I could understand the portal before. Now it’s a gazillion things bolted on.

Enshittification is well in force.

I’d trust the huperscalers a lot more with their workers/lambda like infra to run routine jobs calling LLM APIs or deterministic code instead of Anthropic.

Anthropic is a phenomenal paid model but they have a poor reliability record.

I don’t care much if Claude code hiccups when generating code. But after the code is generated I want it to run with multiple 9s under certain latencies every single time.

Netflix raising prices so YouTube keeps on raising prices.

If they justified the price increase by giving an option to totally disable shorts and control over hiding some parts of UI, I’d understand.

However Google is turning on their rent extraction knob. If people keep paying, they keep on increasing prices. Economics doing its work.

Kids don’t watch TV at all. Our screentime is decreasing. I think time to drop YouTube is close. It’s becoming more of a digital drug.

They are caring less about their users.

As usual anything Israel gets flagged.

From tech perspective, technology builders now need to be extra aware how it’s used.

Is releasing the next autonomous algo/library worth it?, if the US department of war with their friends can use that tech to kill more people.

Especially when more and more weapons become autonomous. Targets are chosen based on LLM hallucinations.

In 2026, we are the baddies.

More than tech we need to vote to reign in the Department of War spending. It will be the death of American Empire.

The white cells meant to protect, have become cancerous.

Saudi funding Ukraine and making Russia collapse to gain their Nuclear Weapons would be quite the future timeline.

Trump family is in bed with a ton of shady firms. Soooo much crypto fraud.

Most openly corrupt president of our time.

The signature of claude is in the piece. “Same X, Same Y, Z later”.

Great piece but sometimes I wonder how much is human thought vs AI fluff.

Or more specifically “We used up all the ammunition”. Americans got nothing for it other than higher inflated costs.

For anyone who really cares about AI in depth, Dwarkesh and the sxyz podcasts are the OG.

have not heard of sxyz podcasts? searching yt did not gave anything meaningful

Likely meant 'swyx', https://www.latent.space/podcast

(came from f5 bot) haha yeah if any ai companies are looking i’ll talk for $200m!!!

I don’t know. I find him pretty hard to listen to. He has admitted that his show prep is AI produced, and I think the gaps in his understanding come across in conversation. I also find his child-like irreverence and familiar tone with his guests to be very distasteful. He also can’t drink a pint of Guinness and you just can’t trust people like that.

His elbows are too pointy?

Did you listen to his interview with Amodei? The guy goes on about, “well don’t you know about such and such” or “yeah but an AI can’t learn like an intern…” like he’s trying to argue with him. Look, this is coming from a guy who hasn’t really done anything with his life other than be friends with people in San Francisco, talking to the head of a company that’s changing the way an entire industry operates. I think Dwarkesh just needs to shut up and listen. The total lack of respect puts me off.

I’d suggest having an adblocker first.

Second not having a ton of extensions. Extensions can do fishy things.

This is Chrome’s broken model. Before installing an extension, one should be able to see all the domains an extension talks to.

The domains should be listed in manifest. But that’s not how it works.

In Android, every app you open needs a gazillion default permissions.