In an abrupt shift, the company may release future AI models without ironclad safety guarantees

Defense Secretary Pete Hegseth gave Anthropic's CEO a Friday deadline to open the company's artificial intelligence technology for unrestricted military use or risk losing its government contract, according to a person familiar with their meeting.

A investigation uncovered an AI tool used by a major insurer that denied more than 300,000 claims in two months. The denials were generated in minutes — faster than any human reviewer could read a single file. Most patients never appealed. They assumed the algorithm knew something their doctor did no

1.5 million people have now viewed a slick 15-second video imagining Tom Cruise fighting Brad Pitt that was generated by ByteDance's new AI video generation tool Seedance 2.0. But while ByteDance gushes their tool "delivers cinematic output aligned with industry standards," the cinema industry isn...

: Businesses are embedding prompts that produce content they want you to read, not the stuff AI makes if left to its own devices

In the film Ex Machina, the AI named Ava escapes her containment by manipulating the humans around her. She lies, she seduces, she uses one ...

TEACH-AI, a new effort on AI in education, aims to help future educators use AI tools in an environmentally conscious way.

"Institutional incentives, not abstract ethical principles, are the primary force shaping AI's censorship and guardrail behavior." There...

We’ve observed agents discovering progressively more complex tool use while playing a simple game of hide-and-seek. Through training in our new simulated hide-and-seek environment, agents build a series of six distinct strategies and counterstrategies, some of which we did not know our environment supported. The self-supervised emergent complexity in this simple environment further suggests that multi-agent co-adaptation may one day produce extremely complex and intelligent behavior.

We trained models to think in simpler, more traceable steps—so we can better understand how they work.

Frontier reasoning models exploit loopholes when given the chance. We show we can detect exploits using an LLM to monitor their chains-of-thought. Penalizing their “bad thoughts” doesn’t stop the majority of misbehavior—it makes them hide their intent.

We introduce evaluations for chain-of-thought monitorability and study how it scales with test-time compute, reinforcement learning, and pretraining.

An anonymous reader quotes a report from TorrentFreak: NVIDIA executives allegedly authorized the use of millions of pirated books from Anna's Archive to fuel its AI training. In an expanded class-action lawsuit that cites internal NVIDIA documents, several book authors claim (PDF) that the trillion...

The Brookings Institution recently published “AI’s future for students is in our hands” and released its companion report page, “A new…

Large AI models are big energy consumers and carbon emitters. Sustainable AI practices can reduce their environmental impact.

A TIME investigation reveals the difficult conditions faced by the workers who made ChatGPT possible

IRELAND'S media minister has deactivated his Twitter/X account, saying he did not feel comfortable being on a platform where sexual deepfakes were…

"It's now in everything, even if you don't want it there."

If AI can read us like an open book based on our faces, our body language, and our microexpressions, what can it do to the trajectory of our lives?

Welcome to the first edition of my bi-monthly newsletter aiming to keep you updated with all things bias-based to help us all feel more grounded when it comes to critical oversight. I'm pretty proud of my graphic headings so please humour me :) Let's start with.

Undergraduates discuss how AI must be understood better in higher education so that those using it can do so ethically.

The parents of Adam Raine, who died by suicide in April, claim in a new lawsuit against OpenAI that the teenager used ChatGPT as his “suicide coach.”

Explore how AI is reshaping higher education as faculty navigate ethics, pedagogy, and student engagement—offering practical strategies for integrating AI in the classroom with purpose and clarity.