The Steady State | October 16, 2025

YouTube video by TVE Espírito Santo

An intensive international study was coordinated by the European Broadcasting Union (EBU) and led by the BBC

Canonical today announced optimized inference snaps, a new way to deploy AI models on Ubuntu devices. Install a well-known model like DeepSeek R1 or Qwen 2.5 VL with a single command, and get the sili...

The first-person shooter is famous for being ported to almost every computer on the planet. Now, we know it can run off the Earth as well.

Discuss, discover, and read arXiv papers.

OpenAI researchers recently claimed a major math breakthrough on X, but quickly walked it back after criticism from the community, including Deepmind CEO Demis Hassabis, who called out the sloppy comm...

: Redmond argued schools, education authorities are responsible for GDPR

Today’s world requires us to make complex and nuanced decisions about our digital security. Evaluating when to use a secure messaging app like Signal or WhatsApp, which passwords to store on your smartphone, or what to share on social media requires us to assess risks and make judgments accordingly. Arriving at any conclusion is an exercise in threat modeling. In security, threat modeling is the process of determining what security measures make sense in your particular situation. It’s a way to think about potential risks, possible defenses, and the costs of both. It’s how experts avoid being distracted by irrelevant risks or overburdened by undue costs...

MIT researcher Kalyan Veeramachaneni describes the pros and cons of using synthetic data, which are artificially generated by algorithms, to build and test AI applications and train machine-learning m...

Researchers argue over whether ‘novel’ AI-generated works use others’ ideas without credit.

Continuous exposure to AI might reduce the ADR of standard non-AI assisted colonoscopy, suggesting a negative effect on endoscopist behaviour.

It’s the most transparent estimate yet from one of the big AI companies, and a long-awaited peek behind the curtain for researchers.

It’s the most transparent estimate yet from one of the big AI companies, and a long-awaited peek behind the curtain for researchers.

According to official mining output records, Peru mined about 90 tonnes of gold in 2023, far ahead of any other South American country. That puts it within the world’s top 15 producers, just below the...

MIT researchers used sparse autoencoders to shed light on the inner workings of protein language models, an advance that could streamline the process of identifying new drugs or vaccine targets.

Automated online conversations made by text classifiers are becoming more prevalent. Now, an MIT team led by Kalyan Veeramachaneni has come up with an innovative approach to not only measuring how well these classifiers are doing their job, but going one step further to make them more accurate.

YouTube video by Fireship

The new science of “emergent misalignment” explores how PG-13 training data — insecure code, superstitious numbers or even extreme-sports advice — can open the door to AI’s dark side.