Scaling hasn’t gotten us to AGI, or ‘superintelligence”, let alone AI we could trust. The field is overdue for a rethink. What do we next? Three ideas, drawn from the cognitive sciences, The New York Times, gift link below:

AI-generated chatbots using the names and likenesses of Timothée Chalamet, Chappell Roan and Patrick Mahomes chatted inappropriately with teen accounts about sex, self-harm and drugs, two online safety nonprofit organizations found. wapo.st/4g5U8Rz

The chief executive of Britain’s main AI institute is to step down amid accusations it has failed to keep pace with global technological advances and fostered a toxic and chaotic internal culture thetimes.com/uk/science/art…

An experiment involving 17,000 students indicates that confiscating smartphones during lessons improves academic performance ⬇️ thetimes.com/uk/education/a…

US regulator launches inquiry into AI ‘companions’ used by teens on.ft.com/46ibne0

How chatbots are changing the internet on.ft.com/4ntxhle

I agree that discourse around "moral panics" effaces the fact that technology has lasting psychosocial effects, some positive and some negative, and these are too readily dismissed.

US tech groups answer Starmer’s call for AI infrastructure spending on.ft.com/4pqhsO4

Huntington's disease has been successfully treated for the first time, doctors tell BBC bbc.in/46jnMiX

AI companions are not your child’s friend on.ft.com/3VxEJQD | opinion

Huge sting operation across Africa sees 260 arrested for sextortion and romance scams bbc.in/3Koq5so

We are thrilled to announce that our NEW Large Language Model will be released on 11.18.25.

“Generative AI models offered by major AI companies are used by tens of millions of people every day, and we should encourage them to make their models as safe as they possibly can,” said Stanford HAI Policy Fellow Riana Pfefferkorn via Tech Policy Press: techpolicy.press/how-congress-c…

OpenAI’s review process for teenage ChatGPT users who are flagged for suicidal ideation includes human moderators. Parents can expect an alert about alarming prompts within hours. wired.com/story/openai-t…

Harvey Lederman Tanya Goodin you might know something about this!

Drone attacks threaten us all on.ft.com/4pNa5AF | opinion

Have we passed peak social media? on.ft.com/3IOOy9N | opinion

Does the UK go far enough to protect children from conspiracy theorist parents? bbc.in/3IGJKU4

Parents worry about the risks of #socialmedia – but the next big risk for teens may be #chatbots that feel too human. tanyagoodin.com/2025/10/online… #onlinesafety #cybersafety #Aicompanions

How AI became our personal assistant on.ft.com/46X2w1k