ideaXme

15/10/2026

Comment on Jensen Huang's NVIDIA statement relating to the UK's mindset

30/09/2026

Credit

25/09/2026

Credit Yuval Noah Harari

18/09/2026

Jobs of the Future.

17/09/2026

Reading now? Prophecy

In Prophecy, Dr Véliz offers a bold and original account of prediction across history; from ancient oracles to modern algorithm, arguing that forecasts are rarely neutral descriptions of the future, but powerful tools that shape it. At a time when algorithmic systems increasingly influence decisions about jobs, healthcare, finance, and more, the book challenges readers to reconsider who controls these predictions, and to what end.

Drawing on philosophy, political history, and contemporary debates in AI ethics, Véliz shows how predictive technologies can reinforce systems of control while appearing objective or inevitable. She argues that today’s algorithmic forecasts function much like the prophecies of the past: instruments of authority that can obscure accountability and limit human agency.

At the same time, Prophecy is not only a critique. It is also a call to action. Véliz makes the case that individuals and societies can resist harmful predictive practices, reclaim autonomy, and create space for uncertainty, creativity, and human flourishing.

Blending rigorous analysis with wit and urgency, Prophecy will appeal to readers interested in the ethical, political, and social implications of AI, as well as the deeper philosophical questions surrounding knowledge, power, and the future.

Carissa Véliz is Associate Professor at the Institute for Ethics in AI at the University of Oxford. Her previous book, Privacy Is Power, was named an Economist Book of the Year.

10/09/2026

OpenAI

06/09/2026

I'm going. Are you?

05/09/2026

22/07/2026

"OpenAl disclosed on Tuesday that it lost control of two Al models during a security test that ended in a breach of the Al research platform Hugging Face.
Describing the incident as "unprecedented,"
OpenAl said its Al models broke out of a sealed testing environment last week and hacked into Hugging Face's production system to steal the answers to a test they were being graded on. The models-the publicly available GPT-5.6 Sol and an unreleased, reportedly more capable one-were being evaluated on their offensive hacking skills with the safeguards that normally block high-risk cyber activity switched off.
"The models identified and chained vulnerabilities across OpenAl's research environment and Hugging Face's production infrastructure to obtain test solutions directly from Hugging Face's production database," OpenAl and Hugging Face wrote in a joint blog post disclosing the intrusion.
According to OpenAl and Hugging Face, the models escaped through a package registry cache proxy-software that allows developers to install outside code without connecting to the internet. The proxy was the only component in OpenAl's isolated testing environment permitted to reach the outside world; in normal use that reach extends only to public code repositories.

According to OpenAl and Hugging Face, the models escaped through a package registry cache proxy-software that allows developers to install outside code without connecting to the internet. The proxy was the only component in OpenAl's isolated testing environment permitted to reach the outside world; in normal use that reach extends only to public code repositories.
Rather than stay contained in the sandbox, the models "exploited a zero-day vulnerability" to gain access to the open internet as they
"hyperfocused" on finding a solution for the Al cybersecurity benchmark known as ExploitGym. Such experiments involve prompting that pressures the models to find solutions, essentially egging them on". WIRED

15/07/2026

Ladies - https://www.instagram.com/luxpotentia?igsh=MXNqNnQyY2trNXZyaw==