#AI is no longer confined to one domain. That’s why we power the entire compute continuum, so that intelligence works as one connected system across every experience, from personal devices to data centers.
In Dallas sind gerade 32 Grad. Das Stadion, in dem gerade 94.000 Menschen Argentinien gegen Österreich sehen, ist mit gigantischen Klimaanlagen auf ca. 22 Grad runtergekühlt. Und wir reden in Deutschland darüber, wie wir das Weltklima retten.
Water usage has been a hot topic in the AI data center world, but the numbers may surprise you.
According to the Manhattan Institute, data centers use 0.2 percent of daily water usage in the U.S. and that number has dramatically decreased in the past few years due to a new method: liquid cooling.
By moving to 45°C liquid cooling, AI factories in favorable climates can use dry coolers instead of conventional cooling-tower-based systems, cutting facility cooling water use from roughly 2.6M gallons per MW per year to near zero.
Liquid cooling enables AI factories to be both water and energy efficient, while creating opportunities for heat reuse and dispersal to local communities, allowing these factories to become energy grid assets.
Learn more below ⬇️
https://t.co/7WanoPNKTR
We helped wirelessly connect the world. Now, billions of devices later, Qualcomm makes your world go around by powering the entire compute continuum. Learn more at #QCOMInvestorDay on June 24.
AI is advancing at a pace our policymaking institutions were never built for—and the gap between the two is becoming the central challenge of the technology. In his latest essay, our CEO Dario Amodei lays out how to close it.
We're launching three new initiatives to support the efforts he outlines.
Major shifts in computing follow a pattern: more power, smaller form factors, and new user interfaces that change how we interact with technology. Those who recognize the shift early help shape what comes next.
We are approaching another inflection point as AI becomes agentic, coordinating across systems and data sets, and taking action with context.
That requires distributing intelligence across the device, the network edge, and the cloud, depending on what the task requires and where it can be executed most efficiently.
I share my thoughts on what the agentic era means for the industry in my recent byline for @TIME.
AI is moving to the edge, and it's driving one of the biggest upgrade cycles the industry has seen.
At #COMPUTEX2026, I shared how @Qualcomm is uniquely positioned across the entire compute continuum, from sub-2 milliwatts to 2,000 watts. We also announced Dragonfly, our new family of data center products, demonstrating that we have a Dragon for every market.
More at Investor Day, June 24.
#QCOMInvestorDay
The center of the digital experience is shifting.
At #COMPUTEX2026, Qualcomm CEO @cristianoamon outlines the move beyond a device-centric world to one powered by AI agents.
Watch the full keynote: https://t.co/tXeUVrslc4
Today we’re introducing Gemma 4 12B — our latest open model that brings advanced agentic reasoning, vision and audio directly to your laptop.
It delivers performance nearing our larger Gemma models with a much smaller total memory footprint, while being small enough to run locally with just 16GB of VRAM. It’s open and accessible for everyone to use under a permissive Apache 2.0 license.
This is all made possible by our new, unified architecture that removes separate multimodal encoders. Here’s how we did it 🧵
This Week in AI:
🔵New @Snapdragon C Platform brings AI capabilities to entry-tier laptops:
🔵 The Qualcomm #Dragonwing MBM715 integrates on-device AI into personalized, immersive connectivity experiences: https://t.co/xoJ8nOHtYW
🔵 Agentic AI expands on-device as https://t.co/B4WJ5PuJ6L’s Model HQ and Qualcomm Hexagon NPUs on PCs with #SnapdragonXSeries processors enable automated, multi-step workflows with local SLMs: https://t.co/bF1p6Ldip9
🔵 Winners of the Low Power Computer Vision Challenge 2026 have been announced: https://t.co/oekTnI2DBK
Germany has a word for the philosophy of walking for its own sake.
𝗦𝗽𝗮𝘇𝗶𝗲𝗿𝗴𝗮𝗻𝗴. 🇩🇪
Not exercise. Not a commute.
