The News About The News (106)

How to see the forest, not the trees.

May 26, 2025

I tried representing an AI avatar presenting the news, talking about “humans”. Not sure if we got there.

This week, like virtually all weeks before it, has seen an explosion of groundbreaking releases from the large AI labs.

In the last seven days, OpenAI launched Codex, an agentic coding assistant and announced a $6.5B investment in an unreleased hardware product, io. Google released Gemini 2.5 Pro, an advanced reasoning model, Veo 3, a next-generation video generation tool for cinematic-quality AI video and Gemini Diffusion, a near-instantaneous language model. Anthropic announced Claude 4 with extended thinking and much better coding abilities. Microsoft unveiled a major Copilot update enabling multiple AI agents to collaborate on coding tasks. Not to mention the thousands of smaller releases, papers and developments taking place worldwide.

My browser tabs are overflowing with thought pieces, fascinating demos and insightful takes about the unleashing of AI capabilities. This has been the case for a few years now, and the sense of ever achieving a “knoweldge parity” has been completely discarded for me.

None of us will ever be able to catch up.

The best you can do is learn some of the fundamentals of machine learning, keeping a finger on the pulse of releases, and introduce as much AI experimentation as possible into your life. Being a reader of this newsletter evidently helps.

To me, the “news behind the news” is the deep realization that despite how disoriented or overwhelmed you might currently be feeling, there might be some respite in realizing that this is the slowest things will ever be.

I hope that brings some of you solace.

Until next week,
MZ

PS. This Thursday 29 May, I’m hosting a Signals demo for foresight practitioners. Join us!

PS2. I’m in London next week – DM if you’re around and up for a coffee.

How Does Claude 4 Think? (2h30)

Sholto Douglas and Trenton Bricken discuss how Claude 4 approaches problem-solving, its alignment with human values, and the challenges faced in creating AI that can think and act responsibly.

Transparency in AI decision-making builds trust and fosters better human-AI collaboration.

AI and the Illusion of Self (90m)

Philosopher and AI researcher Murray Shanahan explores the implications of AI consciousness through the lenses of Buddhism, Wittgenstein, and cognitive science. Rather than debating if machines are conscious in a metaphysical sense, Shanahan emphasizes AI as a mirror for human selfhood.

Human consciousness is constrained to a subject-object dualism and that AI consciousness might not be.

AI Ethics (55 min)

Noema Magazine brought together renowned AI experts Blaise Agüera y Arcas and Prof. Shannon Vallor to discuss whether AI makes us more or less human in a conversation moderated by journalist Kara Swisher.