💯

It's so weird when people assume good deeds must be inspired by feelings of guilt. I don't feel guilty about the state of the world -- I was born into a world full of fires that weren't of my making. But I'm not like "well I'm not the arsonist so I have no obligation to help".

The moral status quo can make people do terrible things. Drifting too much from the moral status quo can also make people do terrible things. So just drift a wee bit from the moral status quo and try to drift in the direction of not doing terrible things.

Relatedly, I would like it if we could stop conflating: - Consciousness (whether AI has subjective experience) - Danger (whether AI has the capability or propensity to cause catastrophe) - Moral patienthood (whether we have an ethical obligation to consider AI welfare)

🫠

GPT-4 can encode secret messages (and hide them from a user) (!!)

Financial Times The above graph is a sketch of an argument. AI abilities (capabilities) are increasing at warp speed. Making them do exactly what we want is moving much more slowly. See the increase in capabilities below, again from the FT:

Croatia Log - Day 2: Despite extensive surveying, the team has managed no sightings of the elusive spotted dog. Research suggests native populations may be in decline due to captive breeding programs in the United Kingdom on "[a] plantation where [the] population can roam."

After a deep review of our potential paths to impact and $$$, we've decided to discontinue Alvea's core operations & return remaining funds (~1y of runway) to investors. It's been a wild ride these past 1.5 yrs—huge thank you to all who've supported us ♥️ Reflections coming soon!

Sometimes I think about this paper and I get a little sad knowing that it’s exceedingly unlikely that I will ever be able to produce satire that is as good or instructive as it is

Not sure how you can look at this data, ostensibly be interested in either meritocracy or equality, and want to move away from standardized tests. It's the subjective measures that are most slanted in favor of the rich kids. nytimes.com/interactive/20…

I think Scott Alexander did a brilliant job at putting a finger on what "irks" populasts: astralcodexten.substack.com/p/a-modest-pro…

10/ We are also delighted to welcome Jade Leung to our leadership team. Jade joins us from OpenAI where she led the firm’s work on AGI governance, with a particular focus on frontier AI regulation, international governance, and safety protocols.

Excited to say that I’ve launched a new charity: Spiro - focused on Tuberculosis screening and prevention for children Working with my wonderful co-founder Roxanne Heston

+++

One of the absolute best routes into AI governance. Strongly recommend applying!

Today I handed in my university laptop, losing a part of my extended mind (albeit replaced with a new one I bought myself). I decided to write down some personal notes elicited by the closure of the Future of Humanity Institute.

Magnify Mentoring is accepting mentee applications! Magnify is for individuals from underrepresented groups who are interested in working in the effective altruism space :) They pair you with a mentor who can help answer your questions and provide career support ✨