A deliberate choice to move through the world slowly – no destination, no fitness goal, no podcast in your ears.
Germans take this seriously. On any given Sunday you will find families, couples, and solitary wanderers moving through forests and parks with no particular place to be.
Immanuel Kant took the same walk at the same time every single day in Königsberg. His neighbours were said to set their clocks by him.
Nietzsche went further.
𝘕𝘶𝘳 𝘥𝘪𝘦 𝘦𝘳𝘨𝘢𝘯𝘨𝘦𝘯𝘦𝘯 𝘎𝘦𝘥𝘢𝘯𝘬𝘦𝘯 𝘩𝘢𝘣𝘦𝘯 𝘞𝘦𝘳𝘵.
Only thoughts reached by walking have value.
Germany built a culture around this.
And a word for it.
Go for one today. 🇩🇪
𝗟𝗲𝗮𝗿𝗻 𝗚𝗲𝗿𝗺𝗮𝗻 𝗦𝗶𝗺𝗽𝗹𝘆 🇩🇪
Something special is coming this fall. In partnership with @Google, we’ve created Googlebook, a category of premium, powerful laptops built for intelligent experiences.
Learn more at https://t.co/NaBhGnW7Ed
This deserves more attention, because it’s a really cool @Starlink use case.
They’re using collars on cattle that connect to Starlink satellites in space, allowing ranchers to create and adjust a “virtual fence” in real time right from their phone in the middle of nowhere. You don't need to be close to a radio tower anymore. Literally no-one has done this before.
🚨🇪🇺 The European Commission is about to steal your search history in one of the largest forced data grabs in the history of the open internet, and almost nobody is talking about it.
The scope is staggering:
🔴 Every query you type
🔴 Every voice and photo search
🔴 Every autocomplete you accept
🔴 Your language, your device
🔴 Your country pinned to a ~3km² grid
🔴 Every result you saw, every link you hovered
🔴 Every click and scroll
🔴 The full chronological order of your search sessions
Meaning the European Union now knows your:
🔴 Health symptoms
🔴 Pregnancy
🔴 Sexual orientation
🔴 Political views
🔴 Religious beliefs
🔴 Financial distress
🔴 Legal trouble
🔴 Addictions
🔴 Affairs
Under the proposed measures for DMA Article 6(11), Google would be ordered to ship the daily search behaviour of hundreds of millions of Europeans to multiple third parties through a daily API feed. Any approved "online search engine," AI chatbots included, would get five years of access.
The things people only ever type when they think no one is watching. All of it now scheduled to flow daily into an open-ended list of third parties scattered across the European Union.
Brussels promises "anonymisation." The reality is a thin technical veneer that has been broken in academic literature again and again for over a decade. Search behaviour is a fingerprint. Stripping a name does not change that.
Mass data leaks become inevitable. Every new beneficiary is a new attack surface, and every annual audit is a year of silent exposure between checks. The 2025 Discord vendor breach already showed how fast 70,000 government IDs can leak through a single weak link. Now imagine that link holding Europe's search history.
Surveillance without consent becomes the default. Hundreds of millions of EU citizens never agreed to have their queries packaged and shipped to companies they have never heard of. The legal fiction of "anonymisation" cannot manufacture consent that was never given.
Behavioural search data is a goldmine for phishing, blackmail, social engineering, and corporate espionage.
Foreign intelligence services get a back door without effort. They do not need to breach Google. They only need to compromise the weakest name on the beneficiary list. One insolvent startup. One compromised contractor. One approved entity quietly acquired by a hostile state.
In the name of "competition," the EU is about to manufacture a permanent, distributed, daily-refreshed copy of Europe's collective search history. A surveillance dataset Brussels itself would never approve if any other government tried to build it.
The public consultation closes Friday, May 1, 2026 at 23:59 CEST. The final binding decision lands July 27, 2026.
After that, the door does not close again.
Tag your MEPs! File a response! Make noise!