Reflective Altruism (@ReflectiveAlt) header
Reflective Altruism (@ReflectiveAlt) profile picture

Reflective Altruism

@ReflectiveAlt

Using academic research to drive positive change within and around the effective altruism movement. https://xitter.cf/t.co/1OdfLBv4W1

 Oxford, UK
Joined December 2022
  • Tweets2.1K
  • Following109
  • Followers1.2K
82Media
Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)Photo by Reflective Altruism (@ReflectiveAlt)
At the risk of repeating myself, please save for retirement folks. futurism.com/artificial-inte…
Top AI Safety Researcher Says He's Stopped Saving for Retirement Because He's So Sure We're All...

Top AI Safety Researcher Says He's Stopped Saving for Retirement Because He's So Sure We're All...

AI researcher Nate Soares claims that he no longer saves for retirement due to the certainty of an AI apocalypse.

futurism.com
5
2
35
17,262
Wouldn't a solved world be boring? Read Part 4 of Bostrom's Deep Utopia with me. reflectivealtruism.com/2026/…
Deep utopia (Part 4: Thursday) - Reflective altruism

Deep utopia (Part 4: Thursday) - Reflective altruism

A solved world threatens not only purpose but also interestingness. Wouldn't a solved world be boring?

reflectivealtruism.com
4
240
I spent some time digging through the METR time horizon paper. I thought I'd hate it. It's actually a lot more rigorous than many things in this space (e.g. AI 2027). Credit where credit is due. It's pretty good.
4
214
Reflective Altruism retweeted
David Thorstad (@DavidThorstad) profile picture
David Thorstad@DavidThorstadOct 9
There's been a lot of recent concern about existential risk from power-seeking AI. I push back against that concern in a new paper: philpapers.org/rec/THOICA-10
Image from David Thorstad (@DavidThorstad)
1
1
8
371
The third lecture of Nick Bostrom's deep utopia deals with two challenges to purpose in a solved world: reflectivealtruism.com/2026/…
Deep Utopia (Part 3: Wednesday) - Reflective altruism

Deep Utopia (Part 3: Wednesday) - Reflective altruism

Wednesday's lecture asks how humans could find meaning in the face of two redundancy threats.

reflectivealtruism.com
113
Worth a look forum.effectivealtruism.org/…
A reply to Austin Chen on his “keen debt” to FTX when hiring Caroline Ellison at Manifund

A reply to Austin Chen on his “keen debt” to FTX when hiring Caroline Ellison at Manifund

New account here, created by an irate outsider to EA. …

forum.effectivealtruism.org
2
204
The second lecture of Deep Utopia deals with our cosmic endowment and barriers to reaching it reflectivealtruism.com/2026/…
Deep Utopia (Part 2: Tuesday) - Reflective altruism

Deep Utopia (Part 2: Tuesday) - Reflective altruism

Tuesday's lecture considers humanity's cosmic endowment and barriers to reaching it.

reflectivealtruism.com
97
Manifund just hired Caroline Ellison. I think this was a poor choice. reflectivealtruism.com/2026/…
Community (Part 1: From FTX to Manifund) - Reflective altruism

Community (Part 1: From FTX to Manifund) - Reflective altruism

Manifund has hired Caroline Ellison, former CEO of Alameda Research. That may be unwise.

reflectivealtruism.com
1
8
376
I wish that important concerns about AI misbehavior were not so often mixed with speculative concerns about existential risk. These letters would be much easier to sign if they weren't mixed. openletter.earth/open-letter…
1
148
This is quite a development lesswrong.com/posts/W3zn5jQa…
Caroline Ellison has joined Manifund — LessWrong

Caroline Ellison has joined Manifund — LessWrong

Austin: I’m happy to announce that Caroline Ellison has accepted a role at Manifund, to develop our funding platform and research how to effectively…

lesswrong.com
7
287
It's about time I read Deep utopia. Read it with me. reflectivealtruism.com/2026/…
Deep Utopia (Part 1: Monday) - Reflective altruism

Deep Utopia (Part 1: Monday) - Reflective altruism

The first lecture of Deep utopia focuses on work, income, and population.

reflectivealtruism.com
129
Effective altruists have shown remarkably improved fiscal responsibility in the post-FTX era reflectivealtruism.com/2026/…
Getting it right (Part 5: Fiscal responsibility) - Reflective altruism

Getting it right (Part 5: Fiscal responsibility) - Reflective altruism

The past several years have seen a remarkable improvement in fiscal responsibility within the effective altruism movement.

reflectivealtruism.com
2
137
Not yet. link.springer.com/article/10…
Against the singularity hypothesis

Against the singularity hypothesis

Philosophical Studies - The singularity hypothesis is a hypothesis about the future of artificial intelligence on which self-improving artificial agents will quickly become orders of magnitude more...

link.springer.com
Ti Morse (@ti_morse) profile pictureTi Morse
@ti_morseJul 25
Sam Altman: We are in the singularity
2
338
Another leading shutdown theorem does not significantly motivate concerns about shutdown-resistance. reflectivealtruism.com/2026/…
Revisiting the shutdown problem (Part 4: Krakovna and Kramar) - Reflective altruism

Revisiting the shutdown problem (Part 4: Krakovna and Kramar) - Reflective altruism

A second shutdown theorem due to Krakovna and Kramar is also unconvincing.

reflectivealtruism.com
2
193
A leading shutdown theorem needs stronger assumptions than it may appear. reflectivealtruism.com/2026/…
Revisiting the shutdown problem (Part 3: Thornley) - Reflective altruism

Revisiting the shutdown problem (Part 3: Thornley) - Reflective altruism

A leading shutdown theorem due to Elliott Thornley needs strong unmentioned assumptions to ground shutdown-resistance.

reflectivealtruism.com
2
126
I'll read this, but I am not optimistic. The AI 2027 report had significant problems with model assumptions and forecasts. reflectivealtruism.com/2025/…
Exaggerating the risks (Part 22: AI 2027 recap) - Reflective altruism

Exaggerating the risks (Part 22: AI 2027 recap) - Reflective altruism

This post draws lessons from our discussion of the AI 2027 report.

reflectivealtruism.com
Daniel Kokotajlo (@DKokotajlo) profile pictureDaniel Kokotajlo
@DKokotajloJul 9
In AI 2027, we predicted that AI would take over the world or irreversibly concentrate power. In AI 2040: Plan A, we've laid out our positive vision for what should happen instead.
Image from Daniel Kokotajlo (@DKokotajlo)
5
697
This article only has 806 online accesses. It's also one of the best articles on instrumental convergence. Fix please? link.springer.com/article/10…
Instrumental divergence

Instrumental divergence

Philosophical Studies - The thesis of instrumental convergence holds that a wide range of ends have common means: for instance, self preservation, desire preservation, self improvement, and...

link.springer.com
1
1
164
Some people argue that it's hard to shut down AI systems by citing (a) instrumental convergence, and (b) empirically observed shutdown-resistance in systems today. reflectivealtruism.com/2026/…
Revisiting the shutdown problem (Part 2: Informal arguments) - Reflective altruism

Revisiting the shutdown problem (Part 2: Informal arguments) - Reflective altruism

Informal arguments for Catastrophic Shutdown Difficulty include the Argument from Instrumental Convergence and the Empirical Argument.

reflectivealtruism.com
1
1
3
234
I argue that (a)-(b) fail. Instrumental convergence doesn't imply sufficiently strong shutdown-resistance. And empirical evidence of shutdown-resistance is scant, mixed, and not from high-stakes catastrophe settings.
2
68
Load more