US scientist John Jumper to leave Google DeepMind for Anthropic ↗
John Jumper, the AlphaFold co-creator, is leaving Google DeepMind for Anthropic. Big move. Neatly timed too, with AI labs still hoovering up top research talent like it’s oxygen.
AlphaFold’s protein-structure work made AI-for-science feel less like a slide deck fantasy and more like a material hammer. Anthropic hasn’t said exactly what Jumper will do there, but you don’t hire that kind of brain for office wallpaper.
Norway imposes near ban on AI in elementary school ↗
Norway is putting a near-ban on generative AI for elementary school pupils, while older children will face tighter, teacher-supervised use.
The government’s worry is pretty plain: kids may skip the unshowy-but-vital bits - reading, writing, maths - if chatbots keep handing them ladders over every small wall. Fair, maybe a bit severe, but not exactly from nowhere.
AI-generated ads should be exempt from EU transparency rules, retail association says ↗
Eurocommerce, representing retailers including Amazon, H&M, Inditex and Ikea, asked the EU to spare AI-generated ads from certain transparency labels.
Their argument: a sofa staged in an AI-made living room isn’t the same beast as a deceptive deepfake. The awkward bit? Retailers are already using AI heavily for product visuals, so disclosure rules could suddenly slap labels on a huge chunk of ordinary-looking ads.
Trump tells Axios he no longer views Anthropic as national security threat ↗
Trump said he no longer sees Anthropic as a national security threat, after previously raising alarms around foreign access to the company’s most advanced models.
The dispute centered on Anthropic disabling access to Fable 5 and Mythos 5 after pressure over foreign-national access. So, a thaw… or so it seems. The Defense Production Act still lurked in the conversation like a toolbox with teeth.
Billionaire Ambani wants AI in every call, app, and home ↗
Mukesh Ambani’s Reliance is pushing AI straight into phone calls, apps and home devices, instead of treating it like yet another standalone chatbot app.
Jio Call Agent can join calls, transcribe, summarize and handle tasks like booking cabs or ordering food. With access to more than 500 million Jio users, this is distribution-as-a-sledgehammer, not a gentle product launch.
Wall St Week Ahead Investors see Micron earnings as pulse check of AI rally momentum ↗
Investors are eyeing Micron as a test of whether the AI-chip boom still has legs, especially around memory demand and data-center spending.
The wider market has been leaning hard on AI optimism, almost like a table with one heroic leg. Micron’s results could either reinforce the “still more juice” story or make traders squint harder at the whole thing.
Over-reliance on chatbots can diminish critical-thinking skills, study finds ↗
A study reported by the Guardian found that relying too heavily on chatbots may weaken people’s ability to spot misinformation on their own.
The sharp trade-off: AI assistance improved immediate accuracy, but users got worse when later working without it. That’s the sticky bit - the tool helps now, then quietly steals a few muscles from tomorrow.
Luca Guadagnino’s Sam Altman movie dropped by Amazon after it announces OpenAI partnership ↗
Amazon dropped Artificial, Luca Guadagnino’s Sam Altman biopic, after deepening its OpenAI partnership. Hmm. Not suspicious-looking at all, naturally.
The film reportedly covers OpenAI’s boardroom drama and stars Andrew Garfield as Altman, with Ike Barinholtz as Elon Musk. Amazon said the movie would be better served elsewhere, which is polite corporate origami for “this got complicated”.
FAQ
Why is John Jumper leaving Google DeepMind for Anthropic significant?
John Jumper is best known as a co-creator of AlphaFold, the protein-structure AI system that helped make AI-for-science feel more practical, tangible and consequential. His move to Anthropic shows how fiercely major AI labs are competing for top research talent. The article does not say what role he will take, but the hire suggests Anthropic wants deeper scientific and technical expertise.
What is Norway changing about AI in elementary schools?
Norway is moving toward a near-ban on generative AI for elementary school pupils. Older students will still be allowed to use AI, but under closer teacher supervision. The concern is that younger children may lean on chatbots instead of developing core reading, writing and maths skills. The policy reflects a cautious approach to AI in early education.
Why do retailers want AI-generated ads exempt from EU transparency rules?
Eurocommerce argues that ordinary AI-generated product visuals should not be treated like deceptive deepfakes. For example, an AI-made living room used to stage a sofa is being presented as different from manipulated political or personal content. Retailers are concerned that broad transparency rules could require labels on many everyday AI-generated ads already used in ecommerce and marketing.
How could AI-generated ad rules affect online shopping?
AI-generated ad rules could make product images, staged scenes and marketing visuals more visibly labelled when AI is involved. For shoppers, that may improve transparency around what is real, edited or synthetic. For retailers, it could add compliance work and alter how ads are designed. The tension sits between clear disclosure and avoiding labels on routine product presentation.
What is Reliance trying to do with AI in India?
Reliance, led by Mukesh Ambani, is trying to embed AI directly into phone calls, apps and home devices rather than keeping it as a separate chatbot. Its Jio Call Agent can join calls, transcribe, summarise and help with tasks like booking cabs or ordering food. With more than 500 million Jio users, the strategy is built around mass distribution.
Can relying too much on chatbots weaken critical thinking?
The article cites a study suggesting that heavy chatbot reliance may reduce people’s ability to identify misinformation without help. AI assistance improved users’ immediate accuracy, but those users performed worse later when working on their own. The practical lesson is not that chatbots have no value, but that people may need to keep practising independent judgment while using them.