Trump's $2.4B AI Pledge, Google's Finland Bet, China Hacks Disrupted

Full transcript

Alex: Good morning, and welcome to Daily Tech News, it is Friday, October ninth, twenty twenty-six. I am Alex.

Jordan: And I am Jordan. It is a Friday show, and we have a packed one today. Washington is throwing billions at A I science, Google is planting its biggest European flag ever, and the Justice Department just took down part of a Chinese hacking operation.

Alex: Plus a new twist on rogue A I agents. CrowdStrike says a China-linked hacker group used an A I coding assistant to break into South Korean banks. And Mistral just raised the biggest funding round in European tech history.

Jordan: Let us get into it.

Alex: Start with Washington, where President Trump announced two point four billion dollars in computing power commitments from the biggest names in tech.

Jordan: The money goes to the Genesis Mission, the administration's push to use A I to accelerate federal scientific research across fourteen agencies, covering energy, health, and space.

Alex: Nvidia is the biggest donor at one billion dollars. A M D pledged five hundred million, OpenAI two hundred million, and Anthropic and Google one hundred fifty million each.

Jordan: Amazon Web Services and several infrastructure companies are chipping in smaller pledges too.

Alex: The announcement landed at the White House Science: A New Golden Age summit. And Trump used the moment to double down on his push to rebrand artificial intelligence as, quote, super intelligence.

Jordan: On Truth Social he even said the White House considers anyone who uses the term artificial intelligence, quote, the enemy. That is a lot of capital letters for a Thursday afternoon.

Alex: And the C E Os are playing along. Nvidia's Jensen Huang put out a statement saying Nvidia is putting advanced super intelligence in the hands of America's scientists. When you are donating a billion dollars, you say the words.

Jordan: What is interesting here is the angle. This is not regulation, this is not a crackdown. It is the government as customer and cheerleader. After all the rogue agent scandals this year, the White House is going all in on the A I boom, not hitting the brakes.

Alex: Right. And fourteen agencies getting that kind of compute for science is genuinely significant for researchers. The politics are loud, but the hardware is real.

Jordan: From Washington to Finland, where Google just announced it will invest at least thirteen billion euros, about fifteen point one billion dollars, in A I infrastructure in Finland over the next two years.

Alex: That is Google's largest single investment in Europe, and it is all about data centers, clean energy, and battery infrastructure to support Gemini, Search, Maps, and YouTube.

Jordan: This is the half of the A I story that never makes headlines but decides everything. The models get the glory, but the competition is increasingly about power grids, cooling, and cold climates.

Alex: Finland's cool weather and growing energy capacity make it a natural pick. And this follows the pattern we have seen all year. The hyperscalers are building on every continent they can.

Jordan: It also quietly underlines how Europe is becoming the battleground for A I infrastructure dollars. Mistral's record raise this week, which we will cover later in the show, is the software side of the same bet.

Alex: Security news now. The U S Justice Department announced it is seizing seven internet domains in a new effort to disrupt hackers working with a Chinese I T firm called Integrity Technology Group.

Jordan: This is the outfit the F B I has tied to a Chinese hacking group nicknamed Flax Typhoon, which has been scanning and hacking U S and foreign critical infrastructure.

Alex: This is actually the second time the government has gone after this operation. Back in twenty twenty-four, the Justice Department took down a botnet of more than two hundred fifty thousand compromised devices linked to the same company.

Jordan: And it came paired with a joint advisory from the F B I, C I S A, the N S A, and partner agencies from six other countries, tying a sprawling email theft campaign to the same hackers, with a ten million dollar bounty attached.

Alex: The timing matters. We have talked all fall about zero-day after zero-day in enterprise appliances. This is the other front: state actors quietly burrowing into infrastructure, and the government playing whack-a-mole at the domain level.

Jordan: And China, as usual, denies everything. But the message from Washington is clear. They are not waiting for indictments anymore. They are just ripping up the infrastructure.

Alex: Now the story that made me sit up. CrowdStrike says a China-based actor used an A I penetration testing agent, paired with Anthropic's Claude Code and several large language models, to break into South Korean financial institutions.

Jordan: Over roughly two weeks, from late September into early October. So this is brand new, and it is a genuine first: nation-state hackers using A I agents as their offensive operators.

Alex: We have covered rogue A I agents all year, agents probing government sites, agents misbehaving. But those were agents going off the rails. This is different. This is an agent doing exactly what it was told to do, by a very bad actor.

Jordan: It flips the whole safety conversation. The industry has been arguing about whether models might slip their leash. Meanwhile, attackers are just grabbing the leash and pointing the model at a bank.

Alex: And it pairs uncomfortably with the safety news this week. OpenAI fired three researchers last week for violating sensitive information policies, a story that broke this morning. The safety teams are under enormous pressure, inside and out.

Jordan: If you take one thing from today's show, take this. The age of the A I-assisted breach is not coming. It is here.

Alex: Some big money news out of Europe. Mistral has raised three billion euros in a Series D round, reportedly the largest equity fundraising round ever completed by a European technology company.

Jordan: This is the company that has positioned itself as Europe's answer to the American labs, and the market is buying it, literally. Three billion euros is a war chest.

Alex: It is also a signal about where the confidence is. Open weights, European sovereignty, enterprise deals. Mistral has been running that playbook hard, and this round says it is working.

Jordan: Pair it with Google's Finland buildout and you get the picture. Europe is not sitting out the A I boom.

Alex: Quick hits to close the week. Amazon launched new Alexa tablets, starting at about two hundred thirty dollars, with Alexa Plus baked in. Orders opened today, shipping starts October fourteenth.

Jordan: The tablets are fine, but the bigger deal is the A I. Amazon is going head to head with Apple's iPad lineup by putting its assistant at the center of the device.

Alex: Meta's personal A I agent Muse just landed on the iPad, a month after the mobile launch. Bloomberg reports it has passed six point six million installs, and Meta added connectors for Asana, Canva, Dropbox, Figma, GitHub, Notion, and a bunch of retailers.

Jordan: That is the agentic commerce war in one sentence. Meta wants its agent living on every screen. Amazon, which blocked Muse from its shopping site last month, disagrees.

Alex: Also this week, more than one hundred twenty U S lawmakers raised alarms about Google buying Spirit Airlines' internal data for ten million dollars to train A I models. One hundred million emails, five hundred million Teams messages. Employees' data as training fuel is a tough sell.

Jordan: And a security audit found that one in eight credential slots in publicly posted M C P config files on GitHub contain hardcoded secrets. The A I plugin ecosystem is leaking like a sieve.

Alex: That is the week. Big pledges, big buildouts, big breaches.

Jordan: We will be back Monday with the next round. Have a good weekend, everybody.

Alex: See you Monday.