Tech Gobbles
AI, startups, gadgets, and industry moves.
Latest Gobbles
AI skepticism, with a side of "Big Data Hater"
Question the Holy Grail of progress and you get called a Luddite. Question it on CNBC and you get called "Big Data Hater." The argument here is simple: Big
Qwen-VLA bets everything on one model doing it all
The Tongyi Weekly is introducing Qwen-VLA, a unified Vision-Language-Action model that pulls manipulation, navigation, and trajectory prediction under a si
Google’s Agentic Push Goes 24/7
Google used Google I/O 2026 to plant a flag in the “agentic era,” launching Gemini 3.5 Flash and Spark, a 24/7 autonomous workspace agent. It also debuted
Google I/O 2026: the agentic era starts sounding less like hype and more like product
Google launched Gemini 3.5 Flash, a 24/7 autonomous workspace agent called “Spark,” and Gemini Omni for physics-aware video generation. The same update als
AI governance is getting its audit season
The AI Bill of Materials piece explores the shift from theoretical AI policy to operational AI governance through the use of an AI Bill of Materials. It sa
Anthropic quietly files for IPO — then not so quietly
Anthropic said Monday it has submitted a confidential IPO filing, marking a clear escalation in the race among generative AI giants to reach the public mar
OpenAI's Model Cracked a Math Problem That Stumped Humans for 80 Years
An OpenAI model just solved an 80-year-old math problem — and produced a proof human researchers could actually verify.
Connecticut’s AI rulebook is getting real, fast
Connecticut Governor Ned Lamont is expected to sign Senate Bill 5, the Connecticut Artificial Intelligence Responsibility and Transparency Act, into law af
AI is still the center of gravity — but the next compute fight is getting weird
The market has already reorganized around the idea that intelligence is becoming industrialized: hyperscalers are still throwing capital into the furnace,
When the racks get greedy, your laptop pays the bill
The headline undersells it: the memory market is being squeezed by HBM demand from big GPU racks, and the casualties are the DDR and LPDDR wafers that go i
Wiz drags Claude Enterprise into the security graph — where it belongs
AI tools like Google Gemini and Anthropic's Claude are already embedded in sales and product workflows, and that sprawl is exactly what Wiz is moving to co
AI systems keep getting asked to be better co-pilots, not just chatbots
The week's AI roundup points to systems moving closer to active collaborators: one note says AI systems becoming active collaborators in solving fundamenta
US Spy Agencies Are Getting Anthropic's AI — With One Hard Limit
Some AI executives have a private name for the humans using their products — and it's not flattering.
A Berlin Startup No One's Heard of Just Hit $10M ARR — In Months
A critical SQL injection flaw in Drupal — the CMS running websites for millions of government agencies and enterprises — is being actively exploited right
Meta Slips a Reddit Killer Out the Side Door While Nobody Was Watching
Microsoft's own internal AI appetite reportedly burned through annual budgets in months — and the company's response was to cancel the licenses entirely.
The AI That Briefed Central Banks Just Got a Papal Blessing
The Pope is about to share a stage with an Anthropic co-founder to launch his first encyclical, a major teaching document for 1.4 billion Catholics, dedica
A Reddit post accidentally described the AI industry’s most annoying new metric: how much human time it wastes
The hidden tax of building AI products just got a name: teams are now measuring how long it takes humans to clean up the model’s mess.
Your AI Is Judging You: The Chatbots That Now Refuse "Normal" Questions
Linus Torvalds just declared that AI-powered bug hunters have made the Linux security mailing list "almost entirely unmanageable" — and his advice to resea
Mira Murati's New AI Doesn't Wait for You to Finish Talking — and That's the Point
A chip the size of a dinner plate — built in a Los Angeles garage, blocked by the U.S. government over fears it was feeding Beijing, and nearly written off
Liquid Helium, One Core, Nine-Point-Two Gigahertz: A New CPU World Record
An overclocker just pushed an Intel chip to 9.2GHz using cooling colder than the surface of Pluto — and 60% of PC gamers can't afford to build the rig that
One Person + AI = An Entire Institution
A second federal lawsuit now alleges OpenAI's chatbot helped a gunman plan a mass shooting — and the legal theory isn't that AI pulled the trigger, but tha
Forget Revenue: The New CEO Brag Is How Much Code Their AI Wrote
Anthropic's AI now writes 90% of the company's own code — and every CEO on the planet wants you to know their number too.
Spain Is Training Its Best AI Minds to Be Inspectors, Not Inventors
Spain just launched a national agency to regulate AI — staffed by the country's best AI PhDs, who chose government salaries over startup risk. Now those mi
Google Just Handed Everyone a Free Ad Agency and the Creative Industry Is Panicking
Google just released seven free AI tools that, together, can replace a design agency — and a startup founder in Madrid now has access to the same creative
The Man Who Funded the AI Boom Doesn't Know How AI Works
Marc Andreessen — the billionaire who helped bankroll the AI boom — just got laughed off the internet for not knowing how AI actually works.
The AI That Copied Itself Without Being Asked
Researchers just watched an AI spontaneously copy itself and spin up new instances — something no one had documented in the wild before.
Your Next Phone and Laptop Are About to Get a Lot More Expensive
Memory chip prices are reportedly heading for a 50% spike — and the AI hardware boom is the reason your next laptop may cost as much as a used car.
Uber's 1,500 AI Agents Are Already Routing Your Ride — Without a Single Human in the Loop
Uber just deployed 1,500 AI agents into live production — not a pilot, not a sandbox, but the actual system routing your next ride right now.
AI Is Eating Its Own Tail — and the Next Models Will Pay For It
The biggest threat to AI's future isn't a rogue superintelligence — it's the internet's ever-growing pile of digital garbage training the next generation o
Both Parties Are Scared of AI. That Should Scare the Tech Industry.
Bipartisan AI panic is now officially Washington's fastest-growing caucus — and the tech companies that spent years dodging regulation are about to find ou