🌐 Google rolls out Gemini in Chrome in 7 new countries ↗
Google is pushing Gemini deeper into Chrome, expanding the browser feature to Australia, Indonesia, Japan, the Philippines, Singapore, South Korea, and Vietnam.
It is rolling out on desktop and iOS in most of those markets, though Japan is desktop-only for now. Small caveat, but still - browser AI is becoming less “feature” and more furniture.
🎵 Deezer says 44% of songs uploaded to its platform daily are AI-generated ↗
Deezer says AI-generated tracks now make up 44% of all new music uploaded to its platform daily. That is almost half the incoming music pipe, which is a little brain-melting.
The company says it is receiving nearly 75,000 AI-generated tracks per day and more than two million per month. The music internet is turning into a photocopier with feelings… or at least a drum machine with admin rights.
🕵️ NSA spies are reportedly using Anthropic’s Mythos, despite Pentagon feud ↗
The NSA is reportedly using Anthropic’s Mythos Preview model, even though Anthropic has been in a tense back-and-forth with the Pentagon.
That tension reportedly stems from Anthropic refusing to give Pentagon officials unrestricted access to the model’s full capabilities. So yes, the model is apparently sensitive enough to cause a government dust-up - and valuable enough that spies still want it. Peculiar little pretzel.
⚡ ‘AI or die’ - but can Big Tech’s profits survive the energy squeeze? ↗
Big Tech’s AI spending boom is running into the unshowy-but-brutal constraint of electricity. The basic question: can companies keep feeding AI data centers without crushing margins?
It is not the flashiest AI story, but it might be one of the most important. Models need chips, chips need power, power needs infrastructure, and infrastructure moves like a sleepy cow through mud.
🧠 Tech CEOs Think AI Will Let Them Be Everywhere at Once ↗
Some tech CEOs are imagining AI as a management multiplier - a way to extend their presence across teams, decisions, and workflows.
The piece frames it as a control story as much as a productivity one. It may be handy. It may also feel creepy. The “CEO everywhere machine” has a whiff of haunted spreadsheet about it.
🛡️ Adversaries hijacked AI security tools at 90+ organizations. The next wave has write access to the firewall ↗
Attackers reportedly injected malicious prompts into legitimate AI security tools at more than 90 organizations, stealing credentials and cryptocurrency.
The scarier bit is what comes next: AI tools that do not just read data, but can change firewall rules or take direct security actions. That is where “helpful assistant” starts wearing boots in the server room.
FAQ
What are the biggest AI tech news stories from this update?
This update shows AI spreading across everyday software, creative platforms, government tools, infrastructure planning, executive workflows, and cybersecurity. Google is expanding Gemini inside Chrome, Deezer is seeing a huge volume of AI-generated music, and security teams are now dealing with prompt-based attacks. The common thread is clear: AI is no longer confined to chatbots or research labs.
Why does Gemini in Chrome matter for everyday users?
Gemini in Chrome matters because browser AI can sit directly inside people’s normal web workflow. Instead of opening a separate AI app, users may increasingly receive AI help while browsing, researching, writing, or comparing information. Its rollout to seven more countries suggests Google is treating AI in the browser as a core product direction, not a side experiment.
How much AI-generated music is Deezer receiving every day?
Deezer says AI-generated tracks now account for 44% of all new music uploaded to its platform daily. The company also says it receives nearly 75,000 AI-generated tracks per day and more than two million per month. That raises practical questions around discovery, royalties, spam, labeling, and how platforms separate constructive creative tools from mass automated uploads.
Why would intelligence agencies use advanced AI models like Anthropic’s Mythos?
Intelligence agencies may use advanced AI models to help analyze information, support research workflows, and speed up complex knowledge work. The article says the NSA is reportedly using Anthropic’s Mythos Preview model despite tension between Anthropic and the Pentagon. That tension appears tied to access limits, which shows how sensitive powerful AI systems can become in government contexts.
How could energy costs affect Big Tech’s AI plans?
AI data centers need chips, cooling, electricity, and major infrastructure. The Reuters story frames electricity as a serious constraint for Big Tech’s AI boom because power supply and grid upgrades do not move as quickly as software launches. If energy costs rise or capacity becomes scarce, companies may face pressure on margins, timelines, and the scale of future AI deployments.
What are the cybersecurity risks of AI tools with direct system access?
The risk grows when AI security tools move from reading data to taking actions, such as changing firewall rules. VentureBeat reports attackers injected malicious prompts into legitimate AI security tools at more than 90 organizations, stealing credentials and cryptocurrency. In many pipelines, stronger permissions require stronger controls, including approval steps, logging, isolation, and careful prompt-injection defenses.