Ritesh Nayak

A staggering 12.6 percent of Americans aged 18 to 25 had serious thoughts of suicide in 2024. Report from the National Alliance on Mental Illness.

Prophecy with no believers is just graffiti

I'm not sure what fascinates me more about this image: the QR code to pay for the weighing machine or the fact that you have to duck a bag of Funyuns to stand on the scale.

Digital Alms: Blind beggars with Qr codes on the streets of Bangalore.

Only SEO and word of mouth helps with organic growth. Paid creates the worst audience. Billions of dollars burnt on startups no one remembers are proof of this.

What I imagine when I hear AI researchers talking about Superalignment

Amazon is hiring a Go To Market lead for AGI 😂

Alaska Board of Game approves revived bear-culling program in caribou grounds The controversial program, aimed at boosting the population of a struggling caribou herd in Western Alaska, had been halted by court rulings because of legal flaws.

Alaska has greenlit a brown bear killing program, believe it or not, so caribou numbers grow so they can hunt them later. Nothing says "first world" like fetishizing the killing of defenseless animals.

alaskapublic.org/news/environ...

From the @newyorker.com

Engineering morale must be falling faster than Facebook's user numbers.

What is your best guess as to the ratio of red-vs-green flags on the platform?

Is anyone left to defend trust and safety? It’s under assault this year from users, politicians, and their own executives — but the industry has responded with silence

www.platformer.news/trustcon-tru... by @caseynewton.bsky.social on the state of Trust and Safety in tech.

T&S teams might be disappearing, but there's more work being automated through AI. Govts are getting in on the act and helping create a new narrative around tech & AI safety.

Chickens in a meeting. One points at a chart and explains, "This year, we doubled our revenue-per-employee numbers! And all we had to do was fire half of the employees!"

Revenue Per Employee. www.patreon.com/posts/revenu...

Many things wrong here:

1. Clearly promotional material ends up in Updates rather than Promotions

2. Breaks the unsubscribe pattern set by the same company.

After failing to convince me to install the app again, Instagram spammed my inbox constantly about what I was missing out on. Unsubscribed from emails only to realize they're spamming me on WhatsApp now.

There's no escape.

Chomsky reminds us, caring about other people is a dangerous idea in America which signals the ongoing drift of the United States from a struggling democracy to an increasingly consolidated authoritarian state.

Took me these many years to realize I have never liked a show with a background laugh track.

Who in Google Pay thought this was a good idea?

The corporation that achieves AGI should ideally have zero employees, or already tending aggressively towards it. Anything else is just manipulation.

It's a great BS test for companies claiming a WIN.

Simple, true AGI = zero employees

The best engineers I know have been vibe coding their entire adult lives

Every frontier AI shop will come up with a definition for AGI which they will self-satisfy and claim WINs, as early as next year.

Unfortunately for some, it's their own past statements which pushes them towards this mass hysteria. I heard the new goal post is narrow AGI

LinkedIn is the strangest network where the ultimate goal is never needing to use it.

Market Research is an interesting thing to try. Timed it late unfortunately.

Chrome's AI assistance feature is fantastic.

The magic bit is where the AI changes the CSS dynamically to test out a hypothesis. Surely one of the best examples of great AI-assisted IDE features yet.

The only people ending up in heaven are the charitable and compassionate

Hey @anthropic.com why do compute credits have a use-by date like produce?

I'm now seeing referrals from ChatGPT on my Substack, which would be cool, but I picked a setting that prevents AI training on my data.

Generations keep getting shorter each time