a dozen proposals.

Hochul chose an executive order over the legislature's own stricter moratorium bill, which she still hasn't signed. Her version sets a higher 50 megawatt bar, leaving smaller clustered projects a loophole to slip through.

China's cyberspace regulator has registered Apple Intelligence, clearing the main hurdle for Apple to launch its AI in the country roughly two years after it debuted in the US.

To meet China's strict AI rules, Apple partnered with local firms. Alibaba's Qwen model powers the main system and handles text and images, while Baidu contributes on a smaller scale.

Apple had little choice but to lean on Chinese models, since foreign AI like OpenAI stays banned there. So in China, the brain behind Siri will largely belong to Alibaba.

A) Yes - 44%

B) No - 56% 🏆