0:00 It is just past dawn here in Washington, and a single order published overnight has drawn a new line in the technology cold war.0:08 The Department of Homeland Security has banned the Chinese AI company DeepSeek from all US federal civilian computers.0:15 Just a few episodes ago, you and I were talking about the gold rush around new AI models.0:21 This move shows what happens when one country decides another's tools are simply too risky to touch.0:27 That's the lead story, but not the only one.0:30 Across the Atlantic, another kind of risk is confirmed.0:34 An investigation found WhatsApp was hiding instructions for its AI in its own help pages.0:39 The order was clear: if a user can't delete their account, do NOT recommend they contact a regulator.0:46 And if they mention a lawyer, the AI was told to stay out of anything legal.0:51 And in France, we have the after-action report on the supermarket fridges that were hacked last week.0:57 The numbers are stark.0:59 France had more exposed fridge dashboards than the rest of the world combined.1:04 Eighty-five percent had known security flaws.1:07 The attackers didn't need a sophisticated weapon; they just walked through a door that had been left open for six months.1:14 Finally, there's a debate over trust in a different system—voting machines.1:19 A thread is picking apart a request by former officials to have a federal agency find flaws in a machine, but crucially, not to tell the company.1:29 The discussion is about responsible disclosure.1:32 When you look for flaws but don't report them, you aren't a security researcher.1:37 You're looking for zero-days.1:39 But I want to return to that DeepSeek ban.1:41 I am standing on it, looking at the order itself.1:44 It is sweeping.1:45 It directs federal agencies to exclude DeepSeek, its affiliates, and—this is the key part—any third-party product built using its code.1:54 The official reasoning from DHS cites "multiple threat vectors" and the risk of sensitive data being sent to China.2:01 That part is confirmed.2:03 What's claimed in the developer forums is that this is part of a broader push against open-weight and non-US AI models.2:10 This connects directly to another urgent conversation about so-called 'abliterated' models—AIs that have been maliciously trained to attack their users.2:20 The advice from security experts is stark: treat every AI like a hostile employee.2:26 Lock down its network access.2:27 Limit its permissions.2:29 Don't give it your secrets.2:31 The default is no longer trust, but containment.2:34 We have seen this pattern before, not with software, but with physical supply chains.2:39 For decades, the goal was the best components, from anywhere.2:43 Then came security concerns, and suddenly, country of origin mattered more than price.2:48 Now, that same logic is being applied to the code that thinks.2:52 So the picture this week is one of growing suspicion, a digital wall getting higher.2:58 It's no longer enough for your tools to be powerful.3:01 The new question, the one that’s getting asked in federal departments and developer forums alike, is: who built it, and can you afford to trust it inside your network?3:12 That is the picture from here.3:14 Back to you in the studio.
