The North Pole’s Icy Past and the Humanoid Data Gold Rush

Last year, a research vessel sailed to the North Pole and found open water. Thin ice. An easy passage. That’s not how it’s supposed to work — getting there used to mean grinding through ice many meters thick for weeks. The Arctic is changing faster than most of us can wrap our heads around.

Now scientists are drilling deep below the seabed to figure out if the Arctic Ocean has ever been completely ice-free before, and what that could mean for the future. It’s a smart move. If you want to predict where we’re headed, you need to understand where we’ve been. The sediment cores they’re pulling up might hold clues about past climate conditions that mirror what we’re seeing today.

Meanwhile, in a completely different corner of tech, something weird is happening. I got invited to join an app that would pay me to film myself doing mundane tasks — putting food in a bowl, microwaving it, that sort of thing. Another site asked if I’d like to remotely control a robotic arm to help improve its dexterity. This isn’t some odd side hustle. It’s part of a growing push by robotics companies to collect data on human movements for training humanoids.

We’re being turned into training data. Every time you reach for a cup, twist a lid, or fold a towel, that’s valuable information for companies trying to teach robots how to function in a world built for humans. The race for real-world data is heating up, and our everyday motions are the prize.

This is one of those moments where the big tech trends collide in interesting ways. On one hand, you have climate scientists digging into the past to understand a rapidly changing Arctic. On the other, you have robotics companies mining the present — literally paying people to perform tasks — to build the future. Both are trying to solve massive problems using data, just from different ends of the timeline.

The spending on this stuff is eye-popping. Google, Microsoft, Amazon, and Meta all set new AI spending records last quarter. Collectively, they’re up 71% compared to the same period last year. Microsoft, Google, and Amazon are reporting big payoffs from the splurge, but Meta’s shares slid after investors got spooked by their plans. The whole thing feels like a high-stakes poker game where everyone’s going all-in.

Anthropic is in the middle of its own drama. The White House opposes their plan to expand access to Mythos, citing cyber risks. There’s also concern the government might lose compute access. Meanwhile, Anthropic is reportedly seeking funding at a valuation over $900 billion. That’s a lot of money for a company that’s still figuring out how to navigate regulatory pushback.

And then there’s Elon Musk. During testimony, he claimed OpenAI’s leaders “looted the nonprofit” and said he “was a fool” for trusting them. But it’s worth remembering that Musk had his own concerns about OpenAI long before this whole legal mess started. The guy has a habit of rewriting history to fit his narrative.

The AI bubble question keeps coming up. Are we in one? Probably. But bubbles don’t always burst cleanly — sometimes they deflate slowly, or the good stuff survives while the hype gets shaken out. The real test will be whether any of this massive spending actually translates into sustainable products people want to use.

For now, I’m more interested in what the Arctic sediment cores will tell us, and whether those humanoid robots will ever learn to microwave a bowl of soup without burning down the kitchen. The future is weird, and it’s happening faster than anyone predicted.