Google's AI energy bill ⚡

Google’s latest environmental report shows just how expensive the AI race has become.

Last year, the company’s electricity use jumped 37%, greenhouse gas emissions rose 18%, and water consumption climbed 34% as it rapidly expanded its AI infrastructure.

Google says its data centers are becoming more efficient and it signed a record amount of clean energy. The problem is that AI demand is growing even faster than those improvements.

The report highlights one of the biggest trade-offs in AI today. Companies want increasingly powerful models, but building them requires enormous amounts of energy, chips, and cooling.

The AI race isn’t just about who builds the smartest model anymore. It’s also about who can power it sustainably.

If your team is still patching selectors and rotating proxies, you’re maintaining infrastructure your competitors stopped owning.

Scraper Studio - launching today from Bright Data turns any public website into a hosted data API from a single prompt:

No sales call. No demo. Just spin up a scraper and watch it run.

One of the largest studies yet on AI adoption found that companies making the biggest investments in AI are growing their workforces, not shrinking them.

Researchers analyzed spending and hiring data from more than 21,000 U.S. companies and found that firms investing most heavily in AI increased headcount by roughly 10% over the following two years. Entry-level hiring grew even faster, rising 12%.

The gains weren’t spread evenly.

Companies making only small AI investments saw no meaningful change in hiring. The biggest increases came from firms that went all in on AI, with hiring expanding across engineering, sales, administration, finance, and customer service. Most of the early gains were concentrated in technology companies.

The researchers also found the effect wasn’t immediate. Hiring tended to pick up several months after adoption, suggesting companies first learned how to use AI before expanding their teams.

The findings push back against one of the biggest fears around AI. So far, the companies investing most heavily in AI aren’t cutting jobs. They’re using AI to grow faster, then hiring more people.

Cognition has unveiled Devin Fusion, a new system that combines multiple AI models instead of relying on a single one for every coding task.

Rather than sending every request to the most powerful and expensive model, Devin Fusion dynamically routes work between frontier models and cheaper “sidekick” models. Cognition says the approach delivers similar coding performance while cutting costs by around 35%.

This reflects a bigger shift happening across AI.

For the past two years, the race has been about building the smartest model. Now it’s becoming just as important to use those models efficiently. As AI costs become a bigger concern for businesses, knowing which model to use, and when, may matter as much as building the model itself.

The future of AI may not belong to a single model. It may belong to the systems that know how to combine many of them.