Every cloud provider faces the same AI infrastructure challenge: chips need to be positioned close together to exchange data quickly, but they generate intense heat, creating unprecedented cooling demands. We needed a strategic solution that allowed us to use our existing air-cooled data centers to do liquid cooling without waiting for new construction. And it needed to be rapidly deployed so we could bring customers these powerful AI capabilities while we transition towards facility-level liquid cooling. Think of a home where only one sunny room needs AC, while the rest stays naturally cool – that’s what we wanted to achieve, allowing us to efficiently land both liquid and air-cooled racks in the same facilities with complete flexibility. The available options weren't great. Either we could wait to build specialized liquid-cooled facilities or adopt off-the-shelf solutions that didn't scale or meet our unique needs. Neither worked for our customers, so we did what we often do at Amazon… we invented our own solution. Our teams designed and delivered our In-Row Heat Exchanger (IRHX), which uses a direct-to-chip approach with a "cold plate" on the chips. The liquid runs through this sealed plate in a closed loop, continuously removing heat without increasing water use. This enables us to support traditional workloads and demanding AI applications in the same facilities. By 2026, our liquid-cooled capacity will grow to over 20% of our ML capacity, which is at multi-gigawatt scale today. While liquid cooling technology itself isn't unique, our approach was. Creating something this effective that could be deployed across our 120 Availability Zones in 38 Regions was significant. Because this solution didn't exist in the market, we developed a system that enables greater liquid cooling capacity with a smaller physical footprint, while maintaining flexibility and efficiency. Our IRHX can support a wide range of racks requiring liquid cooling, uses 9% less water than fully-air cooled sites, and offers a 20% improvement in power efficiency compared to off-the-shelf solutions. And because we invented it in-house, we can deploy it within months in any of our data centers, creating a flexible foundation to serve our customers for decades to come. Reimagining and innovating at scale has been something Amazon has done for a long time and one of the reasons we’ve been the leader in technology infrastructure and data center invention, sustainability, and resilience. We're not done… there's still so much more to invent for customers.

How do you moderate content for 125M athletes while keeping insights intelligent and tone positive? @Strava + Amazon Bedrock + @AnthropicAI 🤝 ✅ personalized training recs ✅ 80%+ accurate content filtering ✅ Harmful content detection at scale The story: go.aws/4hMEhYZ

This was also shocking to see crypto tied with bonds for second place in where people plan to invest. Majorly punching above weight given crypto is 1% of total ETF aum while bonds are 17%. EM better than their assets too vs dev int'l and alts uninspiring to people still.

Announcing a new multi-year, strategic partnership with @OpenAI that will provide our industry-leading infrastructure for them to run and scale ChatGPT inference, training, and agentic AI workloads. This partnership will enable OpenAI to run its advanced AI workloads on AWS’s world-class infrastructure starting immediately. More info in the link below. Learn more: go.aws/495oX7z

Cohere Health’s clinical AI-powered solutions help automate medical necessity reviews securely with #AmazonBedrock AgentCore 🧬⚡🎯 Using #AgentCore, Cohere Health achieves 30-40% faster review times & improved clinical decision accuracy. #AWS 👉 go.aws/43F8TFR

Hospitality that scales effortlessly. Whitbread counts on Amazon DynamoDB to keep tens of thousands of bookings running smoothly every day with consistent performance & seamless scalability. Learn more. ▶️ go.aws/49ruANm

Big news: Project Rainier is now fully operational! Discover how AWS has built infrastructure for the world's most powerful AI supercomputer.

Project Rainier is now live, featuring nearly 500,000 Trainium2 chips dedicated to advancing Claude AI capabilities. The groundbreaking collaboration is expected to expand to over one million chips by late 2025, marking one of the largest dedicated AI compute deployments in the industry.

"Researchers have found that about 80% of kids who believe they’re born in the wrong body ultimately abandon that belief, so long as they don’t receive medical treatment." wsj.com/opinion/the-cdc-can-…

So much attn on $IBIT but $NUKZ (launched same exact day) is also up 185%. Both absolute demons. Proud to say they were both featured in our "24 ETFs to Watch in '24" note. $NUKZ nearly $1b in aum.

$BSOL is beginning life with $220m in assets. Impressive, already half the size of $SSK. Surprised they didn't hold off tho and have it come in on Day One to get volume and flows higher. Good news is now we'll have only organic, easier to measure true demand

I tell ppl it’s like Brooklyn circa 1996 but 60% nyc for 20% the price is better. Gonna steal this.

Five buses. Two cities. One champion. 🚌 💻 ⭐️ On November 30, developers race to re:Invent in a live-streamed hackathon road trip from LA to Vegas, where one team will be crowned champion at Midday Madness. Sign up for livestream notifications. 👉 go.aws/3JrABiw

And yet so much whining and teeth gnashing from the Moon Bois. You’re like Patriots fans back in day crying bc you lost a game amid a 13-3 season. Cheer up, be grateful!

The most interesting people in the world just dropped their best stories. Follow @vivamusverum for the full ride.

"God loves punishing the midwits" is an all-timer quote. You see it all the time, it's practically scripture.

Sixers rookie just broke Bron’s record. Ended up with 34pts in win against the dreaded Celts at home. Sixers needed this boost. Great draft pick by @dmorey !

First thing I did when I landed in U.S. is buy cough drops. You can’t get them in China (no idea why) and I had sore throat when I was there. On flip you CAN get Amoxicillin, which was clutch. ⁦

Texas Qualifying Rodeo 🤠 - F1 Insight Powered by @awscloud Under the Austin sun, it wasn’t brute force — it was balance. 🔵 Max Verstappen mastered the Red Bull’s rotation and rear stability to take pole by 0.291 s over 🔸 Lando Norris, whose McLaren shone in the high-speed arcs but couldn’t quite match Max’s traction and consistency when it counted. 📊 Corner-by-Corner Breakdown: •Turn 1 🔸 (+0.024 s) | Norris dives later on the brakes — confident front end, great rotation uphill — but the McLaren’s rear slides just enough to lose the edge on exit as Verstappen plants earlier power. •Turns 2 → 11 🔵 (-0.06 to -0.11 s) | Red Bull’s platform shines — stable under load and nimble through the S’s. Verstappen carries throttle mid-curve where Lando has to feather slightly to keep it clean. •Back straight 🔵 (-0.03 s) | Slight aero efficiency gain for Red Bull keeps top speed fractionally higher. •Turns 12–15 🔵 (-0.04 s) | Verstappen confident on entry — braking deep and rotating the rear precisely, trusting the RB21’s stability under pressure. •Turns 16–19 🔸 (+0.05 s) | Norris fights back — McLaren more responsive through long-radius corners, less understeer and cleaner throttle modulation. •Turn 20 🔵 (-0.01 s) | Verstappen tidier over the bumps, no mid-corner correction, and rides the kerb clean to close the lap. 💡 Key Differentiator: 🔵 Verstappen spent 59 % of the lap at full throttle vs 🔸 Norris’s 56 % — a testament to Red Bull’s traction platform and Max’s confidence to stay committed. The McLaren danced beautifully when the track opened up, but Verstappen’s ability to control the car at the edge of slip through COTA’s technical middle sector made him the class of the Texas rodeo. #PaidPartnership

“The liquidity of the ETF comes from the underlying assets, mkt makers are liquidity carriers” - Huatai-PineBridge PM (he runs 510300, the biggest ETF by far in China, 9% of total mkt) talking implied liq (hugely imp metric) at our Shanghai ETF event w @JackiWang17