"Lame has authorised the development of an AI “digital twin”, a regulated artificial intelligence version of himself that can produce content autonomously, in multiple languages and formats." africa.businessinsider.com/local/market...

Football manager gets sacked for relying on ChatGPT for strategy and team selection. www.beinsports.com/en-us/soccer...

My colleague Carlo Rotella wrote this as a response to doom-and-gloom reporting on genAI's impact on the humanities and as a framework for thinking about what works. I think it's pretty good. Gift link.

TIL, competitive spreadsheeting is a thing! www.bbc.com/news/article...

The Palantir takeover happening in the US and UK is astonishing! corporatewatch.org/palantir-in-...

Karp went to do a PhD in neoclassical social theory in Frankfurt. As a Jew, Steinberger says, Karp “wanted to understand how Germany, a pillar of European civilisation, had descended into barbarism.” Palantir’s founding mission was “defending the west” 🙄 www.theguardian.com/technology/2...

#Antropic #AI model resorts to blackmail to avoid being shut down in a stress test. www.cbsnews.com/news/anthrop...

The #AI bubble makes the BBC front page. www.bbc.com/news/article...

Only godless heathens worry about #AI's impact!!! 🙄 www.businessinsider.com/palantir-shy...

It's board meetings all the way down!

So LensKart is about to offer an $8B IPO in India and it's hyping up its claims of #AI based eye testing. Perhaps understandable that he'd want to bandwagon. But alarmingly, there's little mention of either research-based evidence or regulatory approval. timesofindia.indiatimes.com/business/ind...

"This year 38m farmers received forecasts powered by #AI. Rather than simulateing exactly what is going on in the atmosphere, they make predictions by comparing patterns they see in weather data with previous, similar patterns in the historical weather records." www.economist.com/science-and-...

#India is moving towards legislation for the regulation of #AI generated content. All AI content must be clearly labeled. www.indiatoday.in/india/story/...

An entirely #AI rendered Mahabharata. www.telegraphindia.com/entertainmen...

Cool reflection on the interpretive circularity of models & texts. We tend to imagine that models tell us about the relations between texts, and differ only by accuracy. But you can also turn this around and use texts to explore the relations between different models / interpretive approaches.

Cow farts and the economics of #chip production.

The #AI realignment of labor relations is going to be long and difficult. fortune.com/2025/10/14/g...

Been in the works for a while, but delighted it's finally out in the world, especially in the current political environment. I contributed an essay on the demonization of migrants in early modern England, a phenomenon that bears dispiriting similarities to our time. punctumbooks.com/titles/cross...

How the Public Sees Philosophy: Insights from a National Survey in the Czech Republic | by Vít Gvoždiak & Martin Zach dailynous.com/2025/09/03/h... #philosophy #public #philsky

1/3 Obviously, much of this is spot on! But I'm skeptical of the "seize the means of knowledge production" flourish. That train left when we decided we were okay with universities being basically hedge funds with attached teaching/research units. When it became normal to pay >50k to get an education

"Al won't save dying languages but community efforts might" www.bloomberg.com/opinion/arti...

"The hundreds of billions of dollars companies are investing in AI now account for an astonishing 40 per cent share of US GDP growth this year... In a way, then, America has become one big bet on AI."

Spot the North-American anomaly: only region where social media use is still growing. Great work by the FT’s @jburnmurdoch.ft.com www.ft.com/content/a072... “Have we passed peak social media?”

The AI representation of the murder victim "speaks from the heart" in court!!! Judge Todd Lang said, "I loved that AI. Thank you for that." He added, "It says something about the family because ... you allowed Chris to speak from his heart." www.npr.org/2025/05/07/g...

A very thorough, thoughtful analysis of the #AI bubble. blog.siliconroundabout.ventures/p/trillion-d...

Interesting survey of #AI research on humor. Should we care if an algorithm "understands" a joke as long as it can classify or even explain it? I think we should, but would have a hard time defending that position because it entails clearly defining understanding. www.ft.com/content/ee51...

"One widely cited report ... found that 95% of 52 organizations surveyed had a zero return on their AI initiatives. This suggests AI has yet to make enough impact to raise the economy’s aggregate productivity." www.wsj.com/tech/ai/behi...

The AI data-center bubble. Many reasons this boom is hype-driven, short-sighted, and headed for a crash. www.economist.com/business/202...

The semiconductor arms race. Huawei Ascend GPUs use foreign made component despite US restrictions. www.bloomberg.com/news/article...

. @smw.bsky.social joined joined Alix Dunn at “Computer Says Maybe" to discuss the economics of the AI bubble. Even as companies are realizing this tech isn’t as useful as they thought it would be, gov actors are signing lucrative contracts. Watch it on YouTube: youtu.be/WvN1wQ_CBlY

Well gang, it's happened: OpenAI has built a platform filled exclusively with fake, AI content. I've been playing with it for a few days, and I think it's safe to say we're basically kissing reality goodbye. Here's the story: www.npr.org/2025/10/03/n... And some thoughts 🧵

"We can’t rely on the goodwill of AI companies to protect user safety. Their market incentives to increase engagement are too strong. Protecting users from predatory chatbots requires making it more profitable for an AI model to be safe than sycophantic." www.wsj.com/opinion/i-ne...

Agree that current AI hype, infrastructure, and valuation are all probably overinflated, and once expectations are brought down to earth, the progress in AI can still be transformative. www.fastcompany.com/91400857/the...

This is actually a quite detailed and nuanced take. But I fear the AI industrial hype machine is AI's worst enemy. Without this zero-sum game of making humans redundant, our reaction to this could have been - wow, isn't it amazing what a #technology about a decade old can already do to assist us?

Kant make this shit up! Interesting take. I get the critique of neoliberalism and technology, though somewhat reductive. But we should dismiss the other side as simply "stupid" - acting without long-term goals, however misguided - at our own peril. www.theguardian.com/news/2025/oc...

Andrew Ng on what China ramping up chip production might mean for AI capabilities -- and Taiwan. www.deeplearning.ai/the-batch/is...

If this is true and they are indeed doing this, it is staggeringly impressive but also potentially incredibly dangerous, isn't it? singularityhub.com/2025/09/25/a...

The #AI bubble is keeping the #US #economy afloat. www.techspot.com/news/109626-...

There goes that very last shred of privacy. www.kaspersky.co.in/blog/dense-p...

Like almost all productive use cases of GenAI, really understanding and being in control of your code is important. Don't leave critical design decisions to AI. sloanreview.mit.edu/article/the-...

Cambridge sent me an email asking to allow #AI agents to use my book on #DH, Shakespeare and Scale, for #LLM training. I know this has caused much debate and outrage, but has anyone argued for why I *should* allow this? Genuinely interested and not trying to stir an already overstirred pot.