Posts

The Evolution and End of Traditional Hardware Journa...

Dedicated hardware journalism fundamentally transformed how consumers evaluat...

Starting reading the AI Snake Oil book online today

The book will be published on September 24

AI companies are pivoting from creating gods to buil...

Turning models into products runs into five challenges

AI existential risk probabilities are too unreliable...

How speculation gets laundered through pseudo-quantification

New paper: AI agents that matter

Rethinking AI agent benchmarking and evaluation

AI scaling myths

Scaling will run out. The question is when.

Scientists should use AI as a tool, not an oracle

How AI hype leads to flawed research that fuels more hype

AI leaderboards are no longer useful. It's time to s...

What spending $2,000 can tell us about evaluating AI agents

AI Snake Oil is now available to preorder

What artificial intelligence can do, what it can't, and how to tell the diffe...

Tech policy is only frustrating 90% of the time

That’s what makes it worthwhile

AI safety is not a model property

Trying to make an AI model that can’t be misused is like trying to make a com...

A safe harbor for AI evaluation and red teaming

An argument for legal and technical safe harbors for AI safety and trustworth...

On the Societal Impact of Open Foundation Models

Adding precision to the debate on openness in AI

Will AI transform law?

The hype is not supported by current evidence

Generative AI’s end-run around copyright won’t be re...

Output similarity is a distraction

Are open foundation models actually more risky than ...

A policy brief on open foundation models