Sigal Samuel

How can you know if an AI is plotting against you? What chimp research can teach us about AI’s ability to scheme.

You may have heard examples of AI "scheming" against us — blackmailing, deceiving, etc. But do those represent the actual tendencies of the AI, or is the bad behavior showing up because researchers are strongly nudging that out of the AI?

We need to avoid groupthink here
www.vox.com/future-perfe...

I write a philosophical advice column for Vox and I want to hear what questions are plaguing you! Here's where you can submit your question anonymously:

I hope you're feeling better!! And that is an AMAZING mug, wow your colleagues really get you :)

You sound like ChatGPT AI isn’t just impacting how we write — it’s changing how we speak and interact with others. And there’s only more to come.

The Verge on how generative AI homogenizes language www.theverge.com/openai/68674...

This looks so good!!

Let us know what moral dilemmas you’re grappling with. You could be featured in the next Your Mileage May Vary column! I'm Sigal Samuel, a senior reporter for Future Perfect, Vox’s section about how to do good in today’s confusing world. My advice column — Your Mileage May Vary — offers people a unique framework for t...

What’s a tricky moral choice you’re struggling with? From how to use tech ethically to how to talk politics with family you disagree with, Sigal Samuel’s advice column Your Mileage May Vary can help.

Ask away here: docs.google.com/forms/d/e/1F...

Let us know what moral dilemmas you’re grappling with. You could be featured in the next Your Mileage May Vary column! I'm Sigal Samuel, a senior reporter for Future Perfect, Vox’s section about how to do good in today’s confusing world. My advice column — Your Mileage May Vary — offers people a unique framework for t...

Have a dilemma you're grappling with? Want me to tackle it in my next advice column? Just fill out this short (totally anonymous!) form:
docs.google.com/forms/d/e/1F...

My response draws on Buddhists & Aristotle & also as a bonus I learned that rich Europeans in the 18th century actually paid men to live in their gardens as "ornamental hermits"! Apparently it was trendy to have an isolated man in a goat’s hair robe wandering around. What a world

The spiritual life calls out to me. But is it self-indulgent? Meditation feels selfish when the world is on fire.

Is it better to spend thousands of hours on meditation and spiritual contemplation, or to pursue a career that concretely helps people?

Someone wrote in to Your Mileage May Vary (my philosophical advice column) to ask that question. Here's my answer!

www.vox.com/future-perfe...

"For living systems there is an end, hence there is an end, a finality, a goal"

Gotta check out this book!

A.I. Is Homogenizing Our Thoughts Recent studies suggest that tools such as ChatGPT make our brains less active and our writing less original.

my @newyorker.com column this week goes in depth on how using AI makes us less original, unique, and creative as writers, thinkers, and communicators. Various new studies are proving that AI is a rampant force of homogenization: www.newyorker.com/culture/infi...

ChatGPT and OCD are a dangerous combo For people with OCD, reassurance can be harmful. ChatGPT provides an infinite supply.

ChatGPT + OCD = dangerous combo

www.vox.com/future-perfe...

He’s the godfather of AI. Now, he has a bold new plan to keep us safe from it. “I should have thought of this 10 years ago,” Yoshua Bengio says.

Enjoyed speaking with @sigalsamuel.bsky.social of @vox.com to mark the launch of @law-zero.bsky.social. We discussed the motivation behind the project, its research direction, and the challenges and risks of increasingly capable and autonomous AI systems.

Full article: www.vox.com/future-perfe...

First comes marriage. Then comes a flirtatious colleague. What to do when your partner wants an open marriage and you don’t.

What to do when your partner wants an open marriage and you don’t

www.vox.com/future-perfe...

My students think it’s fine to cheat with AI. Maybe they’re onto something. To save the humanities, we need to rethink our assumptions about AI — and education.

Students think it’s fine to cheat with AI. What if they’re onto something?

I argue that to save the humanities, we need to rethink our assumptions about AI — and education

www.vox.com/advice/41318... Shoutout to the brilliant @shannonvallor.bsky.social and @freganmitts.bsky.social!

First comes marriage. Then comes a flirtatious colleague. What to do when your partner wants an open marriage and you don’t.

What to do when your partner wants an open marriage and you don’t

www.vox.com/future-perfe...

My students think it’s fine to cheat with AI. Maybe they’re onto something. To save the humanities, we need to rethink our assumptions about AI — and education.

Students think it’s fine to cheat with AI. What if they’re onto something?

I argue that to save the humanities, we need to rethink our assumptions about AI — and education

www.vox.com/advice/41318... Shoutout to the brilliant @shannonvallor.bsky.social and @freganmitts.bsky.social!

AI systems could become conscious. What if they hate their lives? How to not torture ChatGPT and Claude’s successors.

AI systems could become conscious one day. What if they hate their lives?

Is it our duty to make sure they're happy? To make sure Claude is always enjoying spiritual bliss? What could that even mean for an AI?

My new piece:
www.vox.com/future-perfe...

The new pope has strong opinions about AI. Good. Pope Leo’s unusual crusade should be a model for every religion.

The new pope is so focused on AI that he actually chose the name Leo XIV *because of AI*

Is it weird for the Catholic Church to be making this a priority? Not at all! I argue it makes perfect sense — other religions should do the same

www.vox.com/future-perfe...

What's the likeliest explanation for Claude's "spiritual bliss" attractor? Is the Occam's razor here that there's something built deep into the architecture that leads Claude to keep returning to spiritual themes? eg its makers trained it to be philosophical & self-reflective, so it gets spiritual?

Societal and technological progress as sewing an ever-growing, ever-changing, patchy, and polychrome quilt Artificial Intelligence (AI) systems are increasingly placed in positions where their decisions have real consequences, e.g., moderating online spaces, conducting research, and advising on policy. Ens...

My favorite AI paper of the year has arrived! The authors point out something I think should be obvious but seems not to be:

There is no One True Answer when it comes to morality!

It's naive to think the path to AI alignment is to make the AIs converge on a single ethics
arxiv.org/abs/2505.05197

Some truly phenomenal charities here — check out the ones between @pureearthnow.bsky.social and @give-directly.bsky.social. You can buy a year of happiness for someone for just $9!

Details here: www.vox.com/future-perfe...

Does your job feel meaningless? Try “moral ambition.” What if we measured success in terms of how much good we do?

What is "moral ambition" — and should you have more of it? Would it make your career feel a lot more meaningful?

@rutgerbregman.com thinks so. I talked to him about his new book — read an excerpt of our convo here and listen to the podcast version linked in the piece!

www.vox.com/future-perfe...

These Beautiful Birds Form Something Like Lasting Friendships

Bird friendships: "unrelated members of a superb starling flock often help each other raise chicks, trading assistance to one another over years in a behavior that was not previously known." Very cool long-term field study! #birds #nature 🧪🌿🪶🦉

Let us know what moral dilemmas you’re grappling with. You could be featured in the next Your Mileage May Vary column! I'm Sigal Samuel, a senior reporter for Future Perfect, Vox’s section about how to do good in today’s confusing world. My advice column — Your Mileage May Vary — offers people a unique framework for t...

Here's where you can submit a question anonymously to my philosophical advice column, based on value pluralism! This is not your typical advice column because I don't believe there's 1 objectively right answer to moral dilemmas — I'll help you find the answer for you docs.google.com/forms/d/e/1F...

You can’t optimize your way to being a good person I tried to make the perfect moral choice every time. It eroded my humanity.

ICYMI here's my @vox.com article on why you can’t optimize your way to being a good person — and probably shouldn't try! www.vox.com/the-highligh...

SIGAL SAMUEL: optimising morality Lives Well Lived · Episode

Peter Singer invited me to come on his "Lives Well Lived" podcast to talk about my article on the perils of trying to optimize your morality & my philosophical advice column at Vox. We had some productive disagreement — thanks @petersinger.info & Kasia!

Listen here:
open.spotify.com/episode/0vMH...

Combining my passion for Renaissance humanism with my interest in LLMs - thanks @aeon.co :)

Is AI really thinking and reasoning — or just pretending to? The best answer — AI has “jagged intelligence” — lies in between hype and skepticism.

The big AI story of the past 6 months is: Companies now claim that their AI models are capable of genuine reasoning.

And the big question is: Is that true?

I found that the best answer lies in between hype and skepticism.

www.vox.com/future-perfe...