16: Demis Hassabis

At the moment, we are locked in an extremely intense, multilayered commercial and geopolitical race. While these competitive dynamics fuel rapid progress and accelerate the incredible upsides, advances on the frontier are outpacing our understanding of the technology. Nobody in the world knows for sure what is going to happen from here, and even the experts disagree. When there is a large degree of uncertainty and the stakes are this high, proceeding with cautious optimism is the sensible and correct strategy. That calls for public policy that promotes innovation while also incentivising responsibility and security, fosters international collaboration on key safety issues, and encourages careful consideration of how AI is deployed for the benefit of society.

The Odyssey is a 'colossal piece of cinema', critics say in rave reviews

This is not a good comparison. Einstein did theoretical physics. Musk is an entrepreneur. This is like comparing Adam Smith to FDR. https://t.co/xIfVrcx3R2

— Paramendra Kumar Bhagat (@paramendra) July 16, 2026

Hydration breaks are a soccer thing. They come with the game. Coaches use it strategically. Are you an American who does not know a whole lot about soccer, maybe? ;) https://t.co/sNqhG851QN

— Paramendra Kumar Bhagat (@paramendra) July 16, 2026

That’s how making a startup succeed goes pic.twitter.com/CWSuWNFwQr

— Elon Musk (@elonmusk) July 16, 2026

The older I get, the more I realize how many people waste their entire lives waiting to feel ready. Gathering more information. Creating the perfect plan. Simulating progress. Convincing themselves they’ll start tomorrow. Readiness is a myth. Action creates clarity. Do the thing.

— Sahil Bloom (@SahilBloom) July 16, 2026

It is a thoughtful framework overall and certainly a good starting point for discussions https://t.co/BfWxF4jnS5

— Elon Musk (@elonmusk) July 15, 2026

https://t.co/PTeDiv1b6L

— Demis Hassabis (@demishassabis) July 14, 2026

‘Athlete aren’t good investors’ 🙄

Cope from the people who claim to have ‘value-add’.

Pairing the right talent with the right operators to launch a business authentic to their brand is a better asset class than traditional venture.

You need to raise far less (or in this… https://t.co/jvqmM3NqAe

— Alex Macdonald (@alexfmac) July 16, 2026

My bet: @thinkymachines will soon make more money than @AnthropicAI. Not by winning the race to build one standardized frontier model. By becoming the Palantir FDE for enterprise custom models.

The playbook:
1. Release the best American open-weight model.
2. Drive widespread…

— Xiaoyin Qu (@quxiaoyin) July 16, 2026

I had dinner with @bercankilic a few months ago. He collects, and prepares, data to build world models for robotics.

And he is very good at it. Probably because he worked on Formula 1 cars collecting data (he told me the team he worked at had a couple of petabytes of data from… https://t.co/yNxqqf8FLL

— Robert Scoble (@Scobleizer) July 16, 2026

Getting dressed:

Me: How do I look?

Jessica: You look like PG.

— Paul Graham (@paulg) July 15, 2026

I’m done underselling our work.

It is generationally relevant for humanity.

Every founder, I presume, has this moment where they just are fed up.

I am fed up. I am fed up of people not taking this work as seriously as we have for years.

We will make it happen. And no one… https://t.co/gZwyeARXQ5

— Parmita Mishra (@parmita) July 16, 2026

A self regulatory organization (SRO) is a reasonable-ish idea - but it often comes with the government also regulating (classic dual system which is common in financial services). So it often doesn't prevent the government regulation.

So you end up needing approval from both the… https://t.co/LNLSEiJwpl

— Brian Armstrong (@brian_armstrong) July 16, 2026

My bet: @thinkymachines will soon make more money than @AnthropicAI. Not by winning the race to build one standardized frontier model. By becoming the Palantir FDE for enterprise custom models.

The playbook:
1. Release the best American open-weight model.
2. Drive widespread…

— Xiaoyin Qu (@quxiaoyin) July 16, 2026

Chinese models are 112x cheaper than Anthropic per million tokens.

Chamath laid it out on CNBC: a "barrel of intelligence" costs $56 from Anthropic, $26 from OpenAI, $1.50 from Meta, $1 from xAI and Google, and $0.50 from Chinese models.

That is not a pricing quirk. That is the… pic.twitter.com/g71WvrnsH3

— Shruti (@heyshrutimishra) July 16, 2026

robotic sports leagues are gonna be huge pretty soon https://t.co/xvwPML7qIC

— Rand (@rand_longevity) July 16, 2026

I deeply resent that AI has forced me to eliminate em dashes from my writing for fear of signaling slop

— Bret Taylor (@btaylor) July 15, 2026

every few days I stumble across a massive unlock using the @beehiiv MCP. here's the latest:

content tags are great for SEO/AEO, and they let visitors filter by topic and discover relevant content

I used to be diligent about tagging every new post. then somewhere along the way I… pic.twitter.com/QFszjnFKYM

— Tyler Denk 🐝 (@denk_tweets) July 16, 2026

Most startups don’t have a technology problem. They have a behavioral one.

Overwhelmed founders overbuild, overhire, chase too many priorities, and delay hard decisions.

the company stalls when anxiety directs attention toward what feels urgent instead of what ACTUALLY drives…

— Alesya (@AlesyaMacWaters) July 16, 2026

Chinese AI labs are doing serious numbers.

I looked up all the reported and leaked numbers of the 5 big independent labs: Moonshot (Kimi), DeepSeek, Zhipu (GLM), Kuaishou (Kling), Minimax and the 2 hyperscaler labs: Bytedance (Seedance) and Alibaba (Qwen).

The private labs are… pic.twitter.com/0ojD5Mzfy5

— Deedy (@deedydas) July 16, 2026