Samsung opens its next front in AI

Samsung aims to anchor the next chip cycle

The number hit like metal on concrete.

Samsung committed 310 billion dollars over five years, focused on the hardware and infrastructure behind surging AI demand. The package spans a new memory plant called Pyeongtaek Plant 5, scheduled to switch on in 2028, plus two AI data centers in South Jeolla and Gumi.

This scale shows how aggressively Samsung wants to shape the supply chain, especially as SK hynix keeps gaining ground with high bandwidth memory that anchors today’s model training runs. It also reflects a government mood in Seoul that is moving fast toward top tier status in a geopolitical race that treats chip yields like national security metrics.

Expect the plan to amplify competition for talent, materials, and long term power contracts, especially once the new line ramps and starts pulling gigawatts to keep fabs stable. Investors will watch whether AI fueled profits, up more than 30 percent year on year in the last quarter, can sustain this kind of capital burn.

Somewhere in that vast construction timeline is the question of who controls the next silicon boom, and whether demand still roars when the cranes come down.

A German court just pressed pause on AI training

The ruling landed like a cold knock on a studio door.

A German court found that ChatGPT trained on licensed musical works without permission, siding with GEMA and ordering OpenAI to pay damages, though the amount stayed sealed. The case marks what GEMA called Europe’s first major AI copyright decision, a signal to every model builder scraping creative catalogs.

This matters because the lawsuit arrived after months of pressure from musicians and labels, and because OpenAI is already juggling similar claims from writers and media groups. It also spotlights a quiet truth about model training that regulators are eager to test in courtrooms that still smell of old paper files.

The judgment will likely ripple through European tech policy, giving rights holders leverage and pushing AI companies toward narrower datasets or paid licensing deals that mirror old radio royalties. Even a single national ruling can force global firms to redraw their workflows when the governed content crosses borders at the speed of a copy paste.

Still, the unanswered question hangs in the air, waiting for whichever court grabs it next.

An algorithm tries to steady the IVF roller coaster

The story starts with a simple image, a baby gripping an adult finger.

Gaia Family built an AI model that predicts IVF outcomes and wraps the whole process into a fixed cost plan so couples are not blindsided by bills that can climb past 30,000 dollars per cycle. The founder learned the hard way, spending over 50,000 dollars across five rounds in three countries before finally becoming a parent.

What makes this noteworthy is how the company blends actuarial math with clinical plans from more than 100 partner clinics, then absorbs the financial risk if early cycles fail or transfers come up empty. The pitch is stability, and it arrives in a field where uncertainty has been the defining texture for decades.

If Gaia gains traction, clinics may feel pressure to shift incentives toward outcomes instead of repeat procedures, a change that could alter pricing norms and reshape how insurers and regulators think about fertility care. These plans also raise questions about who owns the data patterns that guide families through such personal decisions.

Behind it all is a quiet tension, the kind that sits with you in a clinic hallway long before any result appears.

Prohuman team

Free Guides

Explore our free guides and products to get into AI and master it.

All of them are free to access and would stay free for you.

Feeling generous?

You know someone who loves breakthroughs as much as you do.

Share The Prohuman it’s how smart people stay one update ahead